first

.gitattributes

@@ -0,0 +1,3 @@
+*.gif filter=lfs diff=lfs merge=lfs -text
+*.png filter=lfs diff=lfs merge=lfs -text
+*.jpg filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,156 @@
+# Created by .ignore support plugin (hsz.mobi)
+### Python template
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+env/
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+*.egg-info/
+.installed.cfg
+*.egg
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*,cover
+.hypothesis/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+target/
+
+# IPython Notebook
+.ipynb_checkpoints
+
+# pyenv
+.python-version
+
+# celery beat schedule file
+celerybeat-schedule
+
+# dotenv
+.env
+
+# virtualenv
+venv/
+ENV/
+
+# Spyder project settings
+.spyderproject
+
+# Rope project settings
+.ropeproject
+### VirtualEnv template
+# Virtualenv
+# http://iamzed.com/2009/05/07/a-primer-on-virtualenv/
+.Python
+[Bb]in
+[Ii]nclude
+[Ll]ib
+[Ll]ib64
+[Ll]ocal
+[Ss]cripts
+pyvenv.cfg
+.venv
+pip-selfcheck.json
+### JetBrains template
+# Covers JetBrains IDEs: IntelliJ, RubyMine, PhpStorm, AppCode, PyCharm, CLion, Android Studio and Webstorm
+# Reference: https://intellij-support.jetbrains.com/hc/en-us/articles/206544839
+
+# User-specific stuff:
+.idea/workspace.xml
+.idea/tasks.xml
+.idea/dictionaries
+.idea/vcs.xml
+.idea/jsLibraryMappings.xml
+
+# Sensitive or high-churn files:
+.idea/dataSources.ids
+.idea/dataSources.xml
+.idea/dataSources.local.xml
+.idea/sqlDataSources.xml
+.idea/dynamic.xml
+.idea/uiDesigner.xml
+
+# Gradle:
+.idea/gradle.xml
+.idea/libraries
+
+# Mongo Explorer plugin:
+.idea/mongoSettings.xml
+
+.idea/
+
+## File-based project format:
+*.iws
+
+## Plugin-specific files:
+
+# IntelliJ
+/out/
+
+# mpeltonen/sbt-idea plugin
+.idea_modules/
+
+# JIRA plugin
+atlassian-ide-plugin.xml
+
+# Crashlytics plugin (for Android Studio and IntelliJ)
+com_crashlytics_export_strings.xml
+crashlytics.properties
+crashlytics-build.properties
+fabric.properties
+
+# Mac related
+.DS_Store
+
+checkpoints

LICENSE.txt

@@ -0,0 +1,97 @@
+Copyright (c) 2021, NVIDIA Corporation & affiliates. All rights reserved.
+
+
+NVIDIA Source Code License for StyleGAN3
+
+
+=======================================================================
+
+1. Definitions
+
+"Licensor" means any person or entity that distributes its Work.
+
+"Software" means the original work of authorship made available under
+this License.
+
+"Work" means the Software and any additions to or derivative works of
+the Software that are made available under this License.
+
+The terms "reproduce," "reproduction," "derivative works," and
+"distribution" have the meaning as provided under U.S. copyright law;
+provided, however, that for the purposes of this License, derivative
+works shall not include works that remain separable from, or merely
+link (or bind by name) to the interfaces of, the Work.
+
+Works, including the Software, are "made available" under this License
+by including in or with the Work either (a) a copyright notice
+referencing the applicability of this License to the Work, or (b) a
+copy of this License.
+
+2. License Grants
+
+    2.1 Copyright Grant. Subject to the terms and conditions of this
+    License, each Licensor grants to you a perpetual, worldwide,
+    non-exclusive, royalty-free, copyright license to reproduce,
+    prepare derivative works of, publicly display, publicly perform,
+    sublicense and distribute its Work and any resulting derivative
+    works in any form.
+
+3. Limitations
+
+    3.1 Redistribution. You may reproduce or distribute the Work only
+    if (a) you do so under this License, (b) you include a complete
+    copy of this License with your distribution, and (c) you retain
+    without modification any copyright, patent, trademark, or
+    attribution notices that are present in the Work.
+
+    3.2 Derivative Works. You may specify that additional or different
+    terms apply to the use, reproduction, and distribution of your
+    derivative works of the Work ("Your Terms") only if (a) Your Terms
+    provide that the use limitation in Section 3.3 applies to your
+    derivative works, and (b) you identify the specific derivative
+    works that are subject to Your Terms. Notwithstanding Your Terms,
+    this License (including the redistribution requirements in Section
+    3.1) will continue to apply to the Work itself.
+
+    3.3 Use Limitation. The Work and any derivative works thereof only
+    may be used or intended for use non-commercially. Notwithstanding
+    the foregoing, NVIDIA and its affiliates may use the Work and any
+    derivative works commercially. As used herein, "non-commercially"
+    means for research or evaluation purposes only.
+
+    3.4 Patent Claims. If you bring or threaten to bring a patent claim
+    against any Licensor (including any claim, cross-claim or
+    counterclaim in a lawsuit) to enforce any patents that you allege
+    are infringed by any Work, then your rights under this License from
+    such Licensor (including the grant in Section 2.1) will terminate
+    immediately.
+
+    3.5 Trademarks. This License does not grant any rights to use any
+    Licensor’s or its affiliates’ names, logos, or trademarks, except
+    as necessary to reproduce the notices described in this License.
+
+    3.6 Termination. If you violate any term of this License, then your
+    rights under this License (including the grant in Section 2.1) will
+    terminate immediately.
+
+4. Disclaimer of Warranty.
+
+THE WORK IS PROVIDED "AS IS" WITHOUT WARRANTIES OR CONDITIONS OF ANY
+KIND, EITHER EXPRESS OR IMPLIED, INCLUDING WARRANTIES OR CONDITIONS OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, TITLE OR
+NON-INFRINGEMENT. YOU BEAR THE RISK OF UNDERTAKING ANY ACTIVITIES UNDER
+THIS LICENSE.
+
+5. Limitation of Liability.
+
+EXCEPT AS PROHIBITED BY APPLICABLE LAW, IN NO EVENT AND UNDER NO LEGAL
+THEORY, WHETHER IN TORT (INCLUDING NEGLIGENCE), CONTRACT, OR OTHERWISE
+SHALL ANY LICENSOR BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY DIRECT,
+INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES ARISING OUT OF
+OR RELATED TO THIS LICENSE, THE USE OR INABILITY TO USE THE WORK
+(INCLUDING BUT NOT LIMITED TO LOSS OF GOODWILL, BUSINESS INTERRUPTION,
+LOST PROFITS OR DATA, COMPUTER FAILURE OR MALFUNCTION, OR ANY OTHER
+COMMERCIAL DAMAGES OR LOSSES), EVEN IF THE LICENSOR HAS BEEN ADVISED OF
+THE POSSIBILITY OF SUCH DAMAGES.
+
+=======================================================================

README.md

@@ -0,0 +1,73 @@
+---
+title: DragGan
+emoji: 👀
+colorFrom: purple
+colorTo: pink
+sdk: gradio
+sdk_version: 3.35.2
+app_file: visualizer_drag_gradio.py
+pinned: false
+---
+
+
+# Drag Your GAN: Interactive Point-based Manipulation on the Generative Image Manifold
+
+<p align="center">
+    <img src="DragGAN.gif", width="700">
+</p>
+
+**Figure:** *Drag your GAN.*
+
+> **Drag Your GAN: Interactive Point-based Manipulation on the Generative Image Manifold** <br>
+> Xingang Pan, Ayush Tewari, Thomas Leimkühler, Lingjie Liu, Abhimitra Meka, Christian Theobalt<br>
+> *SIGGRAPH 2023 Conference Proceedings*
+
+## Requirements
+
+Please follow the requirements of [https://github.com/NVlabs/stylegan3](https://github.com/NVlabs/stylegan3).
+
+## Download pre-trained StyleGAN2 weights
+
+To download pre-trained weights, simply run:
+```sh
+sh scripts/download_model.sh
+```
+If you want to try StyleGAN-Human and the Landscapes HQ (LHQ) dataset, please download weights from these links: [StyleGAN-Human](https://drive.google.com/file/d/1dlFEHbu-WzQWJl7nBBZYcTyo000H9hVm/view?usp=sharing), [LHQ](https://drive.google.com/file/d/16twEf0T9QINAEoMsWefoWiyhcTd-aiWc/view?usp=sharing), and put them under `./checkpoints`.
+
+Feel free to try other pretrained StyleGAN.
+
+## Run DragGAN GUI
+
+To start the DragGAN GUI, simply run:
+```sh
+sh scripts/gui.sh
+```
+
+This GUI supports editing GAN-generated images. To edit a real image, you need to first perform GAN inversion using tools like [PTI](https://github.com/danielroich/PTI). Then load the new latent code and model weights to the GUI.
+
+You can run DragGAN Gradio demo as well:
+```sh
+python visualizer_drag_gradio.py
+```
+
+## Acknowledgement
+
+This code is developed based on [StyleGAN3](https://github.com/NVlabs/stylegan3). Part of the code is borrowed from [StyleGAN-Human](https://github.com/stylegan-human/StyleGAN-Human).
+
+## License
+
+The code related to the DragGAN algorithm is licensed under [CC-BY-NC](https://creativecommons.org/licenses/by-nc/4.0/).
+However, most of this project are available under a separate license terms: all codes used or modified from [StyleGAN3](https://github.com/NVlabs/stylegan3) is under the [Nvidia Source Code License](https://github.com/NVlabs/stylegan3/blob/main/LICENSE.txt).
+
+Any form of use and derivative of this code must preserve the watermarking functionality.
+
+## BibTeX
+
+```bibtex
+@inproceedings{pan2023draggan,
+    title={Drag Your GAN: Interactive Point-based Manipulation on the Generative Image Manifold},
+    author={Pan, Xingang and Tewari, Ayush, and Leimk{\"u}hler, Thomas and Liu, Lingjie and Meka, Abhimitra and Theobalt, Christian},
+    booktitle = {ACM SIGGRAPH 2023 Conference Proceedings},
+    year={2023}
+}
+```

dnnlib/__init__.py

@@ -0,0 +1,9 @@
+# Copyright (c) 2021, NVIDIA CORPORATION & AFFILIATES.  All rights reserved.
+#
+# NVIDIA CORPORATION and its licensors retain all intellectual property
+# and proprietary rights in and to this software, related documentation
+# and any modifications thereto.  Any use, reproduction, disclosure or
+# distribution of this software and related documentation without an express
+# license agreement from NVIDIA CORPORATION is strictly prohibited.
+
+from .util import EasyDict, make_cache_dir_path

dnnlib/util.py

@@ -0,0 +1,491 @@
+# Copyright (c) 2021, NVIDIA CORPORATION & AFFILIATES.  All rights reserved.
+#
+# NVIDIA CORPORATION and its licensors retain all intellectual property
+# and proprietary rights in and to this software, related documentation
+# and any modifications thereto.  Any use, reproduction, disclosure or
+# distribution of this software and related documentation without an express
+# license agreement from NVIDIA CORPORATION is strictly prohibited.
+
+"""Miscellaneous utility classes and functions."""
+
+import ctypes
+import fnmatch
+import importlib
+import inspect
+import numpy as np
+import os
+import shutil
+import sys
+import types
+import io
+import pickle
+import re
+import requests
+import html
+import hashlib
+import glob
+import tempfile
+import urllib
+import urllib.request
+import uuid
+
+from distutils.util import strtobool
+from typing import Any, List, Tuple, Union
+
+
+# Util classes
+# ------------------------------------------------------------------------------------------
+
+
+class EasyDict(dict):
+    """Convenience class that behaves like a dict but allows access with the attribute syntax."""
+
+    def __getattr__(self, name: str) -> Any:
+        try:
+            return self[name]
+        except KeyError:
+            raise AttributeError(name)
+
+    def __setattr__(self, name: str, value: Any) -> None:
+        self[name] = value
+
+    def __delattr__(self, name: str) -> None:
+        del self[name]
+
+
+class Logger(object):
+    """Redirect stderr to stdout, optionally print stdout to a file, and optionally force flushing on both stdout and the file."""
+
+    def __init__(self, file_name: str = None, file_mode: str = "w", should_flush: bool = True):
+        self.file = None
+
+        if file_name is not None:
+            self.file = open(file_name, file_mode)
+
+        self.should_flush = should_flush
+        self.stdout = sys.stdout
+        self.stderr = sys.stderr
+
+        sys.stdout = self
+        sys.stderr = self
+
+    def __enter__(self) -> "Logger":
+        return self
+
+    def __exit__(self, exc_type: Any, exc_value: Any, traceback: Any) -> None:
+        self.close()
+
+    def write(self, text: Union[str, bytes]) -> None:
+        """Write text to stdout (and a file) and optionally flush."""
+        if isinstance(text, bytes):
+            text = text.decode()
+        if len(text) == 0: # workaround for a bug in VSCode debugger: sys.stdout.write(''); sys.stdout.flush() => crash
+            return
+
+        if self.file is not None:
+            self.file.write(text)
+
+        self.stdout.write(text)
+
+        if self.should_flush:
+            self.flush()
+
+    def flush(self) -> None:
+        """Flush written text to both stdout and a file, if open."""
+        if self.file is not None:
+            self.file.flush()
+
+        self.stdout.flush()
+
+    def close(self) -> None:
+        """Flush, close possible files, and remove stdout/stderr mirroring."""
+        self.flush()
+
+        # if using multiple loggers, prevent closing in wrong order
+        if sys.stdout is self:
+            sys.stdout = self.stdout
+        if sys.stderr is self:
+            sys.stderr = self.stderr
+
+        if self.file is not None:
+            self.file.close()
+            self.file = None
+
+
+# Cache directories
+# ------------------------------------------------------------------------------------------
+
+_dnnlib_cache_dir = None
+
+def set_cache_dir(path: str) -> None:
+    global _dnnlib_cache_dir
+    _dnnlib_cache_dir = path
+
+def make_cache_dir_path(*paths: str) -> str:
+    if _dnnlib_cache_dir is not None:
+        return os.path.join(_dnnlib_cache_dir, *paths)
+    if 'DNNLIB_CACHE_DIR' in os.environ:
+        return os.path.join(os.environ['DNNLIB_CACHE_DIR'], *paths)
+    if 'HOME' in os.environ:
+        return os.path.join(os.environ['HOME'], '.cache', 'dnnlib', *paths)
+    if 'USERPROFILE' in os.environ:
+        return os.path.join(os.environ['USERPROFILE'], '.cache', 'dnnlib', *paths)
+    return os.path.join(tempfile.gettempdir(), '.cache', 'dnnlib', *paths)
+
+# Small util functions
+# ------------------------------------------------------------------------------------------
+
+
+def format_time(seconds: Union[int, float]) -> str:
+    """Convert the seconds to human readable string with days, hours, minutes and seconds."""
+    s = int(np.rint(seconds))
+
+    if s < 60:
+        return "{0}s".format(s)
+    elif s < 60 * 60:
+        return "{0}m {1:02}s".format(s // 60, s % 60)
+    elif s < 24 * 60 * 60:
+        return "{0}h {1:02}m {2:02}s".format(s // (60 * 60), (s // 60) % 60, s % 60)
+    else:
+        return "{0}d {1:02}h {2:02}m".format(s // (24 * 60 * 60), (s // (60 * 60)) % 24, (s // 60) % 60)
+
+
+def format_time_brief(seconds: Union[int, float]) -> str:
+    """Convert the seconds to human readable string with days, hours, minutes and seconds."""
+    s = int(np.rint(seconds))
+
+    if s < 60:
+        return "{0}s".format(s)
+    elif s < 60 * 60:
+        return "{0}m {1:02}s".format(s // 60, s % 60)
+    elif s < 24 * 60 * 60:
+        return "{0}h {1:02}m".format(s // (60 * 60), (s // 60) % 60)
+    else:
+        return "{0}d {1:02}h".format(s // (24 * 60 * 60), (s // (60 * 60)) % 24)
+
+
+def ask_yes_no(question: str) -> bool:
+    """Ask the user the question until the user inputs a valid answer."""
+    while True:
+        try:
+            print("{0} [y/n]".format(question))
+            return strtobool(input().lower())
+        except ValueError:
+            pass
+
+
+def tuple_product(t: Tuple) -> Any:
+    """Calculate the product of the tuple elements."""
+    result = 1
+
+    for v in t:
+        result *= v
+
+    return result
+
+
+_str_to_ctype = {
+    "uint8": ctypes.c_ubyte,
+    "uint16": ctypes.c_uint16,
+    "uint32": ctypes.c_uint32,
+    "uint64": ctypes.c_uint64,
+    "int8": ctypes.c_byte,
+    "int16": ctypes.c_int16,
+    "int32": ctypes.c_int32,
+    "int64": ctypes.c_int64,
+    "float32": ctypes.c_float,
+    "float64": ctypes.c_double
+}
+
+
+def get_dtype_and_ctype(type_obj: Any) -> Tuple[np.dtype, Any]:
+    """Given a type name string (or an object having a __name__ attribute), return matching Numpy and ctypes types that have the same size in bytes."""
+    type_str = None
+
+    if isinstance(type_obj, str):
+        type_str = type_obj
+    elif hasattr(type_obj, "__name__"):
+        type_str = type_obj.__name__
+    elif hasattr(type_obj, "name"):
+        type_str = type_obj.name
+    else:
+        raise RuntimeError("Cannot infer type name from input")
+
+    assert type_str in _str_to_ctype.keys()
+
+    my_dtype = np.dtype(type_str)
+    my_ctype = _str_to_ctype[type_str]
+
+    assert my_dtype.itemsize == ctypes.sizeof(my_ctype)
+
+    return my_dtype, my_ctype
+
+
+def is_pickleable(obj: Any) -> bool:
+    try:
+        with io.BytesIO() as stream:
+            pickle.dump(obj, stream)
+        return True
+    except:
+        return False
+
+
+# Functionality to import modules/objects by name, and call functions by name
+# ------------------------------------------------------------------------------------------
+
+def get_module_from_obj_name(obj_name: str) -> Tuple[types.ModuleType, str]:
+    """Searches for the underlying module behind the name to some python object.
+    Returns the module and the object name (original name with module part removed)."""
+
+    # allow convenience shorthands, substitute them by full names
+    obj_name = re.sub("^np.", "numpy.", obj_name)
+    obj_name = re.sub("^tf.", "tensorflow.", obj_name)
+
+    # list alternatives for (module_name, local_obj_name)
+    parts = obj_name.split(".")
+    name_pairs = [(".".join(parts[:i]), ".".join(parts[i:])) for i in range(len(parts), 0, -1)]
+
+    # try each alternative in turn
+    for module_name, local_obj_name in name_pairs:
+        try:
+            module = importlib.import_module(module_name) # may raise ImportError
+            get_obj_from_module(module, local_obj_name) # may raise AttributeError
+            return module, local_obj_name
+        except:
+            pass
+
+    # maybe some of the modules themselves contain errors?
+    for module_name, _local_obj_name in name_pairs:
+        try:
+            importlib.import_module(module_name) # may raise ImportError
+        except ImportError:
+            if not str(sys.exc_info()[1]).startswith("No module named '" + module_name + "'"):
+                raise
+
+    # maybe the requested attribute is missing?
+    for module_name, local_obj_name in name_pairs:
+        try:
+            module = importlib.import_module(module_name) # may raise ImportError
+            get_obj_from_module(module, local_obj_name) # may raise AttributeError
+        except ImportError:
+            pass
+
+    # we are out of luck, but we have no idea why
+    raise ImportError(obj_name)
+
+
+def get_obj_from_module(module: types.ModuleType, obj_name: str) -> Any:
+    """Traverses the object name and returns the last (rightmost) python object."""
+    if obj_name == '':
+        return module
+    obj = module
+    for part in obj_name.split("."):
+        obj = getattr(obj, part)
+    return obj
+
+
+def get_obj_by_name(name: str) -> Any:
+    """Finds the python object with the given name."""
+    module, obj_name = get_module_from_obj_name(name)
+    return get_obj_from_module(module, obj_name)
+
+
+def call_func_by_name(*args, func_name: str = None, **kwargs) -> Any:
+    """Finds the python object with the given name and calls it as a function."""
+    assert func_name is not None
+    func_obj = get_obj_by_name(func_name)
+    assert callable(func_obj)
+    return func_obj(*args, **kwargs)
+
+
+def construct_class_by_name(*args, class_name: str = None, **kwargs) -> Any:
+    """Finds the python class with the given name and constructs it with the given arguments."""
+    return call_func_by_name(*args, func_name=class_name, **kwargs)
+
+
+def get_module_dir_by_obj_name(obj_name: str) -> str:
+    """Get the directory path of the module containing the given object name."""
+    module, _ = get_module_from_obj_name(obj_name)
+    return os.path.dirname(inspect.getfile(module))
+
+
+def is_top_level_function(obj: Any) -> bool:
+    """Determine whether the given object is a top-level function, i.e., defined at module scope using 'def'."""
+    return callable(obj) and obj.__name__ in sys.modules[obj.__module__].__dict__
+
+
+def get_top_level_function_name(obj: Any) -> str:
+    """Return the fully-qualified name of a top-level function."""
+    assert is_top_level_function(obj)
+    module = obj.__module__
+    if module == '__main__':
+        module = os.path.splitext(os.path.basename(sys.modules[module].__file__))[0]
+    return module + "." + obj.__name__
+
+
+# File system helpers
+# ------------------------------------------------------------------------------------------
+
+def list_dir_recursively_with_ignore(dir_path: str, ignores: List[str] = None, add_base_to_relative: bool = False) -> List[Tuple[str, str]]:
+    """List all files recursively in a given directory while ignoring given file and directory names.
+    Returns list of tuples containing both absolute and relative paths."""
+    assert os.path.isdir(dir_path)
+    base_name = os.path.basename(os.path.normpath(dir_path))
+
+    if ignores is None:
+        ignores = []
+
+    result = []
+
+    for root, dirs, files in os.walk(dir_path, topdown=True):
+        for ignore_ in ignores:
+            dirs_to_remove = [d for d in dirs if fnmatch.fnmatch(d, ignore_)]
+
+            # dirs need to be edited in-place
+            for d in dirs_to_remove:
+                dirs.remove(d)
+
+            files = [f for f in files if not fnmatch.fnmatch(f, ignore_)]
+
+        absolute_paths = [os.path.join(root, f) for f in files]
+        relative_paths = [os.path.relpath(p, dir_path) for p in absolute_paths]
+
+        if add_base_to_relative:
+            relative_paths = [os.path.join(base_name, p) for p in relative_paths]
+
+        assert len(absolute_paths) == len(relative_paths)
+        result += zip(absolute_paths, relative_paths)
+
+    return result
+
+
+def copy_files_and_create_dirs(files: List[Tuple[str, str]]) -> None:
+    """Takes in a list of tuples of (src, dst) paths and copies files.
+    Will create all necessary directories."""
+    for file in files:
+        target_dir_name = os.path.dirname(file[1])
+
+        # will create all intermediate-level directories
+        if not os.path.exists(target_dir_name):
+            os.makedirs(target_dir_name)
+
+        shutil.copyfile(file[0], file[1])
+
+
+# URL helpers
+# ------------------------------------------------------------------------------------------
+
+def is_url(obj: Any, allow_file_urls: bool = False) -> bool:
+    """Determine whether the given object is a valid URL string."""
+    if not isinstance(obj, str) or not "://" in obj:
+        return False
+    if allow_file_urls and obj.startswith('file://'):
+        return True
+    try:
+        res = requests.compat.urlparse(obj)
+        if not res.scheme or not res.netloc or not "." in res.netloc:
+            return False
+        res = requests.compat.urlparse(requests.compat.urljoin(obj, "/"))
+        if not res.scheme or not res.netloc or not "." in res.netloc:
+            return False
+    except:
+        return False
+    return True
+
+
+def open_url(url: str, cache_dir: str = None, num_attempts: int = 10, verbose: bool = True, return_filename: bool = False, cache: bool = True) -> Any:
+    """Download the given URL and return a binary-mode file object to access the data."""
+    assert num_attempts >= 1
+    assert not (return_filename and (not cache))
+
+    # Doesn't look like an URL scheme so interpret it as a local filename.
+    if not re.match('^[a-z]+://', url):
+        return url if return_filename else open(url, "rb")
+
+    # Handle file URLs.  This code handles unusual file:// patterns that
+    # arise on Windows:
+    #
+    # file:///c:/foo.txt
+    #
+    # which would translate to a local '/c:/foo.txt' filename that's
+    # invalid.  Drop the forward slash for such pathnames.
+    #
+    # If you touch this code path, you should test it on both Linux and
+    # Windows.
+    #
+    # Some internet resources suggest using urllib.request.url2pathname() but
+    # but that converts forward slashes to backslashes and this causes
+    # its own set of problems.
+    if url.startswith('file://'):
+        filename = urllib.parse.urlparse(url).path
+        if re.match(r'^/[a-zA-Z]:', filename):
+            filename = filename[1:]
+        return filename if return_filename else open(filename, "rb")
+
+    assert is_url(url)
+
+    # Lookup from cache.
+    if cache_dir is None:
+        cache_dir = make_cache_dir_path('downloads')
+
+    url_md5 = hashlib.md5(url.encode("utf-8")).hexdigest()
+    if cache:
+        cache_files = glob.glob(os.path.join(cache_dir, url_md5 + "_*"))
+        if len(cache_files) == 1:
+            filename = cache_files[0]
+            return filename if return_filename else open(filename, "rb")
+
+    # Download.
+    url_name = None
+    url_data = None
+    with requests.Session() as session:
+        if verbose:
+            print("Downloading %s ..." % url, end="", flush=True)
+        for attempts_left in reversed(range(num_attempts)):
+            try:
+                with session.get(url) as res:
+                    res.raise_for_status()
+                    if len(res.content) == 0:
+                        raise IOError("No data received")
+
+                    if len(res.content) < 8192:
+                        content_str = res.content.decode("utf-8")
+                        if "download_warning" in res.headers.get("Set-Cookie", ""):
+                            links = [html.unescape(link) for link in content_str.split('"') if "export=download" in link]
+                            if len(links) == 1:
+                                url = requests.compat.urljoin(url, links[0])
+                                raise IOError("Google Drive virus checker nag")
+                        if "Google Drive - Quota exceeded" in content_str:
+                            raise IOError("Google Drive download quota exceeded -- please try again later")
+
+                    match = re.search(r'filename="([^"]*)"', res.headers.get("Content-Disposition", ""))
+                    url_name = match[1] if match else url
+                    url_data = res.content
+                    if verbose:
+                        print(" done")
+                    break
+            except KeyboardInterrupt:
+                raise
+            except:
+                if not attempts_left:
+                    if verbose:
+                        print(" failed")
+                    raise
+                if verbose:
+                    print(".", end="", flush=True)
+
+    # Save to cache.
+    if cache:
+        safe_name = re.sub(r"[^0-9a-zA-Z-._]", "_", url_name)
+        cache_file = os.path.join(cache_dir, url_md5 + "_" + safe_name)
+        temp_file = os.path.join(cache_dir, "tmp_" + uuid.uuid4().hex + "_" + url_md5 + "_" + safe_name)
+        os.makedirs(cache_dir, exist_ok=True)
+        with open(temp_file, "wb") as f:
+            f.write(url_data)
+        os.replace(temp_file, cache_file) # atomic
+        if return_filename:
+            return cache_file
+
+    # Return data as file object.
+    assert not return_filename
+    return io.BytesIO(url_data)

environment.yml

@@ -0,0 +1,24 @@
+name: stylegan3
+channels:
+  - pytorch
+  - nvidia
+dependencies:
+  - python >= 3.8
+  - pip
+  - numpy>=1.20
+  - click>=8.0
+  - pillow=8.3.1
+  - scipy=1.7.1
+  - pytorch=1.9.1
+  - cudatoolkit=11.1
+  - requests=2.26.0
+  - tqdm=4.62.2
+  - ninja=1.10.2
+  - matplotlib=3.4.2
+  - imageio=2.9.0
+  - pip:
+    - imgui==1.3.0
+    - glfw==2.2.0
+    - pyopengl==3.1.5
+    - imageio-ffmpeg==0.4.3
+    - pyspng

gen_images.py

@@ -0,0 +1,149 @@
+# Copyright (c) 2021, NVIDIA CORPORATION & AFFILIATES.  All rights reserved.
+#
+# NVIDIA CORPORATION and its licensors retain all intellectual property
+# and proprietary rights in and to this software, related documentation
+# and any modifications thereto.  Any use, reproduction, disclosure or
+# distribution of this software and related documentation without an express
+# license agreement from NVIDIA CORPORATION is strictly prohibited.
+
+"""Generate images using pretrained network pickle."""
+
+import os
+import re
+from typing import List, Optional, Tuple, Union
+
+import click
+import dnnlib
+import numpy as np
+import PIL.Image
+import torch
+
+import legacy
+
+#----------------------------------------------------------------------------
+
+def parse_range(s: Union[str, List]) -> List[int]:
+    '''Parse a comma separated list of numbers or ranges and return a list of ints.
+
+    Example: '1,2,5-10' returns [1, 2, 5, 6, 7]
+    '''
+    if isinstance(s, list): return s
+    ranges = []
+    range_re = re.compile(r'^(\d+)-(\d+)$')
+    for p in s.split(','):
+        m = range_re.match(p)
+        if m:
+            ranges.extend(range(int(m.group(1)), int(m.group(2))+1))
+        else:
+            ranges.append(int(p))
+    return ranges
+
+#----------------------------------------------------------------------------
+
+def parse_vec2(s: Union[str, Tuple[float, float]]) -> Tuple[float, float]:
+    '''Parse a floating point 2-vector of syntax 'a,b'.
+
+    Example:
+        '0,1' returns (0,1)
+    '''
+    if isinstance(s, tuple): return s
+    parts = s.split(',')
+    if len(parts) == 2:
+        return (float(parts[0]), float(parts[1]))
+    raise ValueError(f'cannot parse 2-vector {s}')
+
+#----------------------------------------------------------------------------
+
+def make_transform(translate: Tuple[float,float], angle: float):
+    m = np.eye(3)
+    s = np.sin(angle/360.0*np.pi*2)
+    c = np.cos(angle/360.0*np.pi*2)
+    m[0][0] = c
+    m[0][1] = s
+    m[0][2] = translate[0]
+    m[1][0] = -s
+    m[1][1] = c
+    m[1][2] = translate[1]
+    return m
+
+#----------------------------------------------------------------------------
+
+@click.command()
+@click.option('--network', 'network_pkl', help='Network pickle filename', required=True)
+@click.option('--seeds', type=parse_range, help='List of random seeds (e.g., \'0,1,4-6\')', required=True)
+@click.option('--trunc', 'truncation_psi', type=float, help='Truncation psi', default=1, show_default=True)
+@click.option('--class', 'class_idx', type=int, help='Class label (unconditional if not specified)')
+@click.option('--noise-mode', help='Noise mode', type=click.Choice(['const', 'random', 'none']), default='const', show_default=True)
+@click.option('--translate', help='Translate XY-coordinate (e.g. \'0.3,1\')', type=parse_vec2, default='0,0', show_default=True, metavar='VEC2')
+@click.option('--rotate', help='Rotation angle in degrees', type=float, default=0, show_default=True, metavar='ANGLE')
+@click.option('--outdir', help='Where to save the output images', type=str, required=True, metavar='DIR')
+def generate_images(
+    network_pkl: str,
+    seeds: List[int],
+    truncation_psi: float,
+    noise_mode: str,
+    outdir: str,
+    translate: Tuple[float,float],
+    rotate: float,
+    class_idx: Optional[int]
+):
+    """Generate images using pretrained network pickle.
+
+    Examples:
+
+    \b
+    # Generate an image using pre-trained AFHQv2 model ("Ours" in Figure 1, left).
+    python gen_images.py --outdir=out --trunc=1 --seeds=2 \\
+        --network=https://api.ngc.nvidia.com/v2/models/nvidia/research/stylegan3/versions/1/files/stylegan3-r-afhqv2-512x512.pkl
+
+    \b
+    # Generate uncurated images with truncation using the MetFaces-U dataset
+    python gen_images.py --outdir=out --trunc=0.7 --seeds=600-605 \\
+        --network=https://api.ngc.nvidia.com/v2/models/nvidia/research/stylegan3/versions/1/files/stylegan3-t-metfacesu-1024x1024.pkl
+    """
+
+    print('Loading networks from "%s"...' % network_pkl)
+    device = torch.device('cuda')
+    with dnnlib.util.open_url(network_pkl) as f:
+        G = legacy.load_network_pkl(f)['G_ema'].to(device) # type: ignore
+        # import pickle
+        # G = legacy.load_network_pkl(f)
+        # output = open('checkpoints/stylegan2-car-config-f-pt.pkl', 'wb')
+        # pickle.dump(G, output)
+
+    os.makedirs(outdir, exist_ok=True)
+
+    # Labels.
+    label = torch.zeros([1, G.c_dim], device=device)
+    if G.c_dim != 0:
+        if class_idx is None:
+            raise click.ClickException('Must specify class label with --class when using a conditional network')
+        label[:, class_idx] = 1
+    else:
+        if class_idx is not None:
+            print ('warn: --class=lbl ignored when running on an unconditional network')
+
+    # Generate images.
+    for seed_idx, seed in enumerate(seeds):
+        print('Generating image for seed %d (%d/%d) ...' % (seed, seed_idx, len(seeds)))
+        z = torch.from_numpy(np.random.RandomState(seed).randn(1, G.z_dim)).to(device)
+
+        # Construct an inverse rotation/translation matrix and pass to the generator.  The
+        # generator expects this matrix as an inverse to avoid potentially failing numerical
+        # operations in the network.
+        if hasattr(G.synthesis, 'input'):
+            m = make_transform(translate, rotate)
+            m = np.linalg.inv(m)
+            G.synthesis.input.transform.copy_(torch.from_numpy(m))
+
+        img = G(z, label, truncation_psi=truncation_psi, noise_mode=noise_mode)
+        img = (img.permute(0, 2, 3, 1) * 127.5 + 128).clamp(0, 255).to(torch.uint8)
+        PIL.Image.fromarray(img[0].cpu().numpy(), 'RGB').save(f'{outdir}/seed{seed:04d}.png')
+
+
+#----------------------------------------------------------------------------
+
+if __name__ == "__main__":
+    generate_images() # pylint: disable=no-value-for-parameter
+
+#----------------------------------------------------------------------------

gradio_utils/__init__.py

@@ -0,0 +1,9 @@
+from .utils import (ImageMask, draw_mask_on_image, draw_points_on_image,
+                    get_latest_points_pair, get_valid_mask,
+                    on_change_single_global_state)
+
+__all__ = [
+    'draw_mask_on_image', 'draw_points_on_image',
+    'on_change_single_global_state', 'get_latest_points_pair',
+    'get_valid_mask', 'ImageMask'
+]

gradio_utils/utils.py

@@ -0,0 +1,154 @@
+import gradio as gr
+import numpy as np
+from PIL import Image, ImageDraw
+
+
+class ImageMask(gr.components.Image):
+    """
+    Sets: source="canvas", tool="sketch"
+    """
+
+    is_template = True
+
+    def __init__(self, **kwargs):
+        super().__init__(source="upload",
+                         tool="sketch",
+                         interactive=False,
+                         **kwargs)
+
+    def preprocess(self, x):
+        if x is None:
+            return x
+        if self.tool == "sketch" and self.source in ["upload", "webcam"
+                                                     ] and type(x) != dict:
+            decode_image = gr.processing_utils.decode_base64_to_image(x)
+            width, height = decode_image.size
+            mask = np.ones((height, width, 4), dtype=np.uint8)
+            mask[..., -1] = 255
+            mask = self.postprocess(mask)
+            x = {'image': x, 'mask': mask}
+        return super().preprocess(x)
+
+
+def get_valid_mask(mask: np.ndarray):
+    """Convert mask from gr.Image(0 to 255, RGBA) to binary mask.
+    """
+    if mask.ndim == 3:
+        mask_pil = Image.fromarray(mask).convert('L')
+        mask = np.array(mask_pil)
+    if mask.max() == 255:
+        mask = mask / 255
+    return mask
+
+
+def draw_points_on_image(image,
+                         points,
+                         curr_point=None,
+                         highlight_all=True,
+                         radius_scale=0.01):
+    overlay_rgba = Image.new("RGBA", image.size, 0)
+    overlay_draw = ImageDraw.Draw(overlay_rgba)
+    for point_key, point in points.items():
+        if ((curr_point is not None and curr_point == point_key)
+                or highlight_all):
+            p_color = (255, 0, 0)
+            t_color = (0, 0, 255)
+
+        else:
+            p_color = (255, 0, 0, 35)
+            t_color = (0, 0, 255, 35)
+
+        rad_draw = int(image.size[0] * radius_scale)
+
+        p_start = point.get("start_temp", point["start"])
+        p_target = point["target"]
+
+        if p_start is not None and p_target is not None:
+            p_draw = int(p_start[0]), int(p_start[1])
+            t_draw = int(p_target[0]), int(p_target[1])
+
+            overlay_draw.line(
+                (p_draw[0], p_draw[1], t_draw[0], t_draw[1]),
+                fill=(255, 255, 0),
+                width=2,
+            )
+
+        if p_start is not None:
+            p_draw = int(p_start[0]), int(p_start[1])
+            overlay_draw.ellipse(
+                (
+                    p_draw[0] - rad_draw,
+                    p_draw[1] - rad_draw,
+                    p_draw[0] + rad_draw,
+                    p_draw[1] + rad_draw,
+                ),
+                fill=p_color,
+            )
+
+            if curr_point is not None and curr_point == point_key:
+                # overlay_draw.text(p_draw, "p", font=font, align="center", fill=(0, 0, 0))
+                overlay_draw.text(p_draw, "p", align="center", fill=(0, 0, 0))
+
+        if p_target is not None:
+            t_draw = int(p_target[0]), int(p_target[1])
+            overlay_draw.ellipse(
+                (
+                    t_draw[0] - rad_draw,
+                    t_draw[1] - rad_draw,
+                    t_draw[0] + rad_draw,
+                    t_draw[1] + rad_draw,
+                ),
+                fill=t_color,
+            )
+
+            if curr_point is not None and curr_point == point_key:
+                # overlay_draw.text(t_draw, "t", font=font, align="center", fill=(0, 0, 0))
+                overlay_draw.text(t_draw, "t", align="center", fill=(0, 0, 0))
+
+    return Image.alpha_composite(image.convert("RGBA"),
+                                 overlay_rgba).convert("RGB")
+
+
+def draw_mask_on_image(image, mask):
+    im_mask = np.uint8(mask * 255)
+    im_mask_rgba = np.concatenate(
+        (
+            np.tile(im_mask[..., None], [1, 1, 3]),
+            45 * np.ones(
+                (im_mask.shape[0], im_mask.shape[1], 1), dtype=np.uint8),
+        ),
+        axis=-1,
+    )
+    im_mask_rgba = Image.fromarray(im_mask_rgba).convert("RGBA")
+
+    return Image.alpha_composite(image.convert("RGBA"),
+                                 im_mask_rgba).convert("RGB")
+
+
+def on_change_single_global_state(keys,
+                                  value,
+                                  global_state,
+                                  map_transform=None):
+    if map_transform is not None:
+        value = map_transform(value)
+
+    curr_state = global_state
+    if isinstance(keys, str):
+        last_key = keys
+
+    else:
+        for k in keys[:-1]:
+            curr_state = curr_state[k]
+
+        last_key = keys[-1]
+
+    curr_state[last_key] = value
+    return global_state
+
+
+def get_latest_points_pair(points_dict):
+    if not points_dict:
+        return None
+    point_idx = list(points_dict.keys())
+    latest_point_idx = max(point_idx)
+    return latest_point_idx

gui_utils/__init__.py

@@ -0,0 +1,9 @@
+# Copyright (c) 2021, NVIDIA CORPORATION & AFFILIATES.  All rights reserved.
+#
+# NVIDIA CORPORATION and its licensors retain all intellectual property
+# and proprietary rights in and to this software, related documentation
+# and any modifications thereto.  Any use, reproduction, disclosure or
+# distribution of this software and related documentation without an express
+# license agreement from NVIDIA CORPORATION is strictly prohibited.
+
+# empty

gui_utils/gl_utils.py

@@ -0,0 +1,416 @@
+# Copyright (c) 2021, NVIDIA CORPORATION & AFFILIATES.  All rights reserved.
+#
+# NVIDIA CORPORATION and its licensors retain all intellectual property
+# and proprietary rights in and to this software, related documentation
+# and any modifications thereto.  Any use, reproduction, disclosure or
+# distribution of this software and related documentation without an express
+# license agreement from NVIDIA CORPORATION is strictly prohibited.
+
+import math
+import os
+import functools
+import contextlib
+import numpy as np
+import OpenGL.GL as gl
+import OpenGL.GL.ARB.texture_float
+import dnnlib
+
+#----------------------------------------------------------------------------
+
+def init_egl():
+    assert os.environ['PYOPENGL_PLATFORM'] == 'egl' # Must be set before importing OpenGL.
+    import OpenGL.EGL as egl
+    import ctypes
+
+    # Initialize EGL.
+    display = egl.eglGetDisplay(egl.EGL_DEFAULT_DISPLAY)
+    assert display != egl.EGL_NO_DISPLAY
+    major = ctypes.c_int32()
+    minor = ctypes.c_int32()
+    ok = egl.eglInitialize(display, major, minor)
+    assert ok
+    assert major.value * 10 + minor.value >= 14
+
+    # Choose config.
+    config_attribs = [
+        egl.EGL_RENDERABLE_TYPE,    egl.EGL_OPENGL_BIT,
+        egl.EGL_SURFACE_TYPE,       egl.EGL_PBUFFER_BIT,
+        egl.EGL_NONE
+    ]
+    configs = (ctypes.c_int32 * 1)()
+    num_configs = ctypes.c_int32()
+    ok = egl.eglChooseConfig(display, config_attribs, configs, 1, num_configs)
+    assert ok
+    assert num_configs.value == 1
+    config = configs[0]
+
+    # Create dummy pbuffer surface.
+    surface_attribs = [
+        egl.EGL_WIDTH,  1,
+        egl.EGL_HEIGHT, 1,
+        egl.EGL_NONE
+    ]
+    surface = egl.eglCreatePbufferSurface(display, config, surface_attribs)
+    assert surface != egl.EGL_NO_SURFACE
+
+    # Setup GL context.
+    ok = egl.eglBindAPI(egl.EGL_OPENGL_API)
+    assert ok
+    context = egl.eglCreateContext(display, config, egl.EGL_NO_CONTEXT, None)
+    assert context != egl.EGL_NO_CONTEXT
+    ok = egl.eglMakeCurrent(display, surface, surface, context)
+    assert ok
+
+#----------------------------------------------------------------------------
+
+_texture_formats = {
+    ('uint8',   1): dnnlib.EasyDict(type=gl.GL_UNSIGNED_BYTE, format=gl.GL_LUMINANCE,       internalformat=gl.GL_LUMINANCE8),
+    ('uint8',   2): dnnlib.EasyDict(type=gl.GL_UNSIGNED_BYTE, format=gl.GL_LUMINANCE_ALPHA, internalformat=gl.GL_LUMINANCE8_ALPHA8),
+    ('uint8',   3): dnnlib.EasyDict(type=gl.GL_UNSIGNED_BYTE, format=gl.GL_RGB,             internalformat=gl.GL_RGB8),
+    ('uint8',   4): dnnlib.EasyDict(type=gl.GL_UNSIGNED_BYTE, format=gl.GL_RGBA,            internalformat=gl.GL_RGBA8),
+    ('float32', 1): dnnlib.EasyDict(type=gl.GL_FLOAT,         format=gl.GL_LUMINANCE,       internalformat=OpenGL.GL.ARB.texture_float.GL_LUMINANCE32F_ARB),
+    ('float32', 2): dnnlib.EasyDict(type=gl.GL_FLOAT,         format=gl.GL_LUMINANCE_ALPHA, internalformat=OpenGL.GL.ARB.texture_float.GL_LUMINANCE_ALPHA32F_ARB),
+    ('float32', 3): dnnlib.EasyDict(type=gl.GL_FLOAT,         format=gl.GL_RGB,             internalformat=gl.GL_RGB32F),
+    ('float32', 4): dnnlib.EasyDict(type=gl.GL_FLOAT,         format=gl.GL_RGBA,            internalformat=gl.GL_RGBA32F),
+}
+
+def get_texture_format(dtype, channels):
+    return _texture_formats[(np.dtype(dtype).name, int(channels))]
+
+#----------------------------------------------------------------------------
+
+def prepare_texture_data(image):
+    image = np.asarray(image)
+    if image.ndim == 2:
+        image = image[:, :, np.newaxis]
+    if image.dtype.name == 'float64':
+        image = image.astype('float32')
+    return image
+
+#----------------------------------------------------------------------------
+
+def draw_pixels(image, *, pos=0, zoom=1, align=0, rint=True):
+    pos = np.broadcast_to(np.asarray(pos, dtype='float32'), [2])
+    zoom = np.broadcast_to(np.asarray(zoom, dtype='float32'), [2])
+    align = np.broadcast_to(np.asarray(align, dtype='float32'), [2])
+    image = prepare_texture_data(image)
+    height, width, channels = image.shape
+    size = zoom * [width, height]
+    pos = pos - size * align
+    if rint:
+        pos = np.rint(pos)
+    fmt = get_texture_format(image.dtype, channels)
+
+    gl.glPushAttrib(gl.GL_CURRENT_BIT | gl.GL_PIXEL_MODE_BIT)
+    gl.glPushClientAttrib(gl.GL_CLIENT_PIXEL_STORE_BIT)
+    gl.glRasterPos2f(pos[0], pos[1])
+    gl.glPixelZoom(zoom[0], -zoom[1])
+    gl.glPixelStorei(gl.GL_UNPACK_ALIGNMENT, 1)
+    gl.glDrawPixels(width, height, fmt.format, fmt.type, image)
+    gl.glPopClientAttrib()
+    gl.glPopAttrib()
+
+#----------------------------------------------------------------------------
+
+def read_pixels(width, height, *, pos=0, dtype='uint8', channels=3):
+    pos = np.broadcast_to(np.asarray(pos, dtype='float32'), [2])
+    dtype = np.dtype(dtype)
+    fmt = get_texture_format(dtype, channels)
+    image = np.empty([height, width, channels], dtype=dtype)
+
+    gl.glPushClientAttrib(gl.GL_CLIENT_PIXEL_STORE_BIT)
+    gl.glPixelStorei(gl.GL_PACK_ALIGNMENT, 1)
+    gl.glReadPixels(int(np.round(pos[0])), int(np.round(pos[1])), width, height, fmt.format, fmt.type, image)
+    gl.glPopClientAttrib()
+    return np.flipud(image)
+
+#----------------------------------------------------------------------------
+
+class Texture:
+    def __init__(self, *, image=None, width=None, height=None, channels=None, dtype=None, bilinear=True, mipmap=True):
+        self.gl_id = None
+        self.bilinear = bilinear
+        self.mipmap = mipmap
+
+        # Determine size and dtype.
+        if image is not None:
+            image = prepare_texture_data(image)
+            self.height, self.width, self.channels = image.shape
+            self.dtype = image.dtype
+        else:
+            assert width is not None and height is not None
+            self.width = width
+            self.height = height
+            self.channels = channels if channels is not None else 3
+            self.dtype = np.dtype(dtype) if dtype is not None else np.uint8
+
+        # Validate size and dtype.
+        assert isinstance(self.width, int) and self.width >= 0
+        assert isinstance(self.height, int) and self.height >= 0
+        assert isinstance(self.channels, int) and self.channels >= 1
+        assert self.is_compatible(width=width, height=height, channels=channels, dtype=dtype)
+
+        # Create texture object.
+        self.gl_id = gl.glGenTextures(1)
+        with self.bind():
+            gl.glTexParameterf(gl.GL_TEXTURE_2D, gl.GL_TEXTURE_WRAP_S, gl.GL_CLAMP_TO_EDGE)
+            gl.glTexParameterf(gl.GL_TEXTURE_2D, gl.GL_TEXTURE_WRAP_T, gl.GL_CLAMP_TO_EDGE)
+            gl.glTexParameterf(gl.GL_TEXTURE_2D, gl.GL_TEXTURE_MAG_FILTER, gl.GL_LINEAR if self.bilinear else gl.GL_NEAREST)
+            gl.glTexParameterf(gl.GL_TEXTURE_2D, gl.GL_TEXTURE_MIN_FILTER, gl.GL_LINEAR_MIPMAP_LINEAR if self.mipmap else gl.GL_NEAREST)
+        self.update(image)
+
+    def delete(self):
+        if self.gl_id is not None:
+            gl.glDeleteTextures([self.gl_id])
+            self.gl_id = None
+
+    def __del__(self):
+        try:
+            self.delete()
+        except:
+            pass
+
+    @contextlib.contextmanager
+    def bind(self):
+        prev_id = gl.glGetInteger(gl.GL_TEXTURE_BINDING_2D)
+        gl.glBindTexture(gl.GL_TEXTURE_2D, self.gl_id)
+        yield
+        gl.glBindTexture(gl.GL_TEXTURE_2D, prev_id)
+
+    def update(self, image):
+        if image is not None:
+            image = prepare_texture_data(image)
+            assert self.is_compatible(image=image)
+        with self.bind():
+            fmt = get_texture_format(self.dtype, self.channels)
+            gl.glPushClientAttrib(gl.GL_CLIENT_PIXEL_STORE_BIT)
+            gl.glPixelStorei(gl.GL_UNPACK_ALIGNMENT, 1)
+            gl.glTexImage2D(gl.GL_TEXTURE_2D, 0, fmt.internalformat, self.width, self.height, 0, fmt.format, fmt.type, image)
+            if self.mipmap:
+                gl.glGenerateMipmap(gl.GL_TEXTURE_2D)
+            gl.glPopClientAttrib()
+
+    def draw(self, *, pos=0, zoom=1, align=0, rint=False, color=1, alpha=1, rounding=0):
+        zoom = np.broadcast_to(np.asarray(zoom, dtype='float32'), [2])
+        size = zoom * [self.width, self.height]
+        with self.bind():
+            gl.glPushAttrib(gl.GL_ENABLE_BIT)
+            gl.glEnable(gl.GL_TEXTURE_2D)
+            draw_rect(pos=pos, size=size, align=align, rint=rint, color=color, alpha=alpha, rounding=rounding)
+            gl.glPopAttrib()
+
+    def is_compatible(self, *, image=None, width=None, height=None, channels=None, dtype=None): # pylint: disable=too-many-return-statements
+        if image is not None:
+            if image.ndim != 3:
+                return False
+            ih, iw, ic = image.shape
+            if not self.is_compatible(width=iw, height=ih, channels=ic, dtype=image.dtype):
+                return False
+        if width is not None and self.width != width:
+            return False
+        if height is not None and self.height != height:
+            return False
+        if channels is not None and self.channels != channels:
+            return False
+        if dtype is not None and self.dtype != dtype:
+            return False
+        return True
+
+#----------------------------------------------------------------------------
+
+class Framebuffer:
+    def __init__(self, *, texture=None, width=None, height=None, channels=None, dtype=None, msaa=0):
+        self.texture = texture
+        self.gl_id = None
+        self.gl_color = None
+        self.gl_depth_stencil = None
+        self.msaa = msaa
+
+        # Determine size and dtype.
+        if texture is not None:
+            assert isinstance(self.texture, Texture)
+            self.width = texture.width
+            self.height = texture.height
+            self.channels = texture.channels
+            self.dtype = texture.dtype
+        else:
+            assert width is not None and height is not None
+            self.width = width
+            self.height = height
+            self.channels = channels if channels is not None else 4
+            self.dtype = np.dtype(dtype) if dtype is not None else np.float32
+
+        # Validate size and dtype.
+        assert isinstance(self.width, int) and self.width >= 0
+        assert isinstance(self.height, int) and self.height >= 0
+        assert isinstance(self.channels, int) and self.channels >= 1
+        assert width is None or width == self.width
+        assert height is None or height == self.height
+        assert channels is None or channels == self.channels
+        assert dtype is None or dtype == self.dtype
+
+        # Create framebuffer object.
+        self.gl_id = gl.glGenFramebuffers(1)
+        with self.bind():
+
+            # Setup color buffer.
+            if self.texture is not None:
+                assert self.msaa == 0
+                gl.glFramebufferTexture2D(gl.GL_FRAMEBUFFER, gl.GL_COLOR_ATTACHMENT0, gl.GL_TEXTURE_2D, self.texture.gl_id, 0)
+            else:
+                fmt = get_texture_format(self.dtype, self.channels)
+                self.gl_color = gl.glGenRenderbuffers(1)
+                gl.glBindRenderbuffer(gl.GL_RENDERBUFFER, self.gl_color)
+                gl.glRenderbufferStorageMultisample(gl.GL_RENDERBUFFER, self.msaa, fmt.internalformat, self.width, self.height)
+                gl.glFramebufferRenderbuffer(gl.GL_FRAMEBUFFER, gl.GL_COLOR_ATTACHMENT0, gl.GL_RENDERBUFFER, self.gl_color)
+
+            # Setup depth/stencil buffer.
+            self.gl_depth_stencil = gl.glGenRenderbuffers(1)
+            gl.glBindRenderbuffer(gl.GL_RENDERBUFFER, self.gl_depth_stencil)
+            gl.glRenderbufferStorageMultisample(gl.GL_RENDERBUFFER, self.msaa, gl.GL_DEPTH24_STENCIL8, self.width, self.height)
+            gl.glFramebufferRenderbuffer(gl.GL_FRAMEBUFFER, gl.GL_DEPTH_STENCIL_ATTACHMENT, gl.GL_RENDERBUFFER, self.gl_depth_stencil)
+
+    def delete(self):
+        if self.gl_id is not None:
+            gl.glDeleteFramebuffers([self.gl_id])
+            self.gl_id = None
+        if self.gl_color is not None:
+            gl.glDeleteRenderbuffers(1, [self.gl_color])
+            self.gl_color = None
+        if self.gl_depth_stencil is not None:
+            gl.glDeleteRenderbuffers(1, [self.gl_depth_stencil])
+            self.gl_depth_stencil = None
+
+    def __del__(self):
+        try:
+            self.delete()
+        except:
+            pass
+
+    @contextlib.contextmanager
+    def bind(self):
+        prev_fbo = gl.glGetInteger(gl.GL_FRAMEBUFFER_BINDING)
+        prev_rbo = gl.glGetInteger(gl.GL_RENDERBUFFER_BINDING)
+        gl.glBindFramebuffer(gl.GL_FRAMEBUFFER, self.gl_id)
+        if self.width is not None and self.height is not None:
+            gl.glViewport(0, 0, self.width, self.height)
+        yield
+        gl.glBindFramebuffer(gl.GL_FRAMEBUFFER, prev_fbo)
+        gl.glBindRenderbuffer(gl.GL_RENDERBUFFER, prev_rbo)
+
+    def blit(self, dst=None):
+        assert dst is None or isinstance(dst, Framebuffer)
+        with self.bind():
+            gl.glBindFramebuffer(gl.GL_DRAW_FRAMEBUFFER, 0 if dst is None else dst.fbo)
+            gl.glBlitFramebuffer(0, 0, self.width, self.height, 0, 0, self.width, self.height, gl.GL_COLOR_BUFFER_BIT, gl.GL_NEAREST)
+
+#----------------------------------------------------------------------------
+
+def draw_shape(vertices, *, mode=gl.GL_TRIANGLE_FAN, pos=0, size=1, color=1, alpha=1):
+    assert vertices.ndim == 2 and vertices.shape[1] == 2
+    pos = np.broadcast_to(np.asarray(pos, dtype='float32'), [2])
+    size = np.broadcast_to(np.asarray(size, dtype='float32'), [2])
+    color = np.broadcast_to(np.asarray(color, dtype='float32'), [3])
+    alpha = np.clip(np.broadcast_to(np.asarray(alpha, dtype='float32'), []), 0, 1)
+
+    gl.glPushClientAttrib(gl.GL_CLIENT_VERTEX_ARRAY_BIT)
+    gl.glPushAttrib(gl.GL_CURRENT_BIT | gl.GL_TRANSFORM_BIT)
+    gl.glMatrixMode(gl.GL_MODELVIEW)
+    gl.glPushMatrix()
+
+    gl.glEnableClientState(gl.GL_VERTEX_ARRAY)
+    gl.glEnableClientState(gl.GL_TEXTURE_COORD_ARRAY)
+    gl.glVertexPointer(2, gl.GL_FLOAT, 0, vertices)
+    gl.glTexCoordPointer(2, gl.GL_FLOAT, 0, vertices)
+    gl.glTranslate(pos[0], pos[1], 0)
+    gl.glScale(size[0], size[1], 1)
+    gl.glColor4f(color[0] * alpha, color[1] * alpha, color[2] * alpha, alpha)
+    gl.glDrawArrays(mode, 0, vertices.shape[0])
+
+    gl.glPopMatrix()
+    gl.glPopAttrib()
+    gl.glPopClientAttrib()
+
+#----------------------------------------------------------------------------
+
+def draw_arrow(x1, y1, x2, y2, l=10, width=1.0):
+    # Compute the length and angle of the arrow
+    dx = x2 - x1
+    dy = y2 - y1
+    length = math.sqrt(dx**2 + dy**2)
+    if length < l:
+        return
+    angle = math.atan2(dy, dx)
+    
+    # Save the current modelview matrix
+    gl.glPushMatrix()
+    
+    # Translate and rotate the coordinate system
+    gl.glTranslatef(x1, y1, 0.0)
+    gl.glRotatef(angle * 180.0 / math.pi, 0.0, 0.0, 1.0)
+    
+    # Set the line width
+    gl.glLineWidth(width)
+    # gl.glColor3f(0.75, 0.75, 0.75)
+    
+    # Begin drawing lines
+    gl.glBegin(gl.GL_LINES)
+    
+    # Draw the shaft of the arrow
+    gl.glVertex2f(0.0, 0.0)
+    gl.glVertex2f(length, 0.0)
+    
+    # Draw the head of the arrow
+    gl.glVertex2f(length, 0.0)
+    gl.glVertex2f(length - 2 * l, l)
+    gl.glVertex2f(length, 0.0)
+    gl.glVertex2f(length - 2 * l, -l)
+    
+    # End drawing lines
+    gl.glEnd()
+    
+    # Restore the modelview matrix
+    gl.glPopMatrix()
+
+#----------------------------------------------------------------------------
+
+def draw_rect(*, pos=0, pos2=None, size=None, align=0, rint=False, color=1, alpha=1, rounding=0):
+    assert pos2 is None or size is None
+    pos = np.broadcast_to(np.asarray(pos, dtype='float32'), [2])
+    pos2 = np.broadcast_to(np.asarray(pos2, dtype='float32'), [2]) if pos2 is not None else None
+    size = np.broadcast_to(np.asarray(size, dtype='float32'), [2]) if size is not None else None
+    size = size if size is not None else pos2 - pos if pos2 is not None else np.array([1, 1], dtype='float32')
+    pos = pos - size * align
+    if rint:
+        pos = np.rint(pos)
+    rounding = np.broadcast_to(np.asarray(rounding, dtype='float32'), [2])
+    rounding = np.minimum(np.abs(rounding) / np.maximum(np.abs(size), 1e-8), 0.5)
+    if np.min(rounding) == 0:
+        rounding *= 0
+    vertices = _setup_rect(float(rounding[0]), float(rounding[1]))
+    draw_shape(vertices, mode=gl.GL_TRIANGLE_FAN, pos=pos, size=size, color=color, alpha=alpha)
+
+@functools.lru_cache(maxsize=10000)
+def _setup_rect(rx, ry):
+    t = np.linspace(0, np.pi / 2, 1 if max(rx, ry) == 0 else 64)
+    s = 1 - np.sin(t); c = 1 - np.cos(t)
+    x = [c * rx, 1 - s * rx, 1 - c * rx, s * rx]
+    y = [s * ry, c * ry, 1 - s * ry, 1 - c * ry]
+    v = np.stack([x, y], axis=-1).reshape(-1, 2)
+    return v.astype('float32')
+
+#----------------------------------------------------------------------------
+
+def draw_circle(*, center=0, radius=100, hole=0, color=1, alpha=1):
+    hole = np.broadcast_to(np.asarray(hole, dtype='float32'), [])
+    vertices = _setup_circle(float(hole))
+    draw_shape(vertices, mode=gl.GL_TRIANGLE_STRIP, pos=center, size=radius, color=color, alpha=alpha)
+
+@functools.lru_cache(maxsize=10000)
+def _setup_circle(hole):
+    t = np.linspace(0, np.pi * 2, 128)
+    s = np.sin(t); c = np.cos(t)
+    v = np.stack([c, s, c * hole, s * hole], axis=-1).reshape(-1, 2)
+    return v.astype('float32')
+
+#----------------------------------------------------------------------------

gui_utils/glfw_window.py

@@ -0,0 +1,229 @@
+# Copyright (c) 2021, NVIDIA CORPORATION & AFFILIATES.  All rights reserved.
+#
+# NVIDIA CORPORATION and its licensors retain all intellectual property
+# and proprietary rights in and to this software, related documentation
+# and any modifications thereto.  Any use, reproduction, disclosure or
+# distribution of this software and related documentation without an express
+# license agreement from NVIDIA CORPORATION is strictly prohibited.
+
+import time
+import glfw
+import OpenGL.GL as gl
+from . import gl_utils
+
+#----------------------------------------------------------------------------
+
+class GlfwWindow: # pylint: disable=too-many-public-methods
+    def __init__(self, *, title='GlfwWindow', window_width=1920, window_height=1080, deferred_show=True, close_on_esc=True):
+        self._glfw_window           = None
+        self._drawing_frame         = False
+        self._frame_start_time      = None
+        self._frame_delta           = 0
+        self._fps_limit             = None
+        self._vsync                 = None
+        self._skip_frames           = 0
+        self._deferred_show         = deferred_show
+        self._close_on_esc          = close_on_esc
+        self._esc_pressed           = False
+        self._drag_and_drop_paths   = None
+        self._capture_next_frame    = False
+        self._captured_frame        = None
+
+        # Create window.
+        glfw.init()
+        glfw.window_hint(glfw.VISIBLE, False)
+        self._glfw_window = glfw.create_window(width=window_width, height=window_height, title=title, monitor=None, share=None)
+        self._attach_glfw_callbacks()
+        self.make_context_current()
+
+        # Adjust window.
+        self.set_vsync(False)
+        self.set_window_size(window_width, window_height)
+        if not self._deferred_show:
+            glfw.show_window(self._glfw_window)
+
+    def close(self):
+        if self._drawing_frame:
+            self.end_frame()
+        if self._glfw_window is not None:
+            glfw.destroy_window(self._glfw_window)
+            self._glfw_window = None
+        #glfw.terminate() # Commented out to play it nice with other glfw clients.
+
+    def __del__(self):
+        try:
+            self.close()
+        except:
+            pass
+
+    @property
+    def window_width(self):
+        return self.content_width
+
+    @property
+    def window_height(self):
+        return self.content_height + self.title_bar_height
+
+    @property
+    def content_width(self):
+        width, _height = glfw.get_window_size(self._glfw_window)
+        return width
+
+    @property
+    def content_height(self):
+        _width, height = glfw.get_window_size(self._glfw_window)
+        return height
+
+    @property
+    def title_bar_height(self):
+        _left, top, _right, _bottom = glfw.get_window_frame_size(self._glfw_window)
+        return top
+
+    @property
+    def monitor_width(self):
+        _, _, width, _height = glfw.get_monitor_workarea(glfw.get_primary_monitor())
+        return width
+
+    @property
+    def monitor_height(self):
+        _, _, _width, height = glfw.get_monitor_workarea(glfw.get_primary_monitor())
+        return height
+
+    @property
+    def frame_delta(self):
+        return self._frame_delta
+
+    def set_title(self, title):
+        glfw.set_window_title(self._glfw_window, title)
+
+    def set_window_size(self, width, height):
+        width = min(width, self.monitor_width)
+        height = min(height, self.monitor_height)
+        glfw.set_window_size(self._glfw_window, width, max(height - self.title_bar_height, 0))
+        if width == self.monitor_width and height == self.monitor_height:
+            self.maximize()
+
+    def set_content_size(self, width, height):
+        self.set_window_size(width, height + self.title_bar_height)
+
+    def maximize(self):
+        glfw.maximize_window(self._glfw_window)
+
+    def set_position(self, x, y):
+        glfw.set_window_pos(self._glfw_window, x, y + self.title_bar_height)
+
+    def center(self):
+        self.set_position((self.monitor_width - self.window_width) // 2, (self.monitor_height - self.window_height) // 2)
+
+    def set_vsync(self, vsync):
+        vsync = bool(vsync)
+        if vsync != self._vsync:
+            glfw.swap_interval(1 if vsync else 0)
+            self._vsync = vsync
+
+    def set_fps_limit(self, fps_limit):
+        self._fps_limit = int(fps_limit)
+
+    def should_close(self):
+        return glfw.window_should_close(self._glfw_window) or (self._close_on_esc and self._esc_pressed)
+
+    def skip_frame(self):
+        self.skip_frames(1)
+
+    def skip_frames(self, num): # Do not update window for the next N frames.
+        self._skip_frames = max(self._skip_frames, int(num))
+
+    def is_skipping_frames(self):
+        return self._skip_frames > 0
+
+    def capture_next_frame(self):
+        self._capture_next_frame = True
+
+    def pop_captured_frame(self):
+        frame = self._captured_frame
+        self._captured_frame = None
+        return frame
+
+    def pop_drag_and_drop_paths(self):
+        paths = self._drag_and_drop_paths
+        self._drag_and_drop_paths = None
+        return paths
+
+    def draw_frame(self): # To be overridden by subclass.
+        self.begin_frame()
+        # Rendering code goes here.
+        self.end_frame()
+
+    def make_context_current(self):
+        if self._glfw_window is not None:
+            glfw.make_context_current(self._glfw_window)
+
+    def begin_frame(self):
+        # End previous frame.
+        if self._drawing_frame:
+            self.end_frame()
+
+        # Apply FPS limit.
+        if self._frame_start_time is not None and self._fps_limit is not None:
+            delay = self._frame_start_time - time.perf_counter() + 1 / self._fps_limit
+            if delay > 0:
+                time.sleep(delay)
+        cur_time = time.perf_counter()
+        if self._frame_start_time is not None:
+            self._frame_delta = cur_time - self._frame_start_time
+        self._frame_start_time = cur_time
+
+        # Process events.
+        glfw.poll_events()
+
+        # Begin frame.
+        self._drawing_frame = True
+        self.make_context_current()
+
+        # Initialize GL state.
+        gl.glViewport(0, 0, self.content_width, self.content_height)
+        gl.glMatrixMode(gl.GL_PROJECTION)
+        gl.glLoadIdentity()
+        gl.glTranslate(-1, 1, 0)
+        gl.glScale(2 / max(self.content_width, 1), -2 / max(self.content_height, 1), 1)
+        gl.glMatrixMode(gl.GL_MODELVIEW)
+        gl.glLoadIdentity()
+        gl.glEnable(gl.GL_BLEND)
+        gl.glBlendFunc(gl.GL_ONE, gl.GL_ONE_MINUS_SRC_ALPHA) # Pre-multiplied alpha.
+
+        # Clear.
+        gl.glClearColor(0, 0, 0, 1)
+        gl.glClear(gl.GL_COLOR_BUFFER_BIT | gl.GL_DEPTH_BUFFER_BIT)
+
+    def end_frame(self):
+        assert self._drawing_frame
+        self._drawing_frame = False
+
+        # Skip frames if requested.
+        if self._skip_frames > 0:
+            self._skip_frames -= 1
+            return
+
+        # Capture frame if requested.
+        if self._capture_next_frame:
+            self._captured_frame = gl_utils.read_pixels(self.content_width, self.content_height)
+            self._capture_next_frame = False
+
+        # Update window.
+        if self._deferred_show:
+            glfw.show_window(self._glfw_window)
+            self._deferred_show = False
+        glfw.swap_buffers(self._glfw_window)
+
+    def _attach_glfw_callbacks(self):
+        glfw.set_key_callback(self._glfw_window, self._glfw_key_callback)
+        glfw.set_drop_callback(self._glfw_window, self._glfw_drop_callback)
+
+    def _glfw_key_callback(self, _window, key, _scancode, action, _mods):
+        if action == glfw.PRESS and key == glfw.KEY_ESCAPE:
+            self._esc_pressed = True
+
+    def _glfw_drop_callback(self, _window, paths):
+        self._drag_and_drop_paths = paths
+
+#----------------------------------------------------------------------------

gui_utils/imgui_utils.py

@@ -0,0 +1,191 @@
+# Copyright (c) 2021, NVIDIA CORPORATION & AFFILIATES.  All rights reserved.
+#
+# NVIDIA CORPORATION and its licensors retain all intellectual property
+# and proprietary rights in and to this software, related documentation
+# and any modifications thereto.  Any use, reproduction, disclosure or
+# distribution of this software and related documentation without an express
+# license agreement from NVIDIA CORPORATION is strictly prohibited.
+
+import contextlib
+import imgui
+
+#----------------------------------------------------------------------------
+
+def set_default_style(color_scheme='dark', spacing=9, indent=23, scrollbar=27):
+    s = imgui.get_style()
+    s.window_padding        = [spacing, spacing]
+    s.item_spacing          = [spacing, spacing]
+    s.item_inner_spacing    = [spacing, spacing]
+    s.columns_min_spacing   = spacing
+    s.indent_spacing        = indent
+    s.scrollbar_size        = scrollbar
+    s.frame_padding         = [4, 3]
+    s.window_border_size    = 1
+    s.child_border_size     = 1
+    s.popup_border_size     = 1
+    s.frame_border_size     = 1
+    s.window_rounding       = 0
+    s.child_rounding        = 0
+    s.popup_rounding        = 3
+    s.frame_rounding        = 3
+    s.scrollbar_rounding    = 3
+    s.grab_rounding         = 3
+
+    getattr(imgui, f'style_colors_{color_scheme}')(s)
+    c0 = s.colors[imgui.COLOR_MENUBAR_BACKGROUND]
+    c1 = s.colors[imgui.COLOR_FRAME_BACKGROUND]
+    s.colors[imgui.COLOR_POPUP_BACKGROUND] = [x * 0.7 + y * 0.3 for x, y in zip(c0, c1)][:3] + [1]
+
+#----------------------------------------------------------------------------
+
+@contextlib.contextmanager
+def grayed_out(cond=True):
+    if cond:
+        s = imgui.get_style()
+        text = s.colors[imgui.COLOR_TEXT_DISABLED]
+        grab = s.colors[imgui.COLOR_SCROLLBAR_GRAB]
+        back = s.colors[imgui.COLOR_MENUBAR_BACKGROUND]
+        imgui.push_style_color(imgui.COLOR_TEXT, *text)
+        imgui.push_style_color(imgui.COLOR_CHECK_MARK, *grab)
+        imgui.push_style_color(imgui.COLOR_SLIDER_GRAB, *grab)
+        imgui.push_style_color(imgui.COLOR_SLIDER_GRAB_ACTIVE, *grab)
+        imgui.push_style_color(imgui.COLOR_FRAME_BACKGROUND, *back)
+        imgui.push_style_color(imgui.COLOR_FRAME_BACKGROUND_HOVERED, *back)
+        imgui.push_style_color(imgui.COLOR_FRAME_BACKGROUND_ACTIVE, *back)
+        imgui.push_style_color(imgui.COLOR_BUTTON, *back)
+        imgui.push_style_color(imgui.COLOR_BUTTON_HOVERED, *back)
+        imgui.push_style_color(imgui.COLOR_BUTTON_ACTIVE, *back)
+        imgui.push_style_color(imgui.COLOR_HEADER, *back)
+        imgui.push_style_color(imgui.COLOR_HEADER_HOVERED, *back)
+        imgui.push_style_color(imgui.COLOR_HEADER_ACTIVE, *back)
+        imgui.push_style_color(imgui.COLOR_POPUP_BACKGROUND, *back)
+        yield
+        imgui.pop_style_color(14)
+    else:
+        yield
+
+#----------------------------------------------------------------------------
+
+@contextlib.contextmanager
+def item_width(width=None):
+    if width is not None:
+        imgui.push_item_width(width)
+        yield
+        imgui.pop_item_width()
+    else:
+        yield
+
+#----------------------------------------------------------------------------
+
+def scoped_by_object_id(method):
+    def decorator(self, *args, **kwargs):
+        imgui.push_id(str(id(self)))
+        res = method(self, *args, **kwargs)
+        imgui.pop_id()
+        return res
+    return decorator
+
+#----------------------------------------------------------------------------
+
+def button(label, width=0, enabled=True):
+    with grayed_out(not enabled):
+        clicked = imgui.button(label, width=width)
+    clicked = clicked and enabled
+    return clicked
+
+#----------------------------------------------------------------------------
+
+def collapsing_header(text, visible=None, flags=0, default=False, enabled=True, show=True):
+    expanded = False
+    if show:
+        if default:
+            flags |= imgui.TREE_NODE_DEFAULT_OPEN
+        if not enabled:
+            flags |= imgui.TREE_NODE_LEAF
+        with grayed_out(not enabled):
+            expanded, visible = imgui.collapsing_header(text, visible=visible, flags=flags)
+        expanded = expanded and enabled
+    return expanded, visible
+
+#----------------------------------------------------------------------------
+
+def popup_button(label, width=0, enabled=True):
+    if button(label, width, enabled):
+        imgui.open_popup(label)
+    opened = imgui.begin_popup(label)
+    return opened
+
+#----------------------------------------------------------------------------
+
+def input_text(label, value, buffer_length, flags, width=None, help_text=''):
+    old_value = value
+    color = list(imgui.get_style().colors[imgui.COLOR_TEXT])
+    if value == '':
+        color[-1] *= 0.5
+    with item_width(width):
+        imgui.push_style_color(imgui.COLOR_TEXT, *color)
+        value = value if value != '' else help_text
+        changed, value = imgui.input_text(label, value, buffer_length, flags)
+        value = value if value != help_text else ''
+        imgui.pop_style_color(1)
+    if not flags & imgui.INPUT_TEXT_ENTER_RETURNS_TRUE:
+        changed = (value != old_value)
+    return changed, value
+
+#----------------------------------------------------------------------------
+
+def drag_previous_control(enabled=True):
+    dragging = False
+    dx = 0
+    dy = 0
+    if imgui.begin_drag_drop_source(imgui.DRAG_DROP_SOURCE_NO_PREVIEW_TOOLTIP):
+        if enabled:
+            dragging = True
+            dx, dy = imgui.get_mouse_drag_delta()
+            imgui.reset_mouse_drag_delta()
+        imgui.end_drag_drop_source()
+    return dragging, dx, dy
+
+#----------------------------------------------------------------------------
+
+def drag_button(label, width=0, enabled=True):
+    clicked = button(label, width=width, enabled=enabled)
+    dragging, dx, dy = drag_previous_control(enabled=enabled)
+    return clicked, dragging, dx, dy
+
+#----------------------------------------------------------------------------
+
+def drag_hidden_window(label, x, y, width, height, enabled=True):
+    imgui.push_style_color(imgui.COLOR_WINDOW_BACKGROUND, 0, 0, 0, 0)
+    imgui.push_style_color(imgui.COLOR_BORDER, 0, 0, 0, 0)
+    imgui.set_next_window_position(x, y)
+    imgui.set_next_window_size(width, height)
+    imgui.begin(label, closable=False, flags=(imgui.WINDOW_NO_TITLE_BAR | imgui.WINDOW_NO_RESIZE | imgui.WINDOW_NO_MOVE))
+    dragging, dx, dy = drag_previous_control(enabled=enabled)
+    imgui.end()
+    imgui.pop_style_color(2)
+    return dragging, dx, dy
+
+#----------------------------------------------------------------------------
+
+def click_hidden_window(label, x, y, width, height, img_w, img_h, enabled=True):
+    imgui.push_style_color(imgui.COLOR_WINDOW_BACKGROUND, 0, 0, 0, 0)
+    imgui.push_style_color(imgui.COLOR_BORDER, 0, 0, 0, 0)
+    imgui.set_next_window_position(x, y)
+    imgui.set_next_window_size(width, height)
+    imgui.begin(label, closable=False, flags=(imgui.WINDOW_NO_TITLE_BAR | imgui.WINDOW_NO_RESIZE | imgui.WINDOW_NO_MOVE))
+    clicked, down = False, False
+    img_x, img_y = 0, 0
+    if imgui.is_mouse_down():
+        posx, posy = imgui.get_mouse_pos()
+        if posx >= x and posx < x + width and posy >= y and posy < y + height:
+            if imgui.is_mouse_clicked():
+                clicked = True
+            down = True
+            img_x = round((posx - x) / (width - 1) * (img_w - 1))
+            img_y = round((posy - y) / (height - 1) * (img_h - 1))
+    imgui.end()
+    imgui.pop_style_color(2)
+    return clicked, down, img_x, img_y
+
+#----------------------------------------------------------------------------

gui_utils/imgui_window.py

@@ -0,0 +1,103 @@
+# Copyright (c) 2021, NVIDIA CORPORATION & AFFILIATES.  All rights reserved.
+#
+# NVIDIA CORPORATION and its licensors retain all intellectual property
+# and proprietary rights in and to this software, related documentation
+# and any modifications thereto.  Any use, reproduction, disclosure or
+# distribution of this software and related documentation without an express
+# license agreement from NVIDIA CORPORATION is strictly prohibited.
+
+import os
+import imgui
+import imgui.integrations.glfw
+
+from . import glfw_window
+from . import imgui_utils
+from . import text_utils
+
+#----------------------------------------------------------------------------
+
+class ImguiWindow(glfw_window.GlfwWindow):
+    def __init__(self, *, title='ImguiWindow', font=None, font_sizes=range(14,24), **glfw_kwargs):
+        if font is None:
+            font = text_utils.get_default_font()
+        font_sizes = {int(size) for size in font_sizes}
+        super().__init__(title=title, **glfw_kwargs)
+
+        # Init fields.
+        self._imgui_context  = None
+        self._imgui_renderer = None
+        self._imgui_fonts    = None
+        self._cur_font_size  = max(font_sizes)
+
+        # Delete leftover imgui.ini to avoid unexpected behavior.
+        if os.path.isfile('imgui.ini'):
+            os.remove('imgui.ini')
+
+        # Init ImGui.
+        self._imgui_context = imgui.create_context()
+        self._imgui_renderer = _GlfwRenderer(self._glfw_window)
+        self._attach_glfw_callbacks()
+        imgui.get_io().ini_saving_rate = 0 # Disable creating imgui.ini at runtime.
+        imgui.get_io().mouse_drag_threshold = 0 # Improve behavior with imgui_utils.drag_custom().
+        self._imgui_fonts = {size: imgui.get_io().fonts.add_font_from_file_ttf(font, size) for size in font_sizes}
+        self._imgui_renderer.refresh_font_texture()
+
+    def close(self):
+        self.make_context_current()
+        self._imgui_fonts = None
+        if self._imgui_renderer is not None:
+            self._imgui_renderer.shutdown()
+            self._imgui_renderer = None
+        if self._imgui_context is not None:
+            #imgui.destroy_context(self._imgui_context) # Commented out to avoid creating imgui.ini at the end.
+            self._imgui_context = None
+        super().close()
+
+    def _glfw_key_callback(self, *args):
+        super()._glfw_key_callback(*args)
+        self._imgui_renderer.keyboard_callback(*args)
+
+    @property
+    def font_size(self):
+        return self._cur_font_size
+
+    @property
+    def spacing(self):
+        return round(self._cur_font_size * 0.4)
+
+    def set_font_size(self, target): # Applied on next frame.
+        self._cur_font_size = min((abs(key - target), key) for key in self._imgui_fonts.keys())[1]
+
+    def begin_frame(self):
+        # Begin glfw frame.
+        super().begin_frame()
+
+        # Process imgui events.
+        self._imgui_renderer.mouse_wheel_multiplier = self._cur_font_size / 10
+        if self.content_width > 0 and self.content_height > 0:
+            self._imgui_renderer.process_inputs()
+
+        # Begin imgui frame.
+        imgui.new_frame()
+        imgui.push_font(self._imgui_fonts[self._cur_font_size])
+        imgui_utils.set_default_style(spacing=self.spacing, indent=self.font_size, scrollbar=self.font_size+4)
+
+    def end_frame(self):
+        imgui.pop_font()
+        imgui.render()
+        imgui.end_frame()
+        self._imgui_renderer.render(imgui.get_draw_data())
+        super().end_frame()
+
+#----------------------------------------------------------------------------
+# Wrapper class for GlfwRenderer to fix a mouse wheel bug on Linux.
+
+class _GlfwRenderer(imgui.integrations.glfw.GlfwRenderer):
+    def __init__(self, *args, **kwargs):
+        super().__init__(*args, **kwargs)
+        self.mouse_wheel_multiplier = 1
+
+    def scroll_callback(self, window, x_offset, y_offset):
+        self.io.mouse_wheel += y_offset * self.mouse_wheel_multiplier
+
+#----------------------------------------------------------------------------

gui_utils/text_utils.py

@@ -0,0 +1,123 @@
+# Copyright (c) 2021, NVIDIA CORPORATION & AFFILIATES.  All rights reserved.
+#
+# NVIDIA CORPORATION and its licensors retain all intellectual property
+# and proprietary rights in and to this software, related documentation
+# and any modifications thereto.  Any use, reproduction, disclosure or
+# distribution of this software and related documentation without an express
+# license agreement from NVIDIA CORPORATION is strictly prohibited.
+
+import functools
+from typing import Optional
+
+import dnnlib
+import numpy as np
+import PIL.Image
+import PIL.ImageFont
+import scipy.ndimage
+
+from . import gl_utils
+
+#----------------------------------------------------------------------------
+
+def get_default_font():
+    url = 'http://fonts.gstatic.com/s/opensans/v17/mem8YaGs126MiZpBA-U1UpcaXcl0Aw.ttf' # Open Sans regular
+    return dnnlib.util.open_url(url, return_filename=True)
+
+#----------------------------------------------------------------------------
+
+@functools.lru_cache(maxsize=None)
+def get_pil_font(font=None, size=32):
+    if font is None:
+        font = get_default_font()
+    return PIL.ImageFont.truetype(font=font, size=size)
+
+#----------------------------------------------------------------------------
+
+def get_array(string, *, dropshadow_radius: int=None, **kwargs):
+    if dropshadow_radius is not None:
+        offset_x = int(np.ceil(dropshadow_radius*2/3))
+        offset_y = int(np.ceil(dropshadow_radius*2/3))
+        return _get_array_priv(string, dropshadow_radius=dropshadow_radius, offset_x=offset_x, offset_y=offset_y, **kwargs)
+    else:
+        return _get_array_priv(string, **kwargs)
+
+@functools.lru_cache(maxsize=10000)
+def _get_array_priv(
+    string: str, *,
+    size: int = 32,
+    max_width: Optional[int]=None,
+    max_height: Optional[int]=None,
+    min_size=10,
+    shrink_coef=0.8,
+    dropshadow_radius: int=None,
+    offset_x: int=None,
+    offset_y: int=None,
+    **kwargs
+):
+    cur_size = size
+    array = None
+    while True:
+        if dropshadow_radius is not None:
+            # separate implementation for dropshadow text rendering
+            array = _get_array_impl_dropshadow(string, size=cur_size, radius=dropshadow_radius, offset_x=offset_x, offset_y=offset_y, **kwargs)
+        else:
+            array = _get_array_impl(string, size=cur_size, **kwargs)
+        height, width, _ = array.shape
+        if (max_width is None or width <= max_width) and (max_height is None or height <= max_height) or (cur_size <= min_size):
+            break
+        cur_size = max(int(cur_size * shrink_coef), min_size)
+    return array
+
+#----------------------------------------------------------------------------
+
+@functools.lru_cache(maxsize=10000)
+def _get_array_impl(string, *, font=None, size=32, outline=0, outline_pad=3, outline_coef=3, outline_exp=2, line_pad: int=None):
+    pil_font = get_pil_font(font=font, size=size)
+    lines = [pil_font.getmask(line, 'L') for line in string.split('\n')]
+    lines = [np.array(line, dtype=np.uint8).reshape([line.size[1], line.size[0]]) for line in lines]
+    width = max(line.shape[1] for line in lines)
+    lines = [np.pad(line, ((0, 0), (0, width - line.shape[1])), mode='constant') for line in lines]
+    line_spacing = line_pad if line_pad is not None else size // 2
+    lines = [np.pad(line, ((0, line_spacing), (0, 0)), mode='constant') for line in lines[:-1]] + lines[-1:]
+    mask = np.concatenate(lines, axis=0)
+    alpha = mask
+    if outline > 0:
+        mask = np.pad(mask, int(np.ceil(outline * outline_pad)), mode='constant', constant_values=0)
+        alpha = mask.astype(np.float32) / 255
+        alpha = scipy.ndimage.gaussian_filter(alpha, outline)
+        alpha = 1 - np.maximum(1 - alpha * outline_coef, 0) ** outline_exp
+        alpha = (alpha * 255 + 0.5).clip(0, 255).astype(np.uint8)
+        alpha = np.maximum(alpha, mask)
+    return np.stack([mask, alpha], axis=-1)
+
+#----------------------------------------------------------------------------
+
+@functools.lru_cache(maxsize=10000)
+def _get_array_impl_dropshadow(string, *, font=None, size=32, radius: int, offset_x: int, offset_y: int, line_pad: int=None, **kwargs):
+    assert (offset_x > 0) and (offset_y > 0)
+    pil_font = get_pil_font(font=font, size=size)
+    lines = [pil_font.getmask(line, 'L') for line in string.split('\n')]
+    lines = [np.array(line, dtype=np.uint8).reshape([line.size[1], line.size[0]]) for line in lines]
+    width = max(line.shape[1] for line in lines)
+    lines = [np.pad(line, ((0, 0), (0, width - line.shape[1])), mode='constant') for line in lines]
+    line_spacing = line_pad if line_pad is not None else size // 2
+    lines = [np.pad(line, ((0, line_spacing), (0, 0)), mode='constant') for line in lines[:-1]] + lines[-1:]
+    mask = np.concatenate(lines, axis=0)
+    alpha = mask
+
+    mask = np.pad(mask, 2*radius + max(abs(offset_x), abs(offset_y)), mode='constant', constant_values=0)
+    alpha = mask.astype(np.float32) / 255
+    alpha = scipy.ndimage.gaussian_filter(alpha, radius)
+    alpha = 1 - np.maximum(1 - alpha * 1.5, 0) ** 1.4
+    alpha = (alpha * 255 + 0.5).clip(0, 255).astype(np.uint8)
+    alpha = np.pad(alpha, [(offset_y, 0), (offset_x, 0)], mode='constant')[:-offset_y, :-offset_x]
+    alpha = np.maximum(alpha, mask)
+    return np.stack([mask, alpha], axis=-1)
+
+#----------------------------------------------------------------------------
+
+@functools.lru_cache(maxsize=10000)
+def get_texture(string, bilinear=True, mipmap=True, **kwargs):
+    return gl_utils.Texture(image=get_array(string, **kwargs), bilinear=bilinear, mipmap=mipmap)
+
+#----------------------------------------------------------------------------

legacy.py

@@ -0,0 +1,323 @@
+# Copyright (c) 2021, NVIDIA CORPORATION & AFFILIATES.  All rights reserved.
+#
+# NVIDIA CORPORATION and its licensors retain all intellectual property
+# and proprietary rights in and to this software, related documentation
+# and any modifications thereto.  Any use, reproduction, disclosure or
+# distribution of this software and related documentation without an express
+# license agreement from NVIDIA CORPORATION is strictly prohibited.
+
+"""Converting legacy network pickle into the new format."""
+
+import click
+import pickle
+import re
+import copy
+import numpy as np
+import torch
+import dnnlib
+from torch_utils import misc
+
+#----------------------------------------------------------------------------
+
+def load_network_pkl(f, force_fp16=False):
+    data = _LegacyUnpickler(f).load()
+
+    # Legacy TensorFlow pickle => convert.
+    if isinstance(data, tuple) and len(data) == 3 and all(isinstance(net, _TFNetworkStub) for net in data):
+        tf_G, tf_D, tf_Gs = data
+        G = convert_tf_generator(tf_G)
+        D = convert_tf_discriminator(tf_D)
+        G_ema = convert_tf_generator(tf_Gs)
+        data = dict(G=G, D=D, G_ema=G_ema)
+
+    # Add missing fields.
+    if 'training_set_kwargs' not in data:
+        data['training_set_kwargs'] = None
+    if 'augment_pipe' not in data:
+        data['augment_pipe'] = None
+
+    # Validate contents.
+    assert isinstance(data['G'], torch.nn.Module)
+    assert isinstance(data['D'], torch.nn.Module)
+    assert isinstance(data['G_ema'], torch.nn.Module)
+    assert isinstance(data['training_set_kwargs'], (dict, type(None)))
+    assert isinstance(data['augment_pipe'], (torch.nn.Module, type(None)))
+
+    # Force FP16.
+    if force_fp16:
+        for key in ['G', 'D', 'G_ema']:
+            old = data[key]
+            kwargs = copy.deepcopy(old.init_kwargs)
+            fp16_kwargs = kwargs.get('synthesis_kwargs', kwargs)
+            fp16_kwargs.num_fp16_res = 4
+            fp16_kwargs.conv_clamp = 256
+            if kwargs != old.init_kwargs:
+                new = type(old)(**kwargs).eval().requires_grad_(False)
+                misc.copy_params_and_buffers(old, new, require_all=True)
+                data[key] = new
+    return data
+
+#----------------------------------------------------------------------------
+
+class _TFNetworkStub(dnnlib.EasyDict):
+    pass
+
+class _LegacyUnpickler(pickle.Unpickler):
+    def find_class(self, module, name):
+        if module == 'dnnlib.tflib.network' and name == 'Network':
+            return _TFNetworkStub
+        return super().find_class(module, name)
+
+#----------------------------------------------------------------------------
+
+def _collect_tf_params(tf_net):
+    # pylint: disable=protected-access
+    tf_params = dict()
+    def recurse(prefix, tf_net):
+        for name, value in tf_net.variables:
+            tf_params[prefix + name] = value
+        for name, comp in tf_net.components.items():
+            recurse(prefix + name + '/', comp)
+    recurse('', tf_net)
+    return tf_params
+
+#----------------------------------------------------------------------------
+
+def _populate_module_params(module, *patterns):
+    for name, tensor in misc.named_params_and_buffers(module):
+        found = False
+        value = None
+        for pattern, value_fn in zip(patterns[0::2], patterns[1::2]):
+            match = re.fullmatch(pattern, name)
+            if match:
+                found = True
+                if value_fn is not None:
+                    value = value_fn(*match.groups())
+                break
+        try:
+            assert found
+            if value is not None:
+                tensor.copy_(torch.from_numpy(np.array(value)))
+        except:
+            print(name, list(tensor.shape))
+            raise
+
+#----------------------------------------------------------------------------
+
+def convert_tf_generator(tf_G):
+    if tf_G.version < 4:
+        raise ValueError('TensorFlow pickle version too low')
+
+    # Collect kwargs.
+    tf_kwargs = tf_G.static_kwargs
+    known_kwargs = set()
+    def kwarg(tf_name, default=None, none=None):
+        known_kwargs.add(tf_name)
+        val = tf_kwargs.get(tf_name, default)
+        return val if val is not None else none
+
+    # Convert kwargs.
+    from training import networks_stylegan2
+    network_class = networks_stylegan2.Generator
+    kwargs = dnnlib.EasyDict(
+        z_dim               = kwarg('latent_size',          512),
+        c_dim               = kwarg('label_size',           0),
+        w_dim               = kwarg('dlatent_size',         512),
+        img_resolution      = kwarg('resolution',           1024),
+        img_channels        = kwarg('num_channels',         3),
+        channel_base        = kwarg('fmap_base',            16384) * 2,
+        channel_max         = kwarg('fmap_max',             512),
+        num_fp16_res        = kwarg('num_fp16_res',         0),
+        conv_clamp          = kwarg('conv_clamp',           None),
+        architecture        = kwarg('architecture',         'skip'),
+        resample_filter     = kwarg('resample_kernel',      [1,3,3,1]),
+        use_noise           = kwarg('use_noise',            True),
+        activation          = kwarg('nonlinearity',         'lrelu'),
+        mapping_kwargs      = dnnlib.EasyDict(
+            num_layers      = kwarg('mapping_layers',       8),
+            embed_features  = kwarg('label_fmaps',          None),
+            layer_features  = kwarg('mapping_fmaps',        None),
+            activation      = kwarg('mapping_nonlinearity', 'lrelu'),
+            lr_multiplier   = kwarg('mapping_lrmul',        0.01),
+            w_avg_beta      = kwarg('w_avg_beta',           0.995,  none=1),
+        ),
+    )
+
+    # Check for unknown kwargs.
+    kwarg('truncation_psi')
+    kwarg('truncation_cutoff')
+    kwarg('style_mixing_prob')
+    kwarg('structure')
+    kwarg('conditioning')
+    kwarg('fused_modconv')
+    unknown_kwargs = list(set(tf_kwargs.keys()) - known_kwargs)
+    if len(unknown_kwargs) > 0:
+        raise ValueError('Unknown TensorFlow kwarg', unknown_kwargs[0])
+
+    # Collect params.
+    tf_params = _collect_tf_params(tf_G)
+    for name, value in list(tf_params.items()):
+        match = re.fullmatch(r'ToRGB_lod(\d+)/(.*)', name)
+        if match:
+            r = kwargs.img_resolution // (2 ** int(match.group(1)))
+            tf_params[f'{r}x{r}/ToRGB/{match.group(2)}'] = value
+            kwargs.synthesis.kwargs.architecture = 'orig'
+    #for name, value in tf_params.items(): print(f'{name:<50s}{list(value.shape)}')
+
+    # Convert params.
+    G = network_class(**kwargs).eval().requires_grad_(False)
+    # pylint: disable=unnecessary-lambda
+    # pylint: disable=f-string-without-interpolation
+    _populate_module_params(G,
+        r'mapping\.w_avg',                                  lambda:     tf_params[f'dlatent_avg'],
+        r'mapping\.embed\.weight',                          lambda:     tf_params[f'mapping/LabelEmbed/weight'].transpose(),
+        r'mapping\.embed\.bias',                            lambda:     tf_params[f'mapping/LabelEmbed/bias'],
+        r'mapping\.fc(\d+)\.weight',                        lambda i:   tf_params[f'mapping/Dense{i}/weight'].transpose(),
+        r'mapping\.fc(\d+)\.bias',                          lambda i:   tf_params[f'mapping/Dense{i}/bias'],
+        r'synthesis\.b4\.const',                            lambda:     tf_params[f'synthesis/4x4/Const/const'][0],
+        r'synthesis\.b4\.conv1\.weight',                    lambda:     tf_params[f'synthesis/4x4/Conv/weight'].transpose(3, 2, 0, 1),
+        r'synthesis\.b4\.conv1\.bias',                      lambda:     tf_params[f'synthesis/4x4/Conv/bias'],
+        r'synthesis\.b4\.conv1\.noise_const',               lambda:     tf_params[f'synthesis/noise0'][0, 0],
+        r'synthesis\.b4\.conv1\.noise_strength',            lambda:     tf_params[f'synthesis/4x4/Conv/noise_strength'],
+        r'synthesis\.b4\.conv1\.affine\.weight',            lambda:     tf_params[f'synthesis/4x4/Conv/mod_weight'].transpose(),
+        r'synthesis\.b4\.conv1\.affine\.bias',              lambda:     tf_params[f'synthesis/4x4/Conv/mod_bias'] + 1,
+        r'synthesis\.b(\d+)\.conv0\.weight',                lambda r:   tf_params[f'synthesis/{r}x{r}/Conv0_up/weight'][::-1, ::-1].transpose(3, 2, 0, 1),
+        r'synthesis\.b(\d+)\.conv0\.bias',                  lambda r:   tf_params[f'synthesis/{r}x{r}/Conv0_up/bias'],
+        r'synthesis\.b(\d+)\.conv0\.noise_const',           lambda r:   tf_params[f'synthesis/noise{int(np.log2(int(r)))*2-5}'][0, 0],
+        r'synthesis\.b(\d+)\.conv0\.noise_strength',        lambda r:   tf_params[f'synthesis/{r}x{r}/Conv0_up/noise_strength'],
+        r'synthesis\.b(\d+)\.conv0\.affine\.weight',        lambda r:   tf_params[f'synthesis/{r}x{r}/Conv0_up/mod_weight'].transpose(),
+        r'synthesis\.b(\d+)\.conv0\.affine\.bias',          lambda r:   tf_params[f'synthesis/{r}x{r}/Conv0_up/mod_bias'] + 1,
+        r'synthesis\.b(\d+)\.conv1\.weight',                lambda r:   tf_params[f'synthesis/{r}x{r}/Conv1/weight'].transpose(3, 2, 0, 1),
+        r'synthesis\.b(\d+)\.conv1\.bias',                  lambda r:   tf_params[f'synthesis/{r}x{r}/Conv1/bias'],
+        r'synthesis\.b(\d+)\.conv1\.noise_const',           lambda r:   tf_params[f'synthesis/noise{int(np.log2(int(r)))*2-4}'][0, 0],
+        r'synthesis\.b(\d+)\.conv1\.noise_strength',        lambda r:   tf_params[f'synthesis/{r}x{r}/Conv1/noise_strength'],
+        r'synthesis\.b(\d+)\.conv1\.affine\.weight',        lambda r:   tf_params[f'synthesis/{r}x{r}/Conv1/mod_weight'].transpose(),
+        r'synthesis\.b(\d+)\.conv1\.affine\.bias',          lambda r:   tf_params[f'synthesis/{r}x{r}/Conv1/mod_bias'] + 1,
+        r'synthesis\.b(\d+)\.torgb\.weight',                lambda r:   tf_params[f'synthesis/{r}x{r}/ToRGB/weight'].transpose(3, 2, 0, 1),
+        r'synthesis\.b(\d+)\.torgb\.bias',                  lambda r:   tf_params[f'synthesis/{r}x{r}/ToRGB/bias'],
+        r'synthesis\.b(\d+)\.torgb\.affine\.weight',        lambda r:   tf_params[f'synthesis/{r}x{r}/ToRGB/mod_weight'].transpose(),
+        r'synthesis\.b(\d+)\.torgb\.affine\.bias',          lambda r:   tf_params[f'synthesis/{r}x{r}/ToRGB/mod_bias'] + 1,
+        r'synthesis\.b(\d+)\.skip\.weight',                 lambda r:   tf_params[f'synthesis/{r}x{r}/Skip/weight'][::-1, ::-1].transpose(3, 2, 0, 1),
+        r'.*\.resample_filter',                             None,
+        r'.*\.act_filter',                                  None,
+    )
+    return G
+
+#----------------------------------------------------------------------------
+
+def convert_tf_discriminator(tf_D):
+    if tf_D.version < 4:
+        raise ValueError('TensorFlow pickle version too low')
+
+    # Collect kwargs.
+    tf_kwargs = tf_D.static_kwargs
+    known_kwargs = set()
+    def kwarg(tf_name, default=None):
+        known_kwargs.add(tf_name)
+        return tf_kwargs.get(tf_name, default)
+
+    # Convert kwargs.
+    kwargs = dnnlib.EasyDict(
+        c_dim                   = kwarg('label_size',           0),
+        img_resolution          = kwarg('resolution',           1024),
+        img_channels            = kwarg('num_channels',         3),
+        architecture            = kwarg('architecture',         'resnet'),
+        channel_base            = kwarg('fmap_base',            16384) * 2,
+        channel_max             = kwarg('fmap_max',             512),
+        num_fp16_res            = kwarg('num_fp16_res',         0),
+        conv_clamp              = kwarg('conv_clamp',           None),
+        cmap_dim                = kwarg('mapping_fmaps',        None),
+        block_kwargs = dnnlib.EasyDict(
+            activation          = kwarg('nonlinearity',         'lrelu'),
+            resample_filter     = kwarg('resample_kernel',      [1,3,3,1]),
+            freeze_layers       = kwarg('freeze_layers',        0),
+        ),
+        mapping_kwargs = dnnlib.EasyDict(
+            num_layers          = kwarg('mapping_layers',       0),
+            embed_features      = kwarg('mapping_fmaps',        None),
+            layer_features      = kwarg('mapping_fmaps',        None),
+            activation          = kwarg('nonlinearity',         'lrelu'),
+            lr_multiplier       = kwarg('mapping_lrmul',        0.1),
+        ),
+        epilogue_kwargs = dnnlib.EasyDict(
+            mbstd_group_size    = kwarg('mbstd_group_size',     None),
+            mbstd_num_channels  = kwarg('mbstd_num_features',   1),
+            activation          = kwarg('nonlinearity',         'lrelu'),
+        ),
+    )
+
+    # Check for unknown kwargs.
+    kwarg('structure')
+    kwarg('conditioning')
+    unknown_kwargs = list(set(tf_kwargs.keys()) - known_kwargs)
+    if len(unknown_kwargs) > 0:
+        raise ValueError('Unknown TensorFlow kwarg', unknown_kwargs[0])
+
+    # Collect params.
+    tf_params = _collect_tf_params(tf_D)
+    for name, value in list(tf_params.items()):
+        match = re.fullmatch(r'FromRGB_lod(\d+)/(.*)', name)
+        if match:
+            r = kwargs.img_resolution // (2 ** int(match.group(1)))
+            tf_params[f'{r}x{r}/FromRGB/{match.group(2)}'] = value
+            kwargs.architecture = 'orig'
+    #for name, value in tf_params.items(): print(f'{name:<50s}{list(value.shape)}')
+
+    # Convert params.
+    from training import networks_stylegan2
+    D = networks_stylegan2.Discriminator(**kwargs).eval().requires_grad_(False)
+    # pylint: disable=unnecessary-lambda
+    # pylint: disable=f-string-without-interpolation
+    _populate_module_params(D,
+        r'b(\d+)\.fromrgb\.weight',     lambda r:       tf_params[f'{r}x{r}/FromRGB/weight'].transpose(3, 2, 0, 1),
+        r'b(\d+)\.fromrgb\.bias',       lambda r:       tf_params[f'{r}x{r}/FromRGB/bias'],
+        r'b(\d+)\.conv(\d+)\.weight',   lambda r, i:    tf_params[f'{r}x{r}/Conv{i}{["","_down"][int(i)]}/weight'].transpose(3, 2, 0, 1),
+        r'b(\d+)\.conv(\d+)\.bias',     lambda r, i:    tf_params[f'{r}x{r}/Conv{i}{["","_down"][int(i)]}/bias'],
+        r'b(\d+)\.skip\.weight',        lambda r:       tf_params[f'{r}x{r}/Skip/weight'].transpose(3, 2, 0, 1),
+        r'mapping\.embed\.weight',      lambda:         tf_params[f'LabelEmbed/weight'].transpose(),
+        r'mapping\.embed\.bias',        lambda:         tf_params[f'LabelEmbed/bias'],
+        r'mapping\.fc(\d+)\.weight',    lambda i:       tf_params[f'Mapping{i}/weight'].transpose(),
+        r'mapping\.fc(\d+)\.bias',      lambda i:       tf_params[f'Mapping{i}/bias'],
+        r'b4\.conv\.weight',            lambda:         tf_params[f'4x4/Conv/weight'].transpose(3, 2, 0, 1),
+        r'b4\.conv\.bias',              lambda:         tf_params[f'4x4/Conv/bias'],
+        r'b4\.fc\.weight',              lambda:         tf_params[f'4x4/Dense0/weight'].transpose(),
+        r'b4\.fc\.bias',                lambda:         tf_params[f'4x4/Dense0/bias'],
+        r'b4\.out\.weight',             lambda:         tf_params[f'Output/weight'].transpose(),
+        r'b4\.out\.bias',               lambda:         tf_params[f'Output/bias'],
+        r'.*\.resample_filter',         None,
+    )
+    return D
+
+#----------------------------------------------------------------------------
+
+@click.command()
+@click.option('--source', help='Input pickle', required=True, metavar='PATH')
+@click.option('--dest', help='Output pickle', required=True, metavar='PATH')
+@click.option('--force-fp16', help='Force the networks to use FP16', type=bool, default=False, metavar='BOOL', show_default=True)
+def convert_network_pickle(source, dest, force_fp16):
+    """Convert legacy network pickle into the native PyTorch format.
+
+    The tool is able to load the main network configurations exported using the TensorFlow version of StyleGAN2 or StyleGAN2-ADA.
+    It does not support e.g. StyleGAN2-ADA comparison methods, StyleGAN2 configs A-D, or StyleGAN1 networks.
+
+    Example:
+
+    \b
+    python legacy.py \\
+        --source=https://nvlabs-fi-cdn.nvidia.com/stylegan2/networks/stylegan2-cat-config-f.pkl \\
+        --dest=stylegan2-cat-config-f.pkl
+    """
+    print(f'Loading "{source}"...')
+    with dnnlib.util.open_url(source) as f:
+        data = load_network_pkl(f, force_fp16=force_fp16)
+    print(f'Saving "{dest}"...')
+    with open(dest, 'wb') as f:
+        pickle.dump(data, f)
+    print('Done.')
+
+#----------------------------------------------------------------------------
+
+if __name__ == "__main__":
+    convert_network_pickle() # pylint: disable=no-value-for-parameter
+
+#----------------------------------------------------------------------------

requirements.txt

@@ -0,0 +1,20 @@
+torch
+pip
+numpy>=1.20
+click>=8.0
+pillow==8.3.1
+scipy==1.7.1
+pytorch==1.9.1
+cudatoolkit==11.1
+requests==2.26.0
+tqdm==4.62.2
+ninja==1.10.2
+matplotlib==3.4.2
+imageio==2.9.0
+imgui==1.3.0
+glfw==2.2.0
+pyopengl==3.1.5
+imageio-ffmpeg==0.4.3
+pyspng
+Ninja
+gradio

scripts/download_model.sh

@@ -0,0 +1,19 @@
+mkdir checkpoints
+cd checkpoints
+
+wget https://storage.googleapis.com/self-distilled-stylegan/lions_512_pytorch.pkl
+mv lions_512_pytorch.pkl stylegan2_lions_512_pytorch.pkl
+
+wget https://storage.googleapis.com/self-distilled-stylegan/dogs_1024_pytorch.pkl
+mv dogs_1024_pytorch.pkl stylegan2_dogs_1024_pytorch.pkl
+
+wget https://storage.googleapis.com/self-distilled-stylegan/horses_256_pytorch.pkl
+mv horses_256_pytorch.pkl stylegan2_horses_256_pytorch.pkl
+
+wget https://storage.googleapis.com/self-distilled-stylegan/elephants_512_pytorch.pkl
+mv elephants_512_pytorch.pkl stylegan2_elephants_512_pytorch.pkl
+
+wget https://api.ngc.nvidia.com/v2/models/nvidia/research/stylegan2/versions/1/files/stylegan2-ffhq-512x512.pkl
+wget https://api.ngc.nvidia.com/v2/models/nvidia/research/stylegan2/versions/1/files/stylegan2-afhqcat-512x512.pkl
+wget http://d36zk2xti64re0.cloudfront.net/stylegan2/networks/stylegan2-car-config-f.pkl
+wget http://d36zk2xti64re0.cloudfront.net/stylegan2/networks/stylegan2-cat-config-f.pkl

scripts/gui.sh

@@ -0,0 +1,11 @@
+python visualizer_drag.py \
+    checkpoints/stylegan2_lions_512_pytorch.pkl \
+    checkpoints/stylegan2-ffhq-512x512.pkl \
+    checkpoints/stylegan2-afhqcat-512x512.pkl \
+    checkpoints/stylegan2-car-config-f.pkl \
+    checkpoints/stylegan2_dogs_1024_pytorch.pkl \
+    checkpoints/stylegan2_horses_256_pytorch.pkl \
+    checkpoints/stylegan2-cat-config-f.pkl \
+    checkpoints/stylegan2_elephants_512_pytorch.pkl \
+    checkpoints/stylegan_human_v2_512.pkl \
+    checkpoints/stylegan2-lhq-256x256.pkl

stylegan_human/.gitignore

@@ -0,0 +1,10 @@
+.DS_Store
+__pycache__
+*.pt
+*.pth
+*.pdparams
+*.pdiparams
+*.pdmodel
+*.pkl
+*.info
+*.yaml

stylegan_human/PP_HumanSeg/deploy/infer.py

@@ -0,0 +1,180 @@
+# Copyright (c) SenseTime Research. All rights reserved.
+
+
+# Copyright (c) 2021 PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#    http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import codecs
+import os
+import time
+
+import yaml
+import numpy as np
+import cv2
+import paddle
+import paddleseg.transforms as T
+from paddle.inference import create_predictor, PrecisionType
+from paddle.inference import Config as PredictConfig
+from paddleseg.core.infer import reverse_transform
+from paddleseg.cvlibs import manager
+from paddleseg.utils import TimeAverager
+
+from ..scripts.optic_flow_process import optic_flow_process
+
+
+class DeployConfig:
+    def __init__(self, path):
+        with codecs.open(path, 'r', 'utf-8') as file:
+            self.dic = yaml.load(file, Loader=yaml.FullLoader)
+
+        self._transforms = self._load_transforms(self.dic['Deploy'][
+            'transforms'])
+        self._dir = os.path.dirname(path)
+
+    @property
+    def transforms(self):
+        return self._transforms
+
+    @property
+    def model(self):
+        return os.path.join(self._dir, self.dic['Deploy']['model'])
+
+    @property
+    def params(self):
+        return os.path.join(self._dir, self.dic['Deploy']['params'])
+
+    def _load_transforms(self, t_list):
+        com = manager.TRANSFORMS
+        transforms = []
+        for t in t_list:
+            ctype = t.pop('type')
+            transforms.append(com[ctype](**t))
+
+        return transforms
+
+
+class Predictor:
+    def __init__(self, args):
+        self.cfg = DeployConfig(args.cfg)
+        self.args = args
+        self.compose = T.Compose(self.cfg.transforms)
+        resize_h, resize_w = args.input_shape
+
+        self.disflow = cv2.DISOpticalFlow_create(
+            cv2.DISOPTICAL_FLOW_PRESET_ULTRAFAST)
+        self.prev_gray = np.zeros((resize_h, resize_w), np.uint8)
+        self.prev_cfd = np.zeros((resize_h, resize_w), np.float32)
+        self.is_init = True
+
+        pred_cfg = PredictConfig(self.cfg.model, self.cfg.params)
+        pred_cfg.disable_glog_info()
+        if self.args.use_gpu:
+            pred_cfg.enable_use_gpu(100, 0)
+
+        self.predictor = create_predictor(pred_cfg)
+        if self.args.test_speed:
+            self.cost_averager = TimeAverager()
+
+    def preprocess(self, img):
+        ori_shapes = []
+        processed_imgs = []
+        processed_img = self.compose(img)[0]
+        processed_imgs.append(processed_img)
+        ori_shapes.append(img.shape)
+        return processed_imgs, ori_shapes
+
+    def run(self, img, bg):
+        input_names = self.predictor.get_input_names()
+        input_handle = self.predictor.get_input_handle(input_names[0])
+        processed_imgs, ori_shapes = self.preprocess(img)
+        data = np.array(processed_imgs)
+        input_handle.reshape(data.shape)
+        input_handle.copy_from_cpu(data)
+        if self.args.test_speed:
+            start = time.time()
+
+        self.predictor.run()
+
+        if self.args.test_speed:
+            self.cost_averager.record(time.time() - start)
+        output_names = self.predictor.get_output_names()
+        output_handle = self.predictor.get_output_handle(output_names[0])
+        output = output_handle.copy_to_cpu()
+        return self.postprocess(output, img, ori_shapes[0], bg)
+
+
+    def postprocess(self, pred, img, ori_shape, bg):
+        if not os.path.exists(self.args.save_dir):
+            os.makedirs(self.args.save_dir)
+        resize_w = pred.shape[-1]
+        resize_h = pred.shape[-2]
+        if self.args.soft_predict:
+            if self.args.use_optic_flow:
+                score_map = pred[:, 1, :, :].squeeze(0)
+                score_map = 255 * score_map
+                cur_gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
+                cur_gray = cv2.resize(cur_gray, (resize_w, resize_h))
+                optflow_map = optic_flow_process(cur_gray, score_map, self.prev_gray, self.prev_cfd, \
+                        self.disflow, self.is_init)
+                self.prev_gray = cur_gray.copy()
+                self.prev_cfd = optflow_map.copy()
+                self.is_init = False
+
+                score_map = np.repeat(optflow_map[:, :, np.newaxis], 3, axis=2)
+                score_map = np.transpose(score_map, [2, 0, 1])[np.newaxis, ...]
+                score_map = reverse_transform(
+                    paddle.to_tensor(score_map),
+                    ori_shape,
+                    self.cfg.transforms,
+                    mode='bilinear')
+                alpha = np.transpose(score_map.numpy().squeeze(0),
+                                     [1, 2, 0]) / 255
+            else:
+                score_map = pred[:, 1, :, :]
+                score_map = score_map[np.newaxis, ...]
+                score_map = reverse_transform(
+                    paddle.to_tensor(score_map),
+                    ori_shape,
+                    self.cfg.transforms,
+                    mode='bilinear')
+                alpha = np.transpose(score_map.numpy().squeeze(0), [1, 2, 0])
+
+        else:
+            if pred.ndim == 3:
+                pred = pred[:, np.newaxis, ...]
+            result = reverse_transform(
+                paddle.to_tensor(
+                    pred, dtype='float32'),
+                ori_shape,
+                self.cfg.transforms,
+                mode='bilinear')
+
+            result = np.array(result)
+            if self.args.add_argmax:
+                result = np.argmax(result, axis=1)
+            else:
+                result = result.squeeze(1)
+            alpha = np.transpose(result, [1, 2, 0])
+
+        # background replace
+        h, w, _ = img.shape
+        if bg is None:
+            bg = np.ones_like(img)*255
+        else:
+            bg = cv2.resize(bg, (w, h))
+            if bg.ndim == 2:
+                bg = bg[..., np.newaxis]
+
+        comb = (alpha * img + (1 - alpha) * bg).astype(np.uint8)
+        return comb, alpha, bg, img

stylegan_human/PP_HumanSeg/export_model/download_export_model.py

@@ -0,0 +1,44 @@
+# coding: utf8
+# Copyright (c) 2021 PaddlePaddle Authors. All Rights Reserve.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#    http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import sys
+import os
+
+LOCAL_PATH = os.path.dirname(os.path.abspath(__file__))
+TEST_PATH = os.path.join(LOCAL_PATH, "../../../", "test")
+sys.path.append(TEST_PATH)
+
+from paddleseg.utils.download import download_file_and_uncompress
+
+model_urls = {
+    "pphumanseg_lite_portrait_398x224_with_softmax":
+    "https://paddleseg.bj.bcebos.com/dygraph/ppseg/ppseg_lite_portrait_398x224_with_softmax.tar.gz",
+    "deeplabv3p_resnet50_os8_humanseg_512x512_100k_with_softmax":
+    "https://paddleseg.bj.bcebos.com/dygraph/humanseg/export/deeplabv3p_resnet50_os8_humanseg_512x512_100k_with_softmax.zip",
+    "fcn_hrnetw18_small_v1_humanseg_192x192_with_softmax":
+    "https://paddleseg.bj.bcebos.com/dygraph/humanseg/export/fcn_hrnetw18_small_v1_humanseg_192x192_with_softmax.zip",
+    "pphumanseg_lite_generic_humanseg_192x192_with_softmax":
+    "https://paddleseg.bj.bcebos.com/dygraph/humanseg/export/pphumanseg_lite_generic_192x192_with_softmax.zip",
+}
+
+if __name__ == "__main__":
+    for model_name, url in model_urls.items():
+        download_file_and_uncompress(
+            url=url,
+            savepath=LOCAL_PATH,
+            extrapath=LOCAL_PATH,
+            extraname=model_name)
+
+    print("Export model download success!")

stylegan_human/PP_HumanSeg/pretrained_model/download_pretrained_model.py

@@ -0,0 +1,44 @@
+# coding: utf8
+# Copyright (c) 2021 PaddlePaddle Authors. All Rights Reserve.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#    http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import sys
+import os
+
+LOCAL_PATH = os.path.dirname(os.path.abspath(__file__))
+TEST_PATH = os.path.join(LOCAL_PATH, "../../../", "test")
+sys.path.append(TEST_PATH)
+
+from paddleseg.utils.download import download_file_and_uncompress
+
+model_urls = {
+    "pphumanseg_lite_portrait_398x224":
+    "https://paddleseg.bj.bcebos.com/dygraph/ppseg/ppseg_lite_portrait_398x224.tar.gz",
+    "deeplabv3p_resnet50_os8_humanseg_512x512_100k":
+    "https://paddleseg.bj.bcebos.com/dygraph/humanseg/train/deeplabv3p_resnet50_os8_humanseg_512x512_100k.zip",
+    "fcn_hrnetw18_small_v1_humanseg_192x192":
+    "https://paddleseg.bj.bcebos.com/dygraph/humanseg/train/fcn_hrnetw18_small_v1_humanseg_192x192.zip",
+    "pphumanseg_lite_generic_human_192x192":
+    "https://paddleseg.bj.bcebos.com/dygraph/humanseg/train/pphumanseg_lite_generic_192x192.zip",
+}
+
+if __name__ == "__main__":
+    for model_name, url in model_urls.items():
+        download_file_and_uncompress(
+            url=url,
+            savepath=LOCAL_PATH,
+            extrapath=LOCAL_PATH,
+            extraname=model_name)
+
+    print("Pretrained model download success!")

stylegan_human/README.md

@@ -0,0 +1,229 @@
+# StyleGAN-Human:  A Data-Centric Odyssey of Human Generation
+<img src="./img/demo_V5_thumbnails-min.png" width="96%" height="96%">
+
+<!--
+**stylegan-human/StyleGAN-Human** is a ✨ _special_ ✨ repository because its `README.md` (this file) appears on your GitHub profile.
+
+-->
+
+> 
+>
+> **Abstract:** *Unconditional human image generation is an important task in vision and graphics, which enables various applications in the creative industry. Existing studies in this field mainly focus on "network engineering" such as designing new components and objective functions. This work takes a data-centric perspective and investigates multiple critical aspects in "data engineering", which we believe would complement the current practice. To facilitate a comprehensive study, we collect and annotate a large-scale human image dataset with over 230K samples capturing diverse poses and textures. Equipped with this large dataset, we rigorously investigate three essential factors in data engineering for StyleGAN-based human generation, namely data size, data distribution, and data alignment. Extensive experiments reveal several valuable observations w.r.t. these aspects: 1) Large-scale data, more than 40K images, are needed to train a high-fidelity unconditional human generation model with vanilla StyleGAN. 2) A balanced training set helps improve the generation quality with rare face poses compared to the long-tailed counterpart, whereas simply balancing the clothing texture distribution does not effectively bring an improvement. 3) Human GAN models with body centers for alignment outperform models trained using face centers or pelvis points as alignment anchors. In addition, a model zoo and human editing applications are demonstrated to facilitate future research in the community.* <br>
+**Keyword:** Human Image Generation, Data-Centric, StyleGAN
+ 
+[Jianglin Fu](mailto:[email protected]), [Shikai Li](mailto:[email protected]), [Yuming Jiang](https://yumingj.github.io/), [Kwan-Yee Lin](https://kwanyeelin.github.io/), [Chen Qian](https://scholar.google.com/citations?user=AerkT0YAAAAJ&hl=zh-CN), [Chen Change Loy](https://www.mmlab-ntu.com/person/ccloy/), [Wayne Wu](https://wywu.github.io/), and [Ziwei Liu](https://liuziwei7.github.io/) <br>
+**[[Demo Video]](https://youtu.be/nIrb9hwsdcI)** | **[[Project Page]](https://stylegan-human.github.io/)** | **[[Paper]](https://arxiv.org/pdf/2204.11823.pdf)**
+
+## Updates
+- [20/07/2022] [SHHQ-1.0](./docs/Dataset.md) dataset with 40K images is released! :sparkles:
+- [15/06/2022] Data alignment and real-image inversion scripts are released. 
+- [26/04/2022] Technical report released!
+- [22/04/2022] Technical report will be released before May.
+- [21/04/2022] The codebase and project page are created.
+
+## Data Download
+The first version SHHQ-1.0, with 40K images is released. To download and use the dataset set, please read the instructions in [Dataset.md](./docs/Dataset.md)
+
+（We are currently facing large incoming applications, and we need to carefully verify all the applicants, please be patient, and we will reply to you as soon as possible.）
+
+## Model Zoo
+
+| Structure | 1024x512 | Metric | Scores |  512x256 | Metric | Scores | 
+| --------- |:----------:| :----------:| :----------:|  :-----: |  :-----: |  :-----: | 
+| StyleGAN1 |[stylegan_human_v1_1024.pkl](https://drive.google.com/file/d/1h-R-IV-INGdPEzj4P9ml6JTEvihuNgLX/view?usp=sharing)| fid50k | 3.79 | to be released | - | - |
+| StyleGAN2 |[stylegan_human_v2_1024.pkl](https://drive.google.com/file/d/1FlAb1rYa0r_--Zj_ML8e6shmaF28hQb5/view?usp=sharing)| fid50k_full | 1.57 |[stylegan_human_v2_512.pkl](https://drive.google.com/file/d/1dlFEHbu-WzQWJl7nBBZYcTyo000H9hVm/view?usp=sharing) | fid50k_full | 1.97 |
+| StyleGAN3 |to be released | - | - |  [stylegan_human_v3_512.pkl](https://drive.google.com/file/d/1_274jk_N6WSCkKWeu7hjHycqGvbuOFf5/view?usp=sharing) | fid50k_full | 2.54 |
+
+
+
+## Web Demo 
+
+Integrated into [Huggingface Spaces 🤗](https://huggingface.co/spaces) using [Gradio](https://github.com/gradio-app/gradio). Try out the Web Demo for generation: [![Hugging Face Spaces](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Spaces-blue)](https://huggingface.co/spaces/hysts/StyleGAN-Human) and interpolation [![Hugging Face Spaces](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Spaces-blue)](https://huggingface.co/spaces/hysts/StyleGAN-Human-Interpolation)
+
+
+
+<a href="https://colab.research.google.com/drive/1sgxoDM55iM07FS54vz9ALg1XckiYA2On"><img src="https://colab.research.google.com/assets/colab-badge.svg" height=22.5></a> 
+
+We prepare a Colab demo to allow you to synthesize images with the provided models, as well as visualize the performance of style-mixing, interpolation, and attributes editing.
+The notebook will guide you to install the necessary environment and download pretrained models. The output images can be found in `./StyleGAN-Human/outputs/`.
+Hope you enjoy!
+
+## Usage
+
+### System requirements
+* The original code bases are [stylegan (tensorflow)](https://github.com/NVlabs/stylegan), [stylegan2-ada (pytorch)](https://github.com/NVlabs/stylegan2-ada-pytorch), [stylegan3 (pytorch)](https://github.com/NVlabs/stylegan3), released by NVidia
+
+* We tested in Python 3.8.5 and PyTorch 1.9.1 with CUDA 11.1. (See https://pytorch.org for PyTorch install instructions.)
+
+### Installation
+To work with this project on your own machine, you need to install the environmnet as follows: 
+
+```
+conda env create -f environment.yml
+conda activate stylehuman
+# [Optional: tensorflow 1.x is required for StyleGAN1. ]
+pip install nvidia-pyindex
+pip install nvidia-tensorflow[horovod]
+pip install nvidia-tensorboard==1.15
+```
+Extra notes:
+1. In case having some conflicts when calling CUDA version, please try to empty the LD_LIBRARY_PATH. For example:
+```
+LD_LIBRARY_PATH=; python generate.py --outdir=out/stylegan_human_v2_1024 --trunc=1 --seeds=1,3,5,7 
+--network=pretrained_models/stylegan_human_v2_1024.pkl --version 2
+```
+
+
+2. We found the following troubleshooting links might be helpful: [1.](https://github.com/NVlabs/stylegan3), [2.](https://github.com/NVlabs/stylegan3/blob/main/docs/troubleshooting.md)
+
+### Train
+The training scripts are based on the original [stylegan1](https://github.com/NVlabs/stylegan), [stylegan2-ada](https://github.com/NVlabs/stylegan2-ada-pytorch), and [stylegan3](https://github.com/NVlabs/stylegan3) with minor changes. Here we only provide the scripts with modifications for SG2 and SG3. You can replace the old files with the provided scripts to train. (assume SHHQ-1.0 is placed under data/)
+
+#### Train Stylegan2-ada-pytorch with SHHQ-1.0
+```
+python train.py --outdir=training_results/sg2/ --data=data/SHHQ-1.0/ \
+    --gpus=8 --aug=noaug --mirror=1 --snap=250 --cfg=shhq --square=False
+```
+#### Train Stylegan3 with SHHQ-1.0 
+```
+python train.py --outdir=training_results/sg3/ --cfg=stylegan3-r --gpus=8 --batch=32 --gamma=12.4 \
+    --mirror=1 --aug=noaug --data=data/SHHQ-1.0/ --square=False --snap=250
+```
+
+### Pretrained models
+Please put the downloaded pretrained models [from above link](#Model-Zoo) under the folder 'pretrained_models'.
+
+
+### Generate full-body human images using our pretrained model
+```
+# Generate human full-body images without truncation
+python generate.py --outdir=outputs/generate/stylegan_human_v2_1024 --trunc=1 --seeds=1,3,5,7 --network=pretrained_models/stylegan_human_v2_1024.pkl --version 2
+
+# Generate human full-body images with truncation 
+python generate.py --outdir=outputs/generate/stylegan_human_v2_1024 --trunc=0.8 --seeds=0-10 --network=pretrained_models/stylegan_human_v2_1024.pkl --version 2
+
+# Generate human full-body images using stylegan V1
+python generate.py --outdir=outputs/generate/stylegan_human_v1_1024 --network=pretrained_models/stylegan_human_v1_1024.pkl --version 1 --seeds=1,3,5
+
+# Generate human full-body images using stylegan V3
+python generate.py --outdir=outputs/generate/stylegan_human_v3_512 --network=pretrained_models/stylegan_human_v3_512.pkl --version 3 --seeds=1,3,5
+```
+
+
+#### Note: The following demos are generated based on models related to StyleGAN V2 (stylegan_human_v2_512.pkl and stylegan_human_v2_1024.pkl). If you want to see results for V1 or V3, you need to change the loading method of the corresponding models.
+
+
+### Interpolation
+```
+python interpolation.py --network=pretrained_models/stylegan_human_v2_1024.pkl  --seeds=85,100 --outdir=outputs/inter_gifs
+```
+
+### Style-mixing **image** using stylegan2
+```
+python style_mixing.py --network=pretrained_models/stylegan_human_v2_1024.pkl --rows=85,100,75,458,1500 \\
+    --cols=55,821,1789,293 --styles=0-3 --outdir=outputs/stylemixing 
+```
+
+### Style-mixing **video** using stylegan2
+```
+python stylemixing_video.py --network=pretrained_models/stylegan_human_v2_1024.pkl --row-seed=3859 \\
+    --col-seeds=3098,31759,3791 --col-styles=8-12 --trunc=0.8 --outdir=outputs/stylemixing_video
+```
+
+### Aligned raw images
+For alignment, we use [openpose-pytorch](https://github.com/Hzzone/pytorch-openpose) for body-keypoints detection and [PaddlePaddle](https://github.com/PaddlePaddle/PaddleSeg/tree/release/2.5/contrib/PP-HumanSeg) for human segmentation.
+Before running the alignment script, few models need to be installed:
+1. download [body_pose_model.pth](https://drive.google.com/drive/folders/1JsvI4M4ZTg98fmnCZLFM-3TeovnCRElG?usp=sharing) and place it into openpose/model/.
+2. download and extract [deeplabv3p_resnet50_os8_humanseg_512x512_100k_with_softmax](https://paddleseg.bj.bcebos.com/dygraph/humanseg/export/deeplabv3p_resnet50_os8_humanseg_512x512_100k_with_softmax.zip) into PP_HumanSeg/export_model/deeplabv3p_resnet50_os8_humanseg_512x512_100k_with_softmax.
+3. download and extract [deeplabv3p_resnet50_os8_humanseg_512x512_100k](https://paddleseg.bj.bcebos.com/dygraph/humanseg/train/deeplabv3p_resnet50_os8_humanseg_512x512_100k.zip) into PP_HumanSeg/pretrained_model/deeplabv3p_resnet50_os8_humanseg_512x512_100k.
+4. install paddlepaddel: ``` pip install paddleseg ```
+
+Then you can start alignment: 
+```
+python alignment.py --image-folder img/test/ --output-folder aligned_image/
+```
+
+### Invert real image with [PTI](https://github.com/danielroich/PTI)
+Before inversion, please download our PTI weights: [e4e_w+.pt](https://drive.google.com/file/d/1NUfSJqLhsrU7c9PwAtlZ9xtrxhzS_6tu/view?usp=sharing) into /pti/.
+
+Few parameters you can change:
+- /pti/pti_configs/hyperparameters.py:
+    - first_inv_type = 'w+' -> Use pretrained e4e encoder
+    - first_inv_type = 'w'  -> Use projection and optimization
+- /pti/pti_configs/paths_config.py:
+    - input_data_path: path of real images
+    - e4e: path of e4e_w+.pt
+    - stylegan2_ada_shhq: pretrained stylegan2-ada model for SHHQ
+
+```
+python run_pti.py
+```
+Note: we used the test image under 'aligned_image/' (the output of alignment.py), the inverted latent code and fine-tuned generator will be saved in 'outputs/pti/'
+
+
+### Editing with InterfaceGAN, StyleSpace, and Sefa
+```
+python edit.py --network pretrained_models/stylegan_human_v2_1024.pkl --attr_name upper_length \\
+    --seeds 61531,61570,61571,61610 --outdir outputs/edit_results
+``` 
+
+### Editing using inverted latent code
+```
+python edit.py ---network outputs/pti/checkpoints/model_test.pkl --attr_name upper_length \\
+    --outdir outputs/edit_results --real True --real_w_path outputs/pti/embeddings/test/PTI/test/0.pt --real_img_path aligned_image/test.png
+```
+
+Note: 
+1. ''upper_length'' and ''bottom_length'' of ''attr_name'' are available for demo.
+2. Layers to control and editing strength are set in edit/edit_config.py.
+
+
+### Demo for [InsetGAN](https://arxiv.org/abs/2203.07293)
+
+We implement a quick demo using the key idea from InsetGAN: combining the face generated by FFHQ with the human-body generated by our pretrained model, optimizing both face and body latent codes to get a coherent full-body image.
+Before running the script, you need to download the [FFHQ face model]( https://docs.google.com/uc?export=download&confirm=t&id=125OG7SMkXI-Kf2aqiwLLHyCvSW-gZk3M), or you can use your own face model, as well as [pretrained face landmark](https://docs.google.com/uc?export=download&confirm=&id=1A82DnJBJzt8wI2J8ZrCK5fgHcQ2-tcWM) and [pretrained CNN face detection model for dlib](https://docs.google.com/uc?export=download&confirm=&id=1MduBgju5KFNrQfDLoQXJ_1_h5MnctCIG)
+```
+python insetgan.py --body_network=pretrained_models/stylegan_human_v2_1024.pkl --face_network=pretrained_models/ffhq.pkl \\
+    --body_seed=82 --face_seed=43  --trunc=0.6 --outdir=outputs/insetgan/ --video 1 
+```
+
+## Results
+
+### Editing with inverted real image
+(from left to right: real image | inverted image | InterFaceGAN result | StyleSpace result | SeFa result)
+
+https://user-images.githubusercontent.com/98547009/173773800-bb7fe54a-84d3-4b30-9864-a6b7b311f8ff.mp4
+
+
+### For more demo, please visit our [**web page**](https://stylegan-human.github.io/) .
+  
+
+## TODO List
+
+- [ ] Release 1024x512 version of StyleGAN-Human based on StyleGAN3 
+- [ ] Release 512x256 version of StyleGAN-Human based on StyleGAN1 
+- [ ] Extension of downstream application (InsetGAN): Add face inversion interface to support fusing user face image and stylegen-human body image
+- [x] Add Inversion Script into the provided editing pipeline
+- [ ] Release Dataset
+
+
+## Related Works
+* (SIGGRAPH 2022) **Text2Human: Text-Driven Controllable Human Image Generation**, Yuming Jiang et al. [[Paper](https://arxiv.org/pdf/2205.15996.pdf)], [[Code](https://github.com/yumingj/Text2Human)], [[Project Page](https://yumingj.github.io/projects/Text2Human.html)], [[Dataset](https://github.com/yumingj/DeepFashion-MultiModal)]
+* (ICCV 2021) **Talk-to-Edit: Fine-Grained Facial Editing via Dialog**, Yuming Jiang et al. [[Paper](https://arxiv.org/abs/2109.04425)], [[Code](https://github.com/yumingj/Talk-to-Edit)], [[Project Page](https://www.mmlab-ntu.com/project/talkedit/)], [[Dataset](https://mmlab.ie.cuhk.edu.hk/projects/CelebA/CelebA_Dialog.html)]
+* (Technical Report 2022) **Generalizable Neural Performer: Learning Robust Radiance Fields for Human Novel View Synthesis**, Wei Cheng et al. [[Paper](https://arxiv.org/pdf/2204.11798.pdf)], [[Code](https://github.com/generalizable-neural-performer/gnr)], [[Project Page](https://generalizable-neural-performer.github.io/)], [[Dataset](https://generalizable-neural-performer.github.io/genebody.html)]
+
+## Citation
+
+If you find this work useful for your research, please consider citing our paper:
+
+```bibtex
+@article{fu2022styleganhuman,
+      title={StyleGAN-Human: A Data-Centric Odyssey of Human Generation}, 
+      author={Fu, Jianglin and Li, Shikai and Jiang, Yuming and Lin, Kwan-Yee and Qian, Chen and Loy, Chen-Change and Wu, Wayne and Liu, Ziwei},
+      journal   = {arXiv preprint},
+      volume    = {arXiv:2204.11823},
+      year    = {2022}
+```
+
+## Acknowlegement
+Part of the code is borrowed from [stylegan (tensorflow)](https://github.com/NVlabs/stylegan), [stylegan2-ada (pytorch)](https://github.com/NVlabs/stylegan2-ada-pytorch), [stylegan3 (pytorch)](https://github.com/NVlabs/stylegan3).

stylegan_human/alignment.py

@@ -0,0 +1,223 @@
+# Copyright (c) SenseTime Research. All rights reserved.
+
+
+import os
+import argparse 
+import numpy as np
+import torch
+from torch.utils.data import DataLoader
+from torchvision.transforms import transforms
+from utils.ImagesDataset import ImagesDataset
+
+import cv2
+import time
+import copy
+import imutils
+
+# for openpose body keypoint detector : # (src:https://github.com/Hzzone/pytorch-openpose)
+from openpose.src import util
+from openpose.src.body import Body
+
+# for paddlepaddle human segmentation : #(src: https://github.com/PaddlePaddle/PaddleSeg/blob/release/2.5/contrib/PP-HumanSeg/)
+from PP_HumanSeg.deploy.infer import Predictor as PP_HumenSeg_Predictor
+
+import math
+def angle_between_points(p0,p1,p2):
+    if p0[1]==-1 or p1[1]==-1 or p2[1]==-1:
+        return -1
+    a = (p1[0]-p0[0])**2 + (p1[1]-p0[1])**2
+    b = (p1[0]-p2[0])**2 + (p1[1]-p2[1])**2
+    c = (p2[0]-p0[0])**2 + (p2[1]-p0[1])**2 
+    if a * b == 0:
+        return -1
+    return math.acos((a+b-c) / math.sqrt(4*a*b)) * 180 / math.pi
+
+
+def crop_img_with_padding(img, keypoints, rect):
+    person_xmin,person_xmax, ymin, ymax= rect
+    img_h,img_w,_ = img.shape    ## find body center using keypoints
+    middle_shoulder_x = keypoints[1][0]
+    middle_hip_x = (keypoints[8][0] + keypoints[11][0]) // 2
+    mid_x = (middle_hip_x + middle_shoulder_x) // 2    
+    mid_y = (ymin + ymax) // 2
+    ## find which side (l or r) is further than center x, use the further side
+    if abs(mid_x-person_xmin) > abs(person_xmax-mid_x): #left further
+        xmin = person_xmin
+        xmax = mid_x + (mid_x-person_xmin)
+    else:
+        ############### may be negtive
+        ### in this case, the script won't output any image, leave the case like this
+        ### since we don't want to pad human body
+        xmin = mid_x - (person_xmax-mid_x)   
+        xmax = person_xmax 
+
+    w = xmax - xmin
+    h = ymax - ymin
+    ## pad rectangle to w:h = 1:2 ## calculate desired border length
+    if h / w >= 2: #pad horizontally
+        target_w = h // 2
+        xmin_prime = int(mid_x - target_w / 2)
+        xmax_prime = int(mid_x + target_w / 2)
+        if xmin_prime < 0:
+            pad_left = abs(xmin_prime)# - xmin
+            xmin = 0
+        else:
+            pad_left = 0
+            xmin = xmin_prime
+        if xmax_prime > img_w:
+            pad_right = xmax_prime - img_w
+            xmax = img_w
+        else:
+            pad_right = 0
+            xmax = xmax_prime
+
+        cropped_img = img[int(ymin):int(ymax), int(xmin):int(xmax)]
+        im_pad = cv2.copyMakeBorder(cropped_img, 0, 0, int(pad_left),  int(pad_right), cv2.BORDER_REPLICATE) 
+    else: #pad vertically
+        target_h = w * 2
+        ymin_prime = mid_y - (target_h / 2)
+        ymax_prime = mid_y + (target_h / 2) 
+        if ymin_prime < 0: 
+            pad_up = abs(ymin_prime)# - ymin
+            ymin = 0
+        else:
+            pad_up = 0
+            ymin = ymin_prime
+        if ymax_prime > img_h:
+            pad_down = ymax_prime - img_h
+            ymax = img_h
+        else:
+            pad_down = 0
+            ymax = ymax_prime
+        print(ymin,ymax, xmin,xmax, img.shape)
+
+        cropped_img = img[int(ymin):int(ymax), int(xmin):int(xmax)]
+        im_pad = cv2.copyMakeBorder(cropped_img, int(pad_up), int(pad_down), 0,
+                                    0, cv2.BORDER_REPLICATE) 
+    result = cv2.resize(im_pad,(512,1024),interpolation = cv2.INTER_AREA)
+    return result
+
+
+def run(args):
+    os.makedirs(args.output_folder, exist_ok=True)
+    dataset = ImagesDataset(args.image_folder, transforms.Compose([transforms.ToTensor()]))
+    dataloader = DataLoader(dataset, batch_size=1, shuffle=False)
+
+    body_estimation = Body('openpose/model/body_pose_model.pth')
+
+    total = len(dataloader)
+    print('Num of dataloader : ', total)
+    os.makedirs(f'{args.output_folder}', exist_ok=True)
+    # os.makedirs(f'{args.output_folder}/middle_result', exist_ok=True)
+    
+    ## initialzide HumenSeg
+    human_seg_args = {}
+    human_seg_args['cfg'] = 'PP_HumanSeg/export_model/deeplabv3p_resnet50_os8_humanseg_512x512_100k_with_softmax/deploy.yaml'
+    human_seg_args['input_shape'] = [1024,512]
+    human_seg_args['save_dir'] = args.output_folder
+    human_seg_args['soft_predict'] = False
+    human_seg_args['use_gpu'] = True
+    human_seg_args['test_speed'] = False
+    human_seg_args['use_optic_flow'] = False
+    human_seg_args['add_argmax'] = True
+    human_seg_args= argparse.Namespace(**human_seg_args)
+    human_seg = PP_HumenSeg_Predictor(human_seg_args)
+
+    from tqdm import tqdm
+    for fname, image in tqdm(dataloader):
+        # try:
+        ## tensor to numpy image
+        fname = fname[0]
+        print(f'Processing \'{fname}\'.')
+        
+        image = (image.permute(0, 2, 3, 1) * 255).clamp(0, 255)
+        image = image.squeeze(0).numpy() # --> tensor to numpy, (H,W,C)
+        # avoid super high res img
+        if image.shape[0] >= 2000: # height  ### for shein image
+            ratio = image.shape[0]/1200 #height
+            dim = (int(image.shape[1]/ratio),1200)#(width, height)
+            image = cv2.resize(image, dim, interpolation = cv2.INTER_AREA)
+        image = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)
+
+        ## create segmentation
+        # mybg = cv2.imread('mybg.png') 
+        comb, segmentation, bg, ori_img = human_seg.run(image,None)  #mybg) 
+        # cv2.imwrite('comb.png',comb)  # [0,255]
+        # cv2.imwrite('alpha.png',segmentation*255) # segmentation [0,1] --> [0.255]
+        # cv2.imwrite('bg.png',bg)  #[0,255]
+        # cv2.imwrite('ori_img.png',ori_img) # [0,255]
+
+        masks_np = (segmentation* 255)# .byte().cpu().numpy() #1024,512,1
+        mask0_np = masks_np[:,:,0].astype(np.uint8)#[0, :, :]
+        contours = cv2.findContours(mask0_np,  cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
+        cnts = imutils.grab_contours(contours)
+        c = max(cnts, key=cv2.contourArea)
+        extTop = tuple(c[c[:, :, 1].argmin()][0])
+        extBot = tuple(c[c[:, :, 1].argmax()][0])
+        extBot = list(extBot)
+        extTop = list(extTop)
+        pad_range = int((extBot[1]-extTop[1])*0.05)
+        if (int(extTop[1])<=5 and int(extTop[1])>0) and (comb.shape[0]>int(extBot[1]) and int(extBot[1])>=comb.shape[0]-5): #seg mask already reaches to the edge
+            #pad with pure white, top 100 px, bottom 100 px
+            comb= cv2.copyMakeBorder(comb,pad_range+5,pad_range+5,0,0,cv2.BORDER_CONSTANT,value=[255,255,255]) 
+        elif int(extTop[1])<=0 or int(extBot[1])>=comb.shape[0]:
+            print('PAD: body out of boundary', fname) #should not happened
+            return {}
+        else:
+            comb = cv2.copyMakeBorder(comb, pad_range+5, pad_range+5, 0, 0, cv2.BORDER_REPLICATE) #105 instead of 100: give some extra space
+        extBot[1] = extBot[1] + pad_range+5
+        extTop[1] = extTop[1] + pad_range+5
+
+        extLeft = tuple(c[c[:, :, 0].argmin()][0])
+        extRight = tuple(c[c[:, :, 0].argmax()][0])
+        extLeft = list(extLeft)
+        extRight = list(extRight)
+        person_ymin = int(extTop[1])-pad_range # 100
+        person_ymax = int(extBot[1])+pad_range # 100 #height
+        if person_ymin<0 or person_ymax>comb.shape[0]: # out of range
+            return {}
+        person_xmin = int(extLeft[0])
+        person_xmax = int(extRight[0])
+        rect =  [person_xmin,person_xmax,person_ymin, person_ymax]
+        # recimg = copy.deepcopy(comb)
+        # cv2.rectangle(recimg,(person_xmin,person_ymin),(person_xmax,person_ymax),(0,255,0),2)
+        # cv2.imwrite(f'{args.output_folder}/middle_result/{fname}_rec.png',recimg)
+
+        ## detect keypoints
+        keypoints, subset = body_estimation(comb)
+        # print(keypoints, subset, len(subset))
+        if len(subset) != 1 or (len(subset)==1 and subset[0][-1]<15): 
+            print(f'Processing \'{fname}\'. Please import image contains one person only. Also can check segmentation mask. ')
+            continue
+
+        # canvas = copy.deepcopy(comb)
+        # canvas = util.draw_bodypose(canvas, keypoints, subset, show_number=True)
+        # cv2.imwrite(f'{args.output_folder}/middle_result/{fname}_keypoints.png',canvas)
+
+        comb = crop_img_with_padding(comb, keypoints, rect)
+
+        
+        cv2.imwrite(f'{args.output_folder}/{fname}.png', comb)
+        print(f' -- Finished processing \'{fname}\'. --')
+        # except:
+        #     print(f'Processing \'{fname}\'. Not satisfied the alignment strategy.')
+        
+        
+if __name__ == '__main__':
+    torch.backends.cudnn.benchmark = True
+    torch.backends.cudnn.deterministic = False
+    
+    t1 = time.time()
+    arg_formatter = argparse.ArgumentDefaultsHelpFormatter
+    description = 'StyleGAN-Human data process'
+    parser = argparse.ArgumentParser(formatter_class=arg_formatter,
+                                     description=description)
+    parser.add_argument('--image-folder', type=str, dest='image_folder')
+    parser.add_argument('--output-folder', dest='output_folder', default='results', type=str)
+    # parser.add_argument('--cfg', dest='cfg for segmentation', default='PP_HumanSeg/export_model/ppseg_lite_portrait_398x224_with_softmax/deploy.yaml', type=str)
+
+    print('parsing arguments')
+    cmd_args = parser.parse_args()
+    run(cmd_args)
+
+    print('total time elapsed: ', str(time.time() - t1))

stylegan_human/bg_white.py

@@ -0,0 +1,57 @@
+# Copyright (c) SenseTime Research. All rights reserved.
+
+import os
+import click
+import cv2
+import numpy as np
+
+def bg_white(seg, raw, blur_level=3, gaussian=81):
+    seg = cv2.blur(seg, (blur_level, blur_level))
+
+    empty = np.ones_like(seg)
+    seg_bg = (empty - seg) * 255 
+    seg_bg = cv2.GaussianBlur(seg_bg,(gaussian,gaussian),0)
+
+    background_mask = cv2.cvtColor(255 - cv2.cvtColor(seg, cv2.COLOR_BGR2GRAY), cv2.COLOR_GRAY2BGR)
+    masked_fg = (raw * (1 / 255)) * (seg * (1 / 255))
+    masked_bg = (seg_bg * (1 / 255)) * (background_mask * (1 / 255))
+
+    frame = np.uint8(cv2.add(masked_bg,masked_fg)*255)
+
+    return frame
+
+
+"""
+To turn background into white.
+
+Examples:
+
+\b
+python bg_white.py  --raw_img_dir=./SHHQ-1.0/no_segment/ --raw_seg_dir=./SHHQ-1.0/segments/ \\
+    --outdir=./SHHQ-1.0/bg_white/
+"""
+
+@click.command()
+@click.pass_context
+@click.option('--raw_img_dir', default="./SHHQ-1.0/no_segment/", help='folder of raw image', required=True)
+@click.option('--raw_seg_dir', default='./SHHQ-1.0/segments/', help='folder of segmentation masks', required=True)
+@click.option('--outdir', help='Where to save the output images', default= "./SHHQ-1.0/bg_white/" , type=str, required=True, metavar='DIR')
+
+def main(
+        ctx: click.Context,
+        raw_img_dir: str,
+        raw_seg_dir: str,
+        outdir: str):
+    os.makedirs(outdir, exist_ok=True)
+    files = os.listdir(raw_img_dir)
+    for file in files: 
+        print(file)
+        raw = cv2.imread(os.path.join(raw_img_dir, file))
+        seg = cv2.imread(os.path.join(raw_seg_dir, file))
+        assert raw is not None
+        assert seg is not None
+        white_frame = bg_white(seg, raw)
+        cv2.imwrite(os.path.join(outdir,file), white_frame)
+
+if __name__ == "__main__":
+    main()

stylegan_human/dnnlib/__init__.py

@@ -0,0 +1,11 @@
+# Copyright (c) SenseTime Research. All rights reserved.
+
+# Copyright (c) 2021, NVIDIA CORPORATION.  All rights reserved.
+#
+# NVIDIA CORPORATION and its licensors retain all intellectual property
+# and proprietary rights in and to this software, related documentation
+# and any modifications thereto.  Any use, reproduction, disclosure or
+# distribution of this software and related documentation without an express
+# license agreement from NVIDIA CORPORATION is strictly prohibited.
+
+from .util import EasyDict, make_cache_dir_path

stylegan_human/dnnlib/tflib/__init__.py

@@ -0,0 +1,20 @@
+# Copyright (c) SenseTime Research. All rights reserved.
+
+# Copyright (c) 2019, NVIDIA Corporation. All rights reserved.
+#
+# This work is made available under the Nvidia Source Code License-NC.
+# To view a copy of this license, visit
+# https://nvlabs.github.io/stylegan2/license.html
+
+from . import autosummary
+from . import network
+from . import optimizer
+from . import tfutil
+from . import custom_ops
+
+from .tfutil import *
+from .network import Network
+
+from .optimizer import Optimizer
+
+from .custom_ops import get_plugin

stylegan_human/dnnlib/tflib/autosummary.py

@@ -0,0 +1,193 @@
+# Copyright (c) SenseTime Research. All rights reserved.
+
+# Copyright (c) 2019, NVIDIA Corporation. All rights reserved.
+#
+# This work is made available under the Nvidia Source Code License-NC.
+# To view a copy of this license, visit
+# https://nvlabs.github.io/stylegan2/license.html
+
+"""Helper for adding automatically tracked values to Tensorboard.
+
+Autosummary creates an identity op that internally keeps track of the input
+values and automatically shows up in TensorBoard. The reported value
+represents an average over input components. The average is accumulated
+constantly over time and flushed when save_summaries() is called.
+
+Notes:
+- The output tensor must be used as an input for something else in the
+  graph. Otherwise, the autosummary op will not get executed, and the average
+  value will not get accumulated.
+- It is perfectly fine to include autosummaries with the same name in
+  several places throughout the graph, even if they are executed concurrently.
+- It is ok to also pass in a python scalar or numpy array. In this case, it
+  is added to the average immediately.
+"""
+
+from collections import OrderedDict
+import numpy as np
+import tensorflow as tf
+from tensorboard import summary as summary_lib
+from tensorboard.plugins.custom_scalar import layout_pb2
+
+from . import tfutil
+from .tfutil import TfExpression
+from .tfutil import TfExpressionEx
+
+# Enable "Custom scalars" tab in TensorBoard for advanced formatting.
+# Disabled by default to reduce tfevents file size.
+enable_custom_scalars = False
+
+_dtype = tf.float64
+_vars = OrderedDict()  # name => [var, ...]
+_immediate = OrderedDict()  # name => update_op, update_value
+_finalized = False
+_merge_op = None
+
+
+def _create_var(name: str, value_expr: TfExpression) -> TfExpression:
+    """Internal helper for creating autosummary accumulators."""
+    assert not _finalized
+    name_id = name.replace("/", "_")
+    v = tf.cast(value_expr, _dtype)
+
+    if v.shape.is_fully_defined():
+        size = np.prod(v.shape.as_list())
+        size_expr = tf.constant(size, dtype=_dtype)
+    else:
+        size = None
+        size_expr = tf.reduce_prod(tf.cast(tf.shape(v), _dtype))
+
+    if size == 1:
+        if v.shape.ndims != 0:
+            v = tf.reshape(v, [])
+        v = [size_expr, v, tf.square(v)]
+    else:
+        v = [size_expr, tf.reduce_sum(v), tf.reduce_sum(tf.square(v))]
+    v = tf.cond(tf.is_finite(v[1]), lambda: tf.stack(v), lambda: tf.zeros(3, dtype=_dtype))
+
+    with tfutil.absolute_name_scope("Autosummary/" + name_id), tf.control_dependencies(None):
+        var = tf.Variable(tf.zeros(3, dtype=_dtype), trainable=False)  # [sum(1), sum(x), sum(x**2)]
+    update_op = tf.cond(tf.is_variable_initialized(var), lambda: tf.assign_add(var, v), lambda: tf.assign(var, v))
+
+    if name in _vars:
+        _vars[name].append(var)
+    else:
+        _vars[name] = [var]
+    return update_op
+
+
+def autosummary(name: str, value: TfExpressionEx, passthru: TfExpressionEx = None, condition: TfExpressionEx = True) -> TfExpressionEx:
+    """Create a new autosummary.
+
+    Args:
+        name:     Name to use in TensorBoard
+        value:    TensorFlow expression or python value to track
+        passthru: Optionally return this TF node without modifications but tack an autosummary update side-effect to this node.
+
+    Example use of the passthru mechanism:
+
+    n = autosummary('l2loss', loss, passthru=n)
+
+    This is a shorthand for the following code:
+
+    with tf.control_dependencies([autosummary('l2loss', loss)]):
+        n = tf.identity(n)
+    """
+    tfutil.assert_tf_initialized()
+    name_id = name.replace("/", "_")
+
+    if tfutil.is_tf_expression(value):
+        with tf.name_scope("summary_" + name_id), tf.device(value.device):
+            condition = tf.convert_to_tensor(condition, name='condition')
+            update_op = tf.cond(condition, lambda: tf.group(_create_var(name, value)), tf.no_op)
+            with tf.control_dependencies([update_op]):
+                return tf.identity(value if passthru is None else passthru)
+
+    else:  # python scalar or numpy array
+        assert not tfutil.is_tf_expression(passthru)
+        assert not tfutil.is_tf_expression(condition)
+        if condition:
+            if name not in _immediate:
+                with tfutil.absolute_name_scope("Autosummary/" + name_id), tf.device(None), tf.control_dependencies(None):
+                    update_value = tf.placeholder(_dtype)
+                    update_op = _create_var(name, update_value)
+                    _immediate[name] = update_op, update_value
+            update_op, update_value = _immediate[name]
+            tfutil.run(update_op, {update_value: value})
+        return value if passthru is None else passthru
+
+
+def finalize_autosummaries() -> None:
+    """Create the necessary ops to include autosummaries in TensorBoard report.
+    Note: This should be done only once per graph.
+    """
+    global _finalized
+    tfutil.assert_tf_initialized()
+
+    if _finalized:
+        return None
+
+    _finalized = True
+    tfutil.init_uninitialized_vars([var for vars_list in _vars.values() for var in vars_list])
+
+    # Create summary ops.
+    with tf.device(None), tf.control_dependencies(None):
+        for name, vars_list in _vars.items():
+            name_id = name.replace("/", "_")
+            with tfutil.absolute_name_scope("Autosummary/" + name_id):
+                moments = tf.add_n(vars_list)
+                moments /= moments[0]
+                with tf.control_dependencies([moments]):  # read before resetting
+                    reset_ops = [tf.assign(var, tf.zeros(3, dtype=_dtype)) for var in vars_list]
+                    with tf.name_scope(None), tf.control_dependencies(reset_ops):  # reset before reporting
+                        mean = moments[1]
+                        std = tf.sqrt(moments[2] - tf.square(moments[1]))
+                        tf.summary.scalar(name, mean)
+                        if enable_custom_scalars:
+                            tf.summary.scalar("xCustomScalars/" + name + "/margin_lo", mean - std)
+                            tf.summary.scalar("xCustomScalars/" + name + "/margin_hi", mean + std)
+
+    # Setup layout for custom scalars.
+    layout = None
+    if enable_custom_scalars:
+        cat_dict = OrderedDict()
+        for series_name in sorted(_vars.keys()):
+            p = series_name.split("/")
+            cat = p[0] if len(p) >= 2 else ""
+            chart = "/".join(p[1:-1]) if len(p) >= 3 else p[-1]
+            if cat not in cat_dict:
+                cat_dict[cat] = OrderedDict()
+            if chart not in cat_dict[cat]:
+                cat_dict[cat][chart] = []
+            cat_dict[cat][chart].append(series_name)
+        categories = []
+        for cat_name, chart_dict in cat_dict.items():
+            charts = []
+            for chart_name, series_names in chart_dict.items():
+                series = []
+                for series_name in series_names:
+                    series.append(layout_pb2.MarginChartContent.Series(
+                        value=series_name,
+                        lower="xCustomScalars/" + series_name + "/margin_lo",
+                        upper="xCustomScalars/" + series_name + "/margin_hi"))
+                margin = layout_pb2.MarginChartContent(series=series)
+                charts.append(layout_pb2.Chart(title=chart_name, margin=margin))
+            categories.append(layout_pb2.Category(title=cat_name, chart=charts))
+        layout = summary_lib.custom_scalar_pb(layout_pb2.Layout(category=categories))
+    return layout
+
+def save_summaries(file_writer, global_step=None):
+    """Call FileWriter.add_summary() with all summaries in the default graph,
+    automatically finalizing and merging them on the first call.
+    """
+    global _merge_op
+    tfutil.assert_tf_initialized()
+
+    if _merge_op is None:
+        layout = finalize_autosummaries()
+        if layout is not None:
+            file_writer.add_summary(layout)
+        with tf.device(None), tf.control_dependencies(None):
+            _merge_op = tf.summary.merge_all()
+
+    file_writer.add_summary(_merge_op.eval(), global_step)

stylegan_human/dnnlib/tflib/custom_ops.py

@@ -0,0 +1,171 @@
+# Copyright (c) SenseTime Research. All rights reserved.
+
+# Copyright (c) 2019, NVIDIA Corporation. All rights reserved.
+#
+# This work is made available under the Nvidia Source Code License-NC.
+# To view a copy of this license, visit
+# https://nvlabs.github.io/stylegan2/license.html
+
+"""TensorFlow custom ops builder.
+"""
+
+import os
+import re
+import uuid
+import hashlib
+import tempfile
+import shutil
+import tensorflow as tf
+from tensorflow.python.client import device_lib # pylint: disable=no-name-in-module
+
+#----------------------------------------------------------------------------
+# Global options.
+
+cuda_cache_path = os.path.join(os.path.dirname(__file__), '_cudacache')
+cuda_cache_version_tag = 'v1'
+do_not_hash_included_headers = False # Speed up compilation by assuming that headers included by the CUDA code never change. Unsafe!
+verbose = True # Print status messages to stdout.
+
+compiler_bindir_search_path = [
+    'C:/Program Files (x86)/Microsoft Visual Studio/2017/Community/VC/Tools/MSVC/14.14.26428/bin/Hostx64/x64',
+    'C:/Program Files (x86)/Microsoft Visual Studio/2019/Community/VC/Tools/MSVC/14.23.28105/bin/Hostx64/x64',
+    'C:/Program Files (x86)/Microsoft Visual Studio 14.0/vc/bin',
+]
+
+#----------------------------------------------------------------------------
+# Internal helper funcs.
+
+def _find_compiler_bindir():
+    for compiler_path in compiler_bindir_search_path:
+        if os.path.isdir(compiler_path):
+            return compiler_path
+    return None
+
+def _get_compute_cap(device):
+    caps_str = device.physical_device_desc
+    m = re.search('compute capability: (\\d+).(\\d+)', caps_str)
+    major = m.group(1)
+    minor = m.group(2)
+    return (major, minor)
+
+def _get_cuda_gpu_arch_string():
+    gpus = [x for x in device_lib.list_local_devices() if x.device_type == 'GPU']
+    if len(gpus) == 0:
+        raise RuntimeError('No GPU devices found')
+    (major, minor) = _get_compute_cap(gpus[0])
+    return 'sm_%s%s' % (major, minor)
+
+def _run_cmd(cmd):
+    with os.popen(cmd) as pipe:
+        output = pipe.read()
+        status = pipe.close()
+    if status is not None:
+        raise RuntimeError('NVCC returned an error. See below for full command line and output log:\n\n%s\n\n%s' % (cmd, output))
+
+def _prepare_nvcc_cli(opts):
+    cmd = 'nvcc ' + opts.strip()
+    cmd += ' --disable-warnings'
+    cmd += ' --include-path "%s"' % tf.sysconfig.get_include()
+    cmd += ' --include-path "%s"' % os.path.join(tf.sysconfig.get_include(), 'external', 'protobuf_archive', 'src')
+    cmd += ' --include-path "%s"' % os.path.join(tf.sysconfig.get_include(), 'external', 'com_google_absl')
+    cmd += ' --include-path "%s"' % os.path.join(tf.sysconfig.get_include(), 'external', 'eigen_archive')
+
+    compiler_bindir = _find_compiler_bindir()
+    if compiler_bindir is None:
+        # Require that _find_compiler_bindir succeeds on Windows.  Allow
+        # nvcc to use whatever is the default on Linux.
+        if os.name == 'nt':
+            raise RuntimeError('Could not find MSVC/GCC/CLANG installation on this computer. Check compiler_bindir_search_path list in "%s".' % __file__)
+    else:
+        cmd += ' --compiler-bindir "%s"' % compiler_bindir
+    cmd += ' 2>&1'
+    return cmd
+
+#----------------------------------------------------------------------------
+# Main entry point.
+
+_plugin_cache = dict()
+
+def get_plugin(cuda_file):
+    cuda_file_base = os.path.basename(cuda_file)
+    cuda_file_name, cuda_file_ext = os.path.splitext(cuda_file_base)
+
+    # Already in cache?
+    if cuda_file in _plugin_cache:
+        return _plugin_cache[cuda_file]
+
+    # Setup plugin.
+    if verbose:
+        print('Setting up TensorFlow plugin "%s": ' % cuda_file_base, end='', flush=True)
+    try:
+        # Hash CUDA source.
+        md5 = hashlib.md5()
+        with open(cuda_file, 'rb') as f:
+            md5.update(f.read())
+        md5.update(b'\n')
+
+        # Hash headers included by the CUDA code by running it through the preprocessor.
+        if not do_not_hash_included_headers:
+            if verbose:
+                print('Preprocessing... ', end='', flush=True)
+            with tempfile.TemporaryDirectory() as tmp_dir:
+                tmp_file = os.path.join(tmp_dir, cuda_file_name + '_tmp' + cuda_file_ext)
+                _run_cmd(_prepare_nvcc_cli('"%s" --preprocess -o "%s" --keep --keep-dir "%s"' % (cuda_file, tmp_file, tmp_dir)))
+                with open(tmp_file, 'rb') as f:
+                    bad_file_str = ('"' + cuda_file.replace('\\', '/') + '"').encode('utf-8') # __FILE__ in error check macros
+                    good_file_str = ('"' + cuda_file_base + '"').encode('utf-8')
+                    for ln in f:
+                        if not ln.startswith(b'# ') and not ln.startswith(b'#line '): # ignore line number pragmas
+                            ln = ln.replace(bad_file_str, good_file_str)
+                            md5.update(ln)
+                    md5.update(b'\n')
+
+        # Select compiler options.
+        compile_opts = ''
+        if os.name == 'nt':
+            compile_opts += '"%s"' % os.path.join(tf.sysconfig.get_lib(), 'python', '_pywrap_tensorflow_internal.lib')
+        elif os.name == 'posix':
+            compile_opts += '"%s"' % os.path.join(tf.sysconfig.get_lib(), 'python', '_pywrap_tensorflow_internal.so')
+            compile_opts += ' --compiler-options \'-fPIC -D_GLIBCXX_USE_CXX11_ABI=0\''
+        else:
+            assert False # not Windows or Linux, w00t?
+        compile_opts += ' --gpu-architecture=%s' % _get_cuda_gpu_arch_string()
+        compile_opts += ' --use_fast_math'
+        nvcc_cmd = _prepare_nvcc_cli(compile_opts)
+
+        # Hash build configuration.
+        md5.update(('nvcc_cmd: ' + nvcc_cmd).encode('utf-8') + b'\n')
+        md5.update(('tf.VERSION: ' + tf.VERSION).encode('utf-8') + b'\n')
+        md5.update(('cuda_cache_version_tag: ' + cuda_cache_version_tag).encode('utf-8') + b'\n')
+
+        # Compile if not already compiled.
+        bin_file_ext = '.dll' if os.name == 'nt' else '.so'
+        bin_file = os.path.join(cuda_cache_path, cuda_file_name + '_' + md5.hexdigest() + bin_file_ext)
+        if not os.path.isfile(bin_file):
+            if verbose:
+                print('Compiling... ', end='', flush=True)
+            with tempfile.TemporaryDirectory() as tmp_dir:
+                tmp_file = os.path.join(tmp_dir, cuda_file_name + '_tmp' + bin_file_ext)
+                _run_cmd(nvcc_cmd + ' "%s" --shared -o "%s" --keep --keep-dir "%s"' % (cuda_file, tmp_file, tmp_dir))
+                os.makedirs(cuda_cache_path, exist_ok=True)
+                intermediate_file = os.path.join(cuda_cache_path, cuda_file_name + '_' + uuid.uuid4().hex + '_tmp' + bin_file_ext)
+                shutil.copyfile(tmp_file, intermediate_file)
+                os.rename(intermediate_file, bin_file) # atomic
+
+        # Load.
+        if verbose:
+            print('Loading... ', end='', flush=True)
+        plugin = tf.load_op_library(bin_file)
+
+        # Add to cache.
+        _plugin_cache[cuda_file] = plugin
+        if verbose:
+            print('Done.', flush=True)
+        return plugin
+
+    except:
+        if verbose:
+            print('Failed!', flush=True)
+        raise
+
+#----------------------------------------------------------------------------

stylegan_human/dnnlib/tflib/network.py

@@ -0,0 +1,592 @@
+# Copyright (c) SenseTime Research. All rights reserved.
+
+# Copyright (c) 2019, NVIDIA Corporation. All rights reserved.
+#
+# This work is made available under the Nvidia Source Code License-NC.
+# To view a copy of this license, visit
+# https://nvlabs.github.io/stylegan2/license.html
+
+"""Helper for managing networks."""
+
+import types
+import inspect
+import re
+import uuid
+import sys
+import numpy as np
+import tensorflow as tf
+
+from collections import OrderedDict
+from typing import Any, List, Tuple, Union
+
+from . import tfutil
+from .. import util
+
+from .tfutil import TfExpression, TfExpressionEx
+
+_import_handlers = []  # Custom import handlers for dealing with legacy data in pickle import.
+_import_module_src = dict()  # Source code for temporary modules created during pickle import.
+
+
+def import_handler(handler_func):
+    """Function decorator for declaring custom import handlers."""
+    _import_handlers.append(handler_func)
+    return handler_func
+
+
+class Network:
+    """Generic network abstraction.
+
+    Acts as a convenience wrapper for a parameterized network construction
+    function, providing several utility methods and convenient access to
+    the inputs/outputs/weights.
+
+    Network objects can be safely pickled and unpickled for long-term
+    archival purposes. The pickling works reliably as long as the underlying
+    network construction function is defined in a standalone Python module
+    that has no side effects or application-specific imports.
+
+    Args:
+        name: Network name. Used to select TensorFlow name and variable scopes.
+        func_name: Fully qualified name of the underlying network construction function, or a top-level function object.
+        static_kwargs: Keyword arguments to be passed in to the network construction function.
+
+    Attributes:
+        name: User-specified name, defaults to build func name if None.
+        scope: Unique TensorFlow scope containing template graph and variables, derived from the user-specified name.
+        static_kwargs: Arguments passed to the user-supplied build func.
+        components: Container for sub-networks. Passed to the build func, and retained between calls.
+        num_inputs: Number of input tensors.
+        num_outputs: Number of output tensors.
+        input_shapes: Input tensor shapes (NC or NCHW), including minibatch dimension.
+        output_shapes: Output tensor shapes (NC or NCHW), including minibatch dimension.
+        input_shape: Short-hand for input_shapes[0].
+        output_shape: Short-hand for output_shapes[0].
+        input_templates: Input placeholders in the template graph.
+        output_templates: Output tensors in the template graph.
+        input_names: Name string for each input.
+        output_names: Name string for each output.
+        own_vars: Variables defined by this network (local_name => var), excluding sub-networks.
+        vars: All variables (local_name => var).
+        trainables: All trainable variables (local_name => var).
+        var_global_to_local: Mapping from variable global names to local names.
+    """
+
+    def __init__(self, name: str = None, func_name: Any = None, **static_kwargs):
+        tfutil.assert_tf_initialized()
+        assert isinstance(name, str) or name is None
+        assert func_name is not None
+        assert isinstance(func_name, str) or util.is_top_level_function(func_name)
+        assert util.is_pickleable(static_kwargs)
+
+        self._init_fields()
+        self.name = name
+        self.static_kwargs = util.EasyDict(static_kwargs)
+
+        # Locate the user-specified network build function.
+        if util.is_top_level_function(func_name):
+            func_name = util.get_top_level_function_name(func_name)
+        module, self._build_func_name = util.get_module_from_obj_name(func_name)
+        self._build_func = util.get_obj_from_module(module, self._build_func_name)
+        assert callable(self._build_func)
+
+        # Dig up source code for the module containing the build function.
+        self._build_module_src = _import_module_src.get(module, None)
+        if self._build_module_src is None:
+            self._build_module_src = inspect.getsource(module)
+
+        # Init TensorFlow graph.
+        self._init_graph()
+        self.reset_own_vars()
+
+    def _init_fields(self) -> None:
+        self.name = None
+        self.scope = None
+        self.static_kwargs = util.EasyDict()
+        self.components = util.EasyDict()
+        self.num_inputs = 0
+        self.num_outputs = 0
+        self.input_shapes = [[]]
+        self.output_shapes = [[]]
+        self.input_shape = []
+        self.output_shape = []
+        self.input_templates = []
+        self.output_templates = []
+        self.input_names = []
+        self.output_names = []
+        self.own_vars = OrderedDict()
+        self.vars = OrderedDict()
+        self.trainables = OrderedDict()
+        self.var_global_to_local = OrderedDict()
+
+        self._build_func = None  # User-supplied build function that constructs the network.
+        self._build_func_name = None  # Name of the build function.
+        self._build_module_src = None  # Full source code of the module containing the build function.
+        self._run_cache = dict()  # Cached graph data for Network.run().
+
+    def _init_graph(self) -> None:
+        # Collect inputs.
+        self.input_names = []
+
+        for param in inspect.signature(self._build_func).parameters.values():
+            if param.kind == param.POSITIONAL_OR_KEYWORD and param.default is param.empty:
+                self.input_names.append(param.name)
+
+        self.num_inputs = len(self.input_names)
+        assert self.num_inputs >= 1
+
+        # Choose name and scope.
+        if self.name is None:
+            self.name = self._build_func_name
+        assert re.match("^[A-Za-z0-9_.\\-]*$", self.name)
+        with tf.name_scope(None):
+            self.scope = tf.get_default_graph().unique_name(self.name, mark_as_used=True)
+
+        # Finalize build func kwargs.
+        build_kwargs = dict(self.static_kwargs)
+        build_kwargs["is_template_graph"] = True
+        build_kwargs["components"] = self.components
+
+        # Build template graph.
+        with tfutil.absolute_variable_scope(self.scope, reuse=False), tfutil.absolute_name_scope(self.scope):  # ignore surrounding scopes
+            assert tf.get_variable_scope().name == self.scope
+            assert tf.get_default_graph().get_name_scope() == self.scope
+            with tf.control_dependencies(None):  # ignore surrounding control dependencies
+                self.input_templates = [tf.placeholder(tf.float32, name=name) for name in self.input_names]
+                out_expr = self._build_func(*self.input_templates, **build_kwargs)
+
+        # Collect outputs.
+        assert tfutil.is_tf_expression(out_expr) or isinstance(out_expr, tuple)
+        self.output_templates = [out_expr] if tfutil.is_tf_expression(out_expr) else list(out_expr)
+        self.num_outputs = len(self.output_templates)
+        assert self.num_outputs >= 1
+        assert all(tfutil.is_tf_expression(t) for t in self.output_templates)
+
+        # Perform sanity checks.
+        if any(t.shape.ndims is None for t in self.input_templates):
+            raise ValueError("Network input shapes not defined. Please call x.set_shape() for each input.")
+        if any(t.shape.ndims is None for t in self.output_templates):
+            raise ValueError("Network output shapes not defined. Please call x.set_shape() where applicable.")
+        if any(not isinstance(comp, Network) for comp in self.components.values()):
+            raise ValueError("Components of a Network must be Networks themselves.")
+        if len(self.components) != len(set(comp.name for comp in self.components.values())):
+            raise ValueError("Components of a Network must have unique names.")
+
+        # List inputs and outputs.
+        self.input_shapes = [t.shape.as_list() for t in self.input_templates]
+        self.output_shapes = [t.shape.as_list() for t in self.output_templates]
+        self.input_shape = self.input_shapes[0]
+        self.output_shape = self.output_shapes[0]
+        self.output_names = [t.name.split("/")[-1].split(":")[0] for t in self.output_templates]
+
+        # List variables.
+        self.own_vars = OrderedDict((var.name[len(self.scope) + 1:].split(":")[0], var) for var in tf.global_variables(self.scope + "/"))
+        self.vars = OrderedDict(self.own_vars)
+        self.vars.update((comp.name + "/" + name, var) for comp in self.components.values() for name, var in comp.vars.items())
+        self.trainables = OrderedDict((name, var) for name, var in self.vars.items() if var.trainable)
+        self.var_global_to_local = OrderedDict((var.name.split(":")[0], name) for name, var in self.vars.items())
+
+    def reset_own_vars(self) -> None:
+        """Re-initialize all variables of this network, excluding sub-networks."""
+        tfutil.run([var.initializer for var in self.own_vars.values()])
+
+    def reset_vars(self) -> None:
+        """Re-initialize all variables of this network, including sub-networks."""
+        tfutil.run([var.initializer for var in self.vars.values()])
+
+    def reset_trainables(self) -> None:
+        """Re-initialize all trainable variables of this network, including sub-networks."""
+        tfutil.run([var.initializer for var in self.trainables.values()])
+
+    def get_output_for(self, *in_expr: TfExpression, return_as_list: bool = False, **dynamic_kwargs) -> Union[TfExpression, List[TfExpression]]:
+        """Construct TensorFlow expression(s) for the output(s) of this network, given the input expression(s)."""
+        assert len(in_expr) == self.num_inputs
+        assert not all(expr is None for expr in in_expr)
+
+        # Finalize build func kwargs.
+        build_kwargs = dict(self.static_kwargs)
+        build_kwargs.update(dynamic_kwargs)
+        build_kwargs["is_template_graph"] = False
+        build_kwargs["components"] = self.components
+
+        # Build TensorFlow graph to evaluate the network.
+        with tfutil.absolute_variable_scope(self.scope, reuse=True), tf.name_scope(self.name):
+            assert tf.get_variable_scope().name == self.scope
+            valid_inputs = [expr for expr in in_expr if expr is not None]
+            final_inputs = []
+            for expr, name, shape in zip(in_expr, self.input_names, self.input_shapes):
+                if expr is not None:
+                    expr = tf.identity(expr, name=name)
+                else:
+                    expr = tf.zeros([tf.shape(valid_inputs[0])[0]] + shape[1:], name=name)
+                final_inputs.append(expr)
+            out_expr = self._build_func(*final_inputs, **build_kwargs)
+
+        # Propagate input shapes back to the user-specified expressions.
+        for expr, final in zip(in_expr, final_inputs):
+            if isinstance(expr, tf.Tensor):
+                expr.set_shape(final.shape)
+
+        # Express outputs in the desired format.
+        assert tfutil.is_tf_expression(out_expr) or isinstance(out_expr, tuple)
+        if return_as_list:
+            out_expr = [out_expr] if tfutil.is_tf_expression(out_expr) else list(out_expr)
+        return out_expr
+
+    def get_var_local_name(self, var_or_global_name: Union[TfExpression, str]) -> str:
+        """Get the local name of a given variable, without any surrounding name scopes."""
+        assert tfutil.is_tf_expression(var_or_global_name) or isinstance(var_or_global_name, str)
+        global_name = var_or_global_name if isinstance(var_or_global_name, str) else var_or_global_name.name
+        return self.var_global_to_local[global_name]
+
+    def find_var(self, var_or_local_name: Union[TfExpression, str]) -> TfExpression:
+        """Find variable by local or global name."""
+        assert tfutil.is_tf_expression(var_or_local_name) or isinstance(var_or_local_name, str)
+        return self.vars[var_or_local_name] if isinstance(var_or_local_name, str) else var_or_local_name
+
+    def get_var(self, var_or_local_name: Union[TfExpression, str]) -> np.ndarray:
+        """Get the value of a given variable as NumPy array.
+        Note: This method is very inefficient -- prefer to use tflib.run(list_of_vars) whenever possible."""
+        return self.find_var(var_or_local_name).eval()
+
+    def set_var(self, var_or_local_name: Union[TfExpression, str], new_value: Union[int, float, np.ndarray]) -> None:
+        """Set the value of a given variable based on the given NumPy array.
+        Note: This method is very inefficient -- prefer to use tflib.set_vars() whenever possible."""
+        tfutil.set_vars({self.find_var(var_or_local_name): new_value})
+
+    def __getstate__(self) -> dict:
+        """Pickle export."""
+        state = dict()
+        state["version"]            = 4
+        state["name"]               = self.name
+        state["static_kwargs"]      = dict(self.static_kwargs)
+        state["components"]         = dict(self.components)
+        state["build_module_src"]   = self._build_module_src
+        state["build_func_name"]    = self._build_func_name
+        state["variables"]          = list(zip(self.own_vars.keys(), tfutil.run(list(self.own_vars.values()))))
+        return state
+
+    def __setstate__(self, state: dict) -> None:
+        """Pickle import."""
+        # pylint: disable=attribute-defined-outside-init
+        tfutil.assert_tf_initialized()
+        self._init_fields()
+
+        # Execute custom import handlers.
+        for handler in _import_handlers:
+            state = handler(state)
+
+        # Set basic fields.
+        assert state["version"] in [2, 3, 4]
+        self.name = state["name"]
+        self.static_kwargs = util.EasyDict(state["static_kwargs"])
+        self.components = util.EasyDict(state.get("components", {}))
+        self._build_module_src = state["build_module_src"]
+        self._build_func_name = state["build_func_name"]
+
+        # Create temporary module from the imported source code.
+        module_name = "_tflib_network_import_" + uuid.uuid4().hex
+        module = types.ModuleType(module_name)
+        sys.modules[module_name] = module
+        _import_module_src[module] = self._build_module_src
+        exec(self._build_module_src, module.__dict__) # pylint: disable=exec-used
+
+        # Locate network build function in the temporary module.
+        self._build_func = util.get_obj_from_module(module, self._build_func_name)
+        assert callable(self._build_func)
+
+        # Init TensorFlow graph.
+        self._init_graph()
+        self.reset_own_vars()
+        tfutil.set_vars({self.find_var(name): value for name, value in state["variables"]})
+
+    def clone(self, name: str = None, **new_static_kwargs) -> "Network":
+        """Create a clone of this network with its own copy of the variables."""
+        # pylint: disable=protected-access
+        net = object.__new__(Network)
+        net._init_fields()
+        net.name = name if name is not None else self.name
+        net.static_kwargs = util.EasyDict(self.static_kwargs)
+        net.static_kwargs.update(new_static_kwargs)
+        net._build_module_src = self._build_module_src
+        net._build_func_name = self._build_func_name
+        net._build_func = self._build_func
+        net._init_graph()
+        net.copy_vars_from(self)
+        return net
+
+    def copy_own_vars_from(self, src_net: "Network") -> None:
+        """Copy the values of all variables from the given network, excluding sub-networks."""
+        names = [name for name in self.own_vars.keys() if name in src_net.own_vars]
+        tfutil.set_vars(tfutil.run({self.vars[name]: src_net.vars[name] for name in names}))
+
+    def copy_vars_from(self, src_net: "Network") -> None:
+        """Copy the values of all variables from the given network, including sub-networks."""
+        names = [name for name in self.vars.keys() if name in src_net.vars]
+        tfutil.set_vars(tfutil.run({self.vars[name]: src_net.vars[name] for name in names}))
+
+    def copy_trainables_from(self, src_net: "Network") -> None:
+        """Copy the values of all trainable variables from the given network, including sub-networks."""
+        names = [name for name in self.trainables.keys() if name in src_net.trainables]
+        tfutil.set_vars(tfutil.run({self.vars[name]: src_net.vars[name] for name in names}))
+
+    def convert(self, new_func_name: str, new_name: str = None, **new_static_kwargs) -> "Network":
+        """Create new network with the given parameters, and copy all variables from this network."""
+        if new_name is None:
+            new_name = self.name
+        static_kwargs = dict(self.static_kwargs)
+        static_kwargs.update(new_static_kwargs)
+        net = Network(name=new_name, func_name=new_func_name, **static_kwargs)
+        net.copy_vars_from(self)
+        return net
+
+    def setup_as_moving_average_of(self, src_net: "Network", beta: TfExpressionEx = 0.99, beta_nontrainable: TfExpressionEx = 0.0) -> tf.Operation:
+        """Construct a TensorFlow op that updates the variables of this network
+        to be slightly closer to those of the given network."""
+        with tfutil.absolute_name_scope(self.scope + "/_MovingAvg"):
+            ops = []
+            for name, var in self.vars.items():
+                if name in src_net.vars:
+                    cur_beta = beta if name in self.trainables else beta_nontrainable
+                    new_value = tfutil.lerp(src_net.vars[name], var, cur_beta)
+                    ops.append(var.assign(new_value))
+            return tf.group(*ops)
+
+    def run(self,
+            *in_arrays: Tuple[Union[np.ndarray, None], ...],
+            input_transform: dict = None,
+            output_transform: dict = None,
+            return_as_list: bool = False,
+            print_progress: bool = False,
+            minibatch_size: int = None,
+            num_gpus: int = 1,
+            assume_frozen: bool = False,
+            **dynamic_kwargs) -> Union[np.ndarray, Tuple[np.ndarray, ...], List[np.ndarray]]:
+        """Run this network for the given NumPy array(s), and return the output(s) as NumPy array(s).
+
+        Args:
+            input_transform:    A dict specifying a custom transformation to be applied to the input tensor(s) before evaluating the network.
+                                The dict must contain a 'func' field that points to a top-level function. The function is called with the input
+                                TensorFlow expression(s) as positional arguments. Any remaining fields of the dict will be passed in as kwargs.
+            output_transform:   A dict specifying a custom transformation to be applied to the output tensor(s) after evaluating the network.
+                                The dict must contain a 'func' field that points to a top-level function. The function is called with the output
+                                TensorFlow expression(s) as positional arguments. Any remaining fields of the dict will be passed in as kwargs.
+            return_as_list:     True = return a list of NumPy arrays, False = return a single NumPy array, or a tuple if there are multiple outputs.
+            print_progress:     Print progress to the console? Useful for very large input arrays.
+            minibatch_size:     Maximum minibatch size to use, None = disable batching.
+            num_gpus:           Number of GPUs to use.
+            assume_frozen:      Improve multi-GPU performance by assuming that the trainable parameters will remain changed between calls.
+            dynamic_kwargs:     Additional keyword arguments to be passed into the network build function.
+        """
+        assert len(in_arrays) == self.num_inputs
+        assert not all(arr is None for arr in in_arrays)
+        assert input_transform is None or util.is_top_level_function(input_transform["func"])
+        assert output_transform is None or util.is_top_level_function(output_transform["func"])
+        output_transform, dynamic_kwargs = _handle_legacy_output_transforms(output_transform, dynamic_kwargs)
+        num_items = in_arrays[0].shape[0]
+        if minibatch_size is None:
+            minibatch_size = num_items
+
+        # Construct unique hash key from all arguments that affect the TensorFlow graph.
+        key = dict(input_transform=input_transform, output_transform=output_transform, num_gpus=num_gpus, assume_frozen=assume_frozen, dynamic_kwargs=dynamic_kwargs)
+        def unwind_key(obj):
+            if isinstance(obj, dict):
+                return [(key, unwind_key(value)) for key, value in sorted(obj.items())]
+            if callable(obj):
+                return util.get_top_level_function_name(obj)
+            return obj
+        key = repr(unwind_key(key))
+
+        # Build graph.
+        if key not in self._run_cache:
+            with tfutil.absolute_name_scope(self.scope + "/_Run"), tf.control_dependencies(None):
+                with tf.device("/cpu:0"):
+                    in_expr = [tf.placeholder(tf.float32, name=name) for name in self.input_names]
+                    in_split = list(zip(*[tf.split(x, num_gpus) for x in in_expr]))
+
+                out_split = []
+                for gpu in range(num_gpus):
+                    with tf.device("/gpu:%d" % gpu):
+                        net_gpu = self.clone() if assume_frozen else self
+                        in_gpu = in_split[gpu]
+
+                        if input_transform is not None:
+                            in_kwargs = dict(input_transform)
+                            in_gpu = in_kwargs.pop("func")(*in_gpu, **in_kwargs)
+                            in_gpu = [in_gpu] if tfutil.is_tf_expression(in_gpu) else list(in_gpu)
+
+                        assert len(in_gpu) == self.num_inputs
+                        out_gpu = net_gpu.get_output_for(*in_gpu, return_as_list=True, **dynamic_kwargs)
+
+                        if output_transform is not None:
+                            out_kwargs = dict(output_transform)
+                            out_gpu = out_kwargs.pop("func")(*out_gpu, **out_kwargs)
+                            out_gpu = [out_gpu] if tfutil.is_tf_expression(out_gpu) else list(out_gpu)
+
+                        assert len(out_gpu) == self.num_outputs
+                        out_split.append(out_gpu)
+
+                with tf.device("/cpu:0"):
+                    out_expr = [tf.concat(outputs, axis=0) for outputs in zip(*out_split)]
+                    self._run_cache[key] = in_expr, out_expr
+
+        # Run minibatches.
+        in_expr, out_expr = self._run_cache[key]
+        out_arrays = [np.empty([num_items] + expr.shape.as_list()[1:], expr.dtype.name) for expr in out_expr]
+
+        for mb_begin in range(0, num_items, minibatch_size):
+            if print_progress:
+                print("\r%d / %d" % (mb_begin, num_items), end="")
+
+            mb_end = min(mb_begin + minibatch_size, num_items)
+            mb_num = mb_end - mb_begin
+            mb_in = [src[mb_begin : mb_end] if src is not None else np.zeros([mb_num] + shape[1:]) for src, shape in zip(in_arrays, self.input_shapes)]
+            mb_out = tf.get_default_session().run(out_expr, dict(zip(in_expr, mb_in)))
+
+            for dst, src in zip(out_arrays, mb_out):
+                dst[mb_begin: mb_end] = src
+
+        # Done.
+        if print_progress:
+            print("\r%d / %d" % (num_items, num_items))
+
+        if not return_as_list:
+            out_arrays = out_arrays[0] if len(out_arrays) == 1 else tuple(out_arrays)
+        return out_arrays
+
+    def list_ops(self) -> List[TfExpression]:
+        include_prefix = self.scope + "/"
+        exclude_prefix = include_prefix + "_"
+        ops = tf.get_default_graph().get_operations()
+        ops = [op for op in ops if op.name.startswith(include_prefix)]
+        ops = [op for op in ops if not op.name.startswith(exclude_prefix)]
+        return ops
+
+    def list_layers(self) -> List[Tuple[str, TfExpression, List[TfExpression]]]:
+        """Returns a list of (layer_name, output_expr, trainable_vars) tuples corresponding to
+        individual layers of the network. Mainly intended to be used for reporting."""
+        layers = []
+
+        def recurse(scope, parent_ops, parent_vars, level):
+            # Ignore specific patterns.
+            if any(p in scope for p in ["/Shape", "/strided_slice", "/Cast", "/concat", "/Assign"]):
+                return
+
+            # Filter ops and vars by scope.
+            global_prefix = scope + "/"
+            local_prefix = global_prefix[len(self.scope) + 1:]
+            cur_ops = [op for op in parent_ops if op.name.startswith(global_prefix) or op.name == global_prefix[:-1]]
+            cur_vars = [(name, var) for name, var in parent_vars if name.startswith(local_prefix) or name == local_prefix[:-1]]
+            if not cur_ops and not cur_vars:
+                return
+
+            # Filter out all ops related to variables.
+            for var in [op for op in cur_ops if op.type.startswith("Variable")]:
+                var_prefix = var.name + "/"
+                cur_ops = [op for op in cur_ops if not op.name.startswith(var_prefix)]
+
+            # Scope does not contain ops as immediate children => recurse deeper.
+            contains_direct_ops = any("/" not in op.name[len(global_prefix):] and op.type not in ["Identity", "Cast", "Transpose"] for op in cur_ops)
+            if (level == 0 or not contains_direct_ops) and (len(cur_ops) + len(cur_vars)) > 1:
+                visited = set()
+                for rel_name in [op.name[len(global_prefix):] for op in cur_ops] + [name[len(local_prefix):] for name, _var in cur_vars]:
+                    token = rel_name.split("/")[0]
+                    if token not in visited:
+                        recurse(global_prefix + token, cur_ops, cur_vars, level + 1)
+                        visited.add(token)
+                return
+
+            # Report layer.
+            layer_name = scope[len(self.scope) + 1:]
+            layer_output = cur_ops[-1].outputs[0] if cur_ops else cur_vars[-1][1]
+            layer_trainables = [var for _name, var in cur_vars if var.trainable]
+            layers.append((layer_name, layer_output, layer_trainables))
+
+        recurse(self.scope, self.list_ops(), list(self.vars.items()), 0)
+        return layers
+
+    def print_layers(self, title: str = None, hide_layers_with_no_params: bool = False) -> None:
+        """Print a summary table of the network structure."""
+        rows = [[title if title is not None else self.name, "Params", "OutputShape", "WeightShape"]]
+        rows += [["---"] * 4]
+        total_params = 0
+
+        for layer_name, layer_output, layer_trainables in self.list_layers():
+            num_params = sum(int(np.prod(var.shape.as_list())) for var in layer_trainables)
+            weights = [var for var in layer_trainables if var.name.endswith("/weight:0")]
+            weights.sort(key=lambda x: len(x.name))
+            if len(weights) == 0 and len(layer_trainables) == 1:
+                weights = layer_trainables
+            total_params += num_params
+
+            if not hide_layers_with_no_params or num_params != 0:
+                num_params_str = str(num_params) if num_params > 0 else "-"
+                output_shape_str = str(layer_output.shape)
+                weight_shape_str = str(weights[0].shape) if len(weights) >= 1 else "-"
+                rows += [[layer_name, num_params_str, output_shape_str, weight_shape_str]]
+
+        rows += [["---"] * 4]
+        rows += [["Total", str(total_params), "", ""]]
+
+        widths = [max(len(cell) for cell in column) for column in zip(*rows)]
+        print()
+        for row in rows:
+            print("  ".join(cell + " " * (width - len(cell)) for cell, width in zip(row, widths)))
+        print()
+
+    def setup_weight_histograms(self, title: str = None) -> None:
+        """Construct summary ops to include histograms of all trainable parameters in TensorBoard."""
+        if title is None:
+            title = self.name
+
+        with tf.name_scope(None), tf.device(None), tf.control_dependencies(None):
+            for local_name, var in self.trainables.items():
+                if "/" in local_name:
+                    p = local_name.split("/")
+                    name = title + "_" + p[-1] + "/" + "_".join(p[:-1])
+                else:
+                    name = title + "_toplevel/" + local_name
+
+                tf.summary.histogram(name, var)
+
+#----------------------------------------------------------------------------
+# Backwards-compatible emulation of legacy output transformation in Network.run().
+
+_print_legacy_warning = True
+
+def _handle_legacy_output_transforms(output_transform, dynamic_kwargs):
+    global _print_legacy_warning
+    legacy_kwargs = ["out_mul", "out_add", "out_shrink", "out_dtype"]
+    if not any(kwarg in dynamic_kwargs for kwarg in legacy_kwargs):
+        return output_transform, dynamic_kwargs
+
+    if _print_legacy_warning:
+        _print_legacy_warning = False
+        print()
+        print("WARNING: Old-style output transformations in Network.run() are deprecated.")
+        print("Consider using 'output_transform=dict(func=tflib.convert_images_to_uint8)'")
+        print("instead of 'out_mul=127.5, out_add=127.5, out_dtype=np.uint8'.")
+        print()
+    assert output_transform is None
+
+    new_kwargs = dict(dynamic_kwargs)
+    new_transform = {kwarg: new_kwargs.pop(kwarg) for kwarg in legacy_kwargs if kwarg in dynamic_kwargs}
+    new_transform["func"] = _legacy_output_transform_func
+    return new_transform, new_kwargs
+
+def _legacy_output_transform_func(*expr, out_mul=1.0, out_add=0.0, out_shrink=1, out_dtype=None):
+    if out_mul != 1.0:
+        expr = [x * out_mul for x in expr]
+
+    if out_add != 0.0:
+        expr = [x + out_add for x in expr]
+
+    if out_shrink > 1:
+        ksize = [1, 1, out_shrink, out_shrink]
+        expr = [tf.nn.avg_pool(x, ksize=ksize, strides=ksize, padding="VALID", data_format="NCHW") for x in expr]
+
+    if out_dtype is not None:
+        if tf.as_dtype(out_dtype).is_integer:
+            expr = [tf.round(x) for x in expr]
+        expr = [tf.saturate_cast(x, out_dtype) for x in expr]
+    return expr

stylegan_human/dnnlib/tflib/ops/__init__.py

@@ -0,0 +1,9 @@
+# Copyright (c) SenseTime Research. All rights reserved.
+
+# Copyright (c) 2019, NVIDIA Corporation. All rights reserved.
+#
+# This work is made available under the Nvidia Source Code License-NC.
+# To view a copy of this license, visit
+# https://nvlabs.github.io/stylegan2/license.html
+
+# empty

stylegan_human/dnnlib/tflib/ops/fused_bias_act.cu

@@ -0,0 +1,190 @@
+// Copyright (c) SenseTime Research. All rights reserved.
+
+// Copyright (c) 2019, NVIDIA Corporation. All rights reserved.
+//
+// This work is made available under the Nvidia Source Code License-NC.
+// To view a copy of this license, visit
+// https://nvlabs.github.io/stylegan2/license.html
+
+#define EIGEN_USE_GPU
+#define __CUDA_INCLUDE_COMPILER_INTERNAL_HEADERS__
+#include "tensorflow/core/framework/op.h"
+#include "tensorflow/core/framework/op_kernel.h"
+#include "tensorflow/core/framework/shape_inference.h"
+#include <stdio.h>
+
+using namespace tensorflow;
+using namespace tensorflow::shape_inference;
+
+#define OP_CHECK_CUDA_ERROR(CTX, CUDA_CALL) do { cudaError_t err = CUDA_CALL; OP_REQUIRES(CTX, err == cudaSuccess, errors::Internal(cudaGetErrorName(err))); } while (false)
+
+//------------------------------------------------------------------------
+// CUDA kernel.
+
+template <class T>
+struct FusedBiasActKernelParams
+{
+    const T*    x;      // [sizeX]
+    const T*    b;      // [sizeB] or NULL
+    const T*    ref;    // [sizeX] or NULL
+    T*          y;      // [sizeX]
+
+    int         grad;
+    int         axis;
+    int         act;
+    float       alpha;
+    float       gain;
+
+    int         sizeX;
+    int         sizeB;
+    int         stepB;
+    int         loopX;
+};
+
+template <class T>
+static __global__ void FusedBiasActKernel(const FusedBiasActKernelParams<T> p)
+{
+    const float expRange        = 80.0f;
+    const float halfExpRange    = 40.0f;
+    const float seluScale       = 1.0507009873554804934193349852946f;
+    const float seluAlpha       = 1.6732632423543772848170429916717f;
+
+    // Loop over elements.
+    int xi = blockIdx.x * p.loopX * blockDim.x + threadIdx.x;
+    for (int loopIdx = 0; loopIdx < p.loopX && xi < p.sizeX; loopIdx++, xi += blockDim.x)
+    {
+        // Load and apply bias.
+        float x = (float)p.x[xi];
+        if (p.b)
+            x += (float)p.b[(xi / p.stepB) % p.sizeB];
+        float ref = (p.ref) ? (float)p.ref[xi] : 0.0f;
+        if (p.gain != 0.0f & p.act != 9)
+            ref /= p.gain;
+
+        // Evaluate activation func.
+        float y;
+        switch (p.act * 10 + p.grad)
+        {
+            // linear
+            default:
+            case 10: y = x; break;
+            case 11: y = x; break;
+            case 12: y = 0.0f; break;
+
+            // relu
+            case 20: y = (x > 0.0f) ? x : 0.0f; break;
+            case 21: y = (ref > 0.0f) ? x : 0.0f; break;
+            case 22: y = 0.0f; break;
+
+            // lrelu
+            case 30: y = (x > 0.0f) ? x : x * p.alpha; break;
+            case 31: y = (ref > 0.0f) ? x : x * p.alpha; break;
+            case 32: y = 0.0f; break;
+
+            // tanh
+            case 40: { float c = expf(x); float d = 1.0f / c; y = (x < -expRange) ? -1.0f : (x > expRange) ? 1.0f : (c - d) / (c + d); } break;
+            case 41: y = x * (1.0f - ref * ref); break;
+            case 42: y = x * (1.0f - ref * ref) * (-2.0f * ref); break;
+
+            // sigmoid
+            case 50: y = (x < -expRange) ? 0.0f : 1.0f / (expf(-x) + 1.0f); break;
+            case 51: y = x * ref * (1.0f - ref); break;
+            case 52: y = x * ref * (1.0f - ref) * (1.0f - 2.0f * ref); break;
+
+            // elu
+            case 60: y = (x >= 0.0f) ? x : expf(x) - 1.0f; break;
+            case 61: y = (ref >= 0.0f) ? x : x * (ref + 1.0f); break;
+            case 62: y = (ref >= 0.0f) ? 0.0f : x * (ref + 1.0f); break;
+
+            // selu
+            case 70: y = (x >= 0.0f) ? seluScale * x : (seluScale * seluAlpha) * (expf(x) - 1.0f); break;
+            case 71: y = (ref >= 0.0f) ? x * seluScale : x * (ref + seluScale * seluAlpha); break;
+            case 72: y = (ref >= 0.0f) ? 0.0f : x * (ref + seluScale * seluAlpha); break;
+
+            // softplus
+            case 80: y = (x > expRange) ? x : logf(expf(x) + 1.0f); break;
+            case 81: y = x * (1.0f - expf(-ref)); break;
+            case 82: { float c = expf(-ref); y = x * c * (1.0f - c); } break;
+
+            // swish
+            case 90: y = (x < -expRange) ? 0.0f : x / (expf(-x) + 1.0f); break;
+            case 91: { float c = expf(ref); float d = c + 1.0f; y = (ref > halfExpRange) ? x : x * c * (ref + d) / (d * d); } break;
+            case 92: { float c = expf(ref); float d = c + 1.0f; y = (ref > halfExpRange) ? 0.0f : x * c * (ref * (2.0f - d) + 2.0f * d) / (d * d * d); } break;
+        }
+
+        // Apply gain and store.
+        p.y[xi] = (T)(y * p.gain);
+    }
+}
+
+//------------------------------------------------------------------------
+// TensorFlow op.
+
+template <class T>
+struct FusedBiasActOp : public OpKernel
+{
+    FusedBiasActKernelParams<T> m_attribs;
+
+    FusedBiasActOp(OpKernelConstruction* ctx) : OpKernel(ctx)
+    {
+        memset(&m_attribs, 0, sizeof(m_attribs));
+        OP_REQUIRES_OK(ctx, ctx->GetAttr("grad", &m_attribs.grad));
+        OP_REQUIRES_OK(ctx, ctx->GetAttr("axis", &m_attribs.axis));
+        OP_REQUIRES_OK(ctx, ctx->GetAttr("act", &m_attribs.act));
+        OP_REQUIRES_OK(ctx, ctx->GetAttr("alpha", &m_attribs.alpha));
+        OP_REQUIRES_OK(ctx, ctx->GetAttr("gain", &m_attribs.gain));
+        OP_REQUIRES(ctx, m_attribs.grad >= 0, errors::InvalidArgument("grad must be non-negative"));
+        OP_REQUIRES(ctx, m_attribs.axis >= 0, errors::InvalidArgument("axis must be non-negative"));
+        OP_REQUIRES(ctx, m_attribs.act >= 0, errors::InvalidArgument("act must be non-negative"));
+    }
+
+    void Compute(OpKernelContext* ctx)
+    {
+        FusedBiasActKernelParams<T> p = m_attribs;
+        cudaStream_t stream = ctx->eigen_device<Eigen::GpuDevice>().stream();
+
+        const Tensor& x     = ctx->input(0); // [...]
+        const Tensor& b     = ctx->input(1); // [sizeB] or [0]
+        const Tensor& ref   = ctx->input(2); // x.shape or [0]
+        p.x = x.flat<T>().data();
+        p.b = (b.NumElements()) ? b.flat<T>().data() : NULL;
+        p.ref = (ref.NumElements()) ? ref.flat<T>().data() : NULL;
+        OP_REQUIRES(ctx, b.NumElements() == 0 || m_attribs.axis < x.dims(), errors::InvalidArgument("axis out of bounds"));
+        OP_REQUIRES(ctx, b.dims() == 1, errors::InvalidArgument("b must have rank 1"));
+        OP_REQUIRES(ctx, b.NumElements() == 0 || b.NumElements() == x.dim_size(m_attribs.axis), errors::InvalidArgument("b has wrong number of elements"));
+        OP_REQUIRES(ctx, ref.NumElements() == ((p.grad == 0) ? 0 : x.NumElements()), errors::InvalidArgument("ref has wrong number of elements"));
+        OP_REQUIRES(ctx, x.NumElements() <= kint32max, errors::InvalidArgument("x is too large"));
+
+        p.sizeX = (int)x.NumElements();
+        p.sizeB = (int)b.NumElements();
+        p.stepB = 1;
+        for (int i = m_attribs.axis + 1; i < x.dims(); i++)
+            p.stepB *= (int)x.dim_size(i);
+
+        Tensor* y = NULL; // x.shape
+        OP_REQUIRES_OK(ctx, ctx->allocate_output(0, x.shape(), &y));
+        p.y = y->flat<T>().data();
+
+        p.loopX = 4;
+        int blockSize = 4 * 32;
+        int gridSize = (p.sizeX - 1) / (p.loopX * blockSize) + 1;
+        void* args[] = {&p};
+        OP_CHECK_CUDA_ERROR(ctx, cudaLaunchKernel((void*)FusedBiasActKernel<T>, gridSize, blockSize, args, 0, stream));
+    }
+};
+
+REGISTER_OP("FusedBiasAct")
+    .Input      ("x: T")
+    .Input      ("b: T")
+    .Input      ("ref: T")
+    .Output     ("y: T")
+    .Attr       ("T: {float, half}")
+    .Attr       ("grad: int = 0")
+    .Attr       ("axis: int = 1")
+    .Attr       ("act: int = 0")
+    .Attr       ("alpha: float = 0.0")
+    .Attr       ("gain: float = 1.0");
+REGISTER_KERNEL_BUILDER(Name("FusedBiasAct").Device(DEVICE_GPU).TypeConstraint<float>("T"), FusedBiasActOp<float>);
+REGISTER_KERNEL_BUILDER(Name("FusedBiasAct").Device(DEVICE_GPU).TypeConstraint<Eigen::half>("T"), FusedBiasActOp<Eigen::half>);
+
+//------------------------------------------------------------------------

stylegan_human/dnnlib/tflib/ops/fused_bias_act.py

@@ -0,0 +1,198 @@
+# Copyright (c) SenseTime Research. All rights reserved.
+
+# Copyright (c) 2019, NVIDIA Corporation. All rights reserved.
+#
+# This work is made available under the Nvidia Source Code License-NC.
+# To view a copy of this license, visit
+# https://nvlabs.github.io/stylegan2/license.html
+
+"""Custom TensorFlow ops for efficient bias and activation."""
+
+import os
+import numpy as np
+import tensorflow as tf
+from .. import custom_ops
+from ...util import EasyDict
+
+def _get_plugin():
+    return custom_ops.get_plugin(os.path.splitext(__file__)[0] + '.cu')
+
+#----------------------------------------------------------------------------
+
+activation_funcs = {
+    'linear':   EasyDict(func=lambda x, **_:        x,                          def_alpha=None, def_gain=1.0,           cuda_idx=1, ref='y', zero_2nd_grad=True),
+    'relu':     EasyDict(func=lambda x, **_:        tf.nn.relu(x),              def_alpha=None, def_gain=np.sqrt(2),    cuda_idx=2, ref='y', zero_2nd_grad=True),
+    'lrelu':    EasyDict(func=lambda x, alpha, **_: tf.nn.leaky_relu(x, alpha), def_alpha=0.2,  def_gain=np.sqrt(2),    cuda_idx=3, ref='y', zero_2nd_grad=True),
+    'tanh':     EasyDict(func=lambda x, **_:        tf.nn.tanh(x),              def_alpha=None, def_gain=1.0,           cuda_idx=4, ref='y', zero_2nd_grad=False),
+    'sigmoid':  EasyDict(func=lambda x, **_:        tf.nn.sigmoid(x),           def_alpha=None, def_gain=1.0,           cuda_idx=5, ref='y', zero_2nd_grad=False),
+    'elu':      EasyDict(func=lambda x, **_:        tf.nn.elu(x),               def_alpha=None, def_gain=1.0,           cuda_idx=6, ref='y', zero_2nd_grad=False),
+    'selu':     EasyDict(func=lambda x, **_:        tf.nn.selu(x),              def_alpha=None, def_gain=1.0,           cuda_idx=7, ref='y', zero_2nd_grad=False),
+    'softplus': EasyDict(func=lambda x, **_:        tf.nn.softplus(x),          def_alpha=None, def_gain=1.0,           cuda_idx=8, ref='y', zero_2nd_grad=False),
+    'swish':    EasyDict(func=lambda x, **_:        tf.nn.sigmoid(x) * x,       def_alpha=None, def_gain=np.sqrt(2),    cuda_idx=9, ref='x', zero_2nd_grad=False),
+}
+
+#----------------------------------------------------------------------------
+
+def fused_bias_act(x, b=None, axis=1, act='linear', alpha=None, gain=None, impl='cuda'):
+    r"""Fused bias and activation function.
+
+    Adds bias `b` to activation tensor `x`, evaluates activation function `act`,
+    and scales the result by `gain`. Each of the steps is optional. In most cases,
+    the fused op is considerably more efficient than performing the same calculation
+    using standard TensorFlow ops. It supports first and second order gradients,
+    but not third order gradients.
+
+    Args:
+        x:      Input activation tensor. Can have any shape, but if `b` is defined, the
+                dimension corresponding to `axis`, as well as the rank, must be known.
+        b:      Bias vector, or `None` to disable. Must be a 1D tensor of the same type
+                as `x`. The shape must be known, and it must match the dimension of `x`
+                corresponding to `axis`.
+        axis:   The dimension in `x` corresponding to the elements of `b`.
+                The value of `axis` is ignored if `b` is not specified.
+        act:    Name of the activation function to evaluate, or `"linear"` to disable.
+                Can be e.g. `"relu"`, `"lrelu"`, `"tanh"`, `"sigmoid"`, `"swish"`, etc.
+                See `activation_funcs` for a full list. `None` is not allowed.
+        alpha:  Shape parameter for the activation function, or `None` to use the default.
+        gain:   Scaling factor for the output tensor, or `None` to use default.
+                See `activation_funcs` for the default scaling of each activation function.
+                If unsure, consider specifying `1.0`.
+        impl:   Name of the implementation to use. Can be `"ref"` or `"cuda"` (default).
+
+    Returns:
+        Tensor of the same shape and datatype as `x`.
+    """
+
+    impl_dict = {
+        'ref':  _fused_bias_act_ref,
+        'cuda': _fused_bias_act_cuda,
+    }
+    return impl_dict[impl](x=x, b=b, axis=axis, act=act, alpha=alpha, gain=gain)
+
+#----------------------------------------------------------------------------
+
+def _fused_bias_act_ref(x, b, axis, act, alpha, gain):
+    """Slow reference implementation of `fused_bias_act()` using standard TensorFlow ops."""
+
+    # Validate arguments.
+    x = tf.convert_to_tensor(x)
+    b = tf.convert_to_tensor(b) if b is not None else tf.constant([], dtype=x.dtype)
+    act_spec = activation_funcs[act]
+    assert b.shape.rank == 1 and (b.shape[0] == 0 or b.shape[0] == x.shape[axis])
+    assert b.shape[0] == 0 or 0 <= axis < x.shape.rank
+    if alpha is None:
+        alpha = act_spec.def_alpha
+    if gain is None:
+        gain = act_spec.def_gain
+
+    # Add bias.
+    if b.shape[0] != 0:
+        x += tf.reshape(b, [-1 if i == axis else 1 for i in range(x.shape.rank)])
+
+    # Evaluate activation function.
+    x = act_spec.func(x, alpha=alpha)
+
+    # Scale by gain.
+    if gain != 1:
+        x *= gain
+    return x
+
+#----------------------------------------------------------------------------
+
+def _fused_bias_act_cuda(x, b, axis, act, alpha, gain):
+    """Fast CUDA implementation of `fused_bias_act()` using custom ops."""
+
+    # Validate arguments.
+    x = tf.convert_to_tensor(x)
+    empty_tensor = tf.constant([], dtype=x.dtype)
+    b = tf.convert_to_tensor(b) if b is not None else empty_tensor
+    act_spec = activation_funcs[act]
+    assert b.shape.rank == 1 and (b.shape[0] == 0 or b.shape[0] == x.shape[axis])
+    assert b.shape[0] == 0 or 0 <= axis < x.shape.rank
+    if alpha is None:
+        alpha = act_spec.def_alpha
+    if gain is None:
+        gain = act_spec.def_gain
+
+    # Special cases.
+    if act == 'linear' and b is None and gain == 1.0:
+        return x
+    if act_spec.cuda_idx is None:
+        return _fused_bias_act_ref(x=x, b=b, axis=axis, act=act, alpha=alpha, gain=gain)
+
+    # CUDA kernel.
+    cuda_kernel = _get_plugin().fused_bias_act
+    cuda_kwargs = dict(axis=axis, act=act_spec.cuda_idx, alpha=alpha, gain=gain)
+
+    # Forward pass: y = func(x, b).
+    def func_y(x, b):
+        y = cuda_kernel(x=x, b=b, ref=empty_tensor, grad=0, **cuda_kwargs)
+        y.set_shape(x.shape)
+        return y
+
+    # Backward pass: dx, db = grad(dy, x, y)
+    def grad_dx(dy, x, y):
+        ref = {'x': x, 'y': y}[act_spec.ref]
+        dx = cuda_kernel(x=dy, b=empty_tensor, ref=ref, grad=1, **cuda_kwargs)
+        dx.set_shape(x.shape)
+        return dx
+    def grad_db(dx):
+        if b.shape[0] == 0:
+            return empty_tensor
+        db = dx
+        if axis < x.shape.rank - 1:
+            db = tf.reduce_sum(db, list(range(axis + 1, x.shape.rank)))
+        if axis > 0:
+            db = tf.reduce_sum(db, list(range(axis)))
+        db.set_shape(b.shape)
+        return db
+
+    # Second order gradients: d_dy, d_x = grad2(d_dx, d_db, x, y)
+    def grad2_d_dy(d_dx, d_db, x, y):
+        ref = {'x': x, 'y': y}[act_spec.ref]
+        d_dy = cuda_kernel(x=d_dx, b=d_db, ref=ref, grad=1, **cuda_kwargs)
+        d_dy.set_shape(x.shape)
+        return d_dy
+    def grad2_d_x(d_dx, d_db, x, y):
+        ref = {'x': x, 'y': y}[act_spec.ref]
+        d_x = cuda_kernel(x=d_dx, b=d_db, ref=ref, grad=2, **cuda_kwargs)
+        d_x.set_shape(x.shape)
+        return d_x
+
+    # Fast version for piecewise-linear activation funcs.
+    @tf.custom_gradient
+    def func_zero_2nd_grad(x, b):
+        y = func_y(x, b)
+        @tf.custom_gradient
+        def grad(dy):
+            dx = grad_dx(dy, x, y)
+            db = grad_db(dx)
+            def grad2(d_dx, d_db):
+                d_dy = grad2_d_dy(d_dx, d_db, x, y)
+                return d_dy
+            return (dx, db), grad2
+        return y, grad
+
+    # Slow version for general activation funcs.
+    @tf.custom_gradient
+    def func_nonzero_2nd_grad(x, b):
+        y = func_y(x, b)
+        def grad_wrap(dy):
+            @tf.custom_gradient
+            def grad_impl(dy, x):
+                dx = grad_dx(dy, x, y)
+                db = grad_db(dx)
+                def grad2(d_dx, d_db):
+                    d_dy = grad2_d_dy(d_dx, d_db, x, y)
+                    d_x = grad2_d_x(d_dx, d_db, x, y)
+                    return d_dy, d_x
+                return (dx, db), grad2
+            return grad_impl(dy, x)
+        return y, grad_wrap
+
+    # Which version to use?
+    if act_spec.zero_2nd_grad:
+        return func_zero_2nd_grad(x, b)
+    return func_nonzero_2nd_grad(x, b)
+
+#----------------------------------------------------------------------------

stylegan_human/dnnlib/tflib/ops/upfirdn_2d.cu

@@ -0,0 +1,328 @@
+// Copyright (c) SenseTime Research. All rights reserved.
+
+// Copyright (c) 2019, NVIDIA Corporation. All rights reserved.
+//
+// This work is made available under the Nvidia Source Code License-NC.
+// To view a copy of this license, visit
+// https://nvlabs.github.io/stylegan2/license.html
+
+#define EIGEN_USE_GPU
+#define __CUDA_INCLUDE_COMPILER_INTERNAL_HEADERS__
+#include "tensorflow/core/framework/op.h"
+#include "tensorflow/core/framework/op_kernel.h"
+#include "tensorflow/core/framework/shape_inference.h"
+#include <stdio.h>
+
+using namespace tensorflow;
+using namespace tensorflow::shape_inference;
+
+//------------------------------------------------------------------------
+// Helpers.
+
+#define OP_CHECK_CUDA_ERROR(CTX, CUDA_CALL) do { cudaError_t err = CUDA_CALL; OP_REQUIRES(CTX, err == cudaSuccess, errors::Internal(cudaGetErrorName(err))); } while (false)
+
+static __host__ __device__ __forceinline__ int floorDiv(int a, int b)
+{
+    int c = a / b;
+    if (c * b > a)
+        c--;
+    return c;
+}
+
+//------------------------------------------------------------------------
+// CUDA kernel params.
+
+template <class T>
+struct UpFirDn2DKernelParams
+{
+    const T*    x;          // [majorDim, inH, inW, minorDim]
+    const T*    k;          // [kernelH, kernelW]
+    T*          y;          // [majorDim, outH, outW, minorDim]
+
+    int         upx;
+    int         upy;
+    int         downx;
+    int         downy;
+    int         padx0;
+    int         padx1;
+    int         pady0;
+    int         pady1;
+
+    int         majorDim;
+    int         inH;
+    int         inW;
+    int         minorDim;
+    int         kernelH;
+    int         kernelW;
+    int         outH;
+    int         outW;
+    int         loopMajor;
+    int         loopX;
+};
+
+//------------------------------------------------------------------------
+// General CUDA implementation for large filter kernels.
+
+template <class T>
+static __global__ void UpFirDn2DKernel_large(const UpFirDn2DKernelParams<T> p)
+{
+    // Calculate thread index.
+    int minorIdx = blockIdx.x * blockDim.x + threadIdx.x;
+    int outY = minorIdx / p.minorDim;
+    minorIdx -= outY * p.minorDim;
+    int outXBase = blockIdx.y * p.loopX * blockDim.y + threadIdx.y;
+    int majorIdxBase = blockIdx.z * p.loopMajor;
+    if (outXBase >= p.outW || outY >= p.outH || majorIdxBase >= p.majorDim)
+        return;
+
+    // Setup Y receptive field.
+    int midY = outY * p.downy + p.upy - 1 - p.pady0;
+    int inY = min(max(floorDiv(midY, p.upy), 0), p.inH);
+    int h = min(max(floorDiv(midY + p.kernelH, p.upy), 0), p.inH) - inY;
+    int kernelY = midY + p.kernelH - (inY + 1) * p.upy;
+
+    // Loop over majorDim and outX.
+    for (int loopMajor = 0, majorIdx = majorIdxBase; loopMajor < p.loopMajor && majorIdx < p.majorDim; loopMajor++, majorIdx++)
+    for (int loopX = 0, outX = outXBase; loopX < p.loopX && outX < p.outW; loopX++, outX += blockDim.y)
+    {
+        // Setup X receptive field.
+        int midX = outX * p.downx + p.upx - 1 - p.padx0;
+        int inX = min(max(floorDiv(midX, p.upx), 0), p.inW);
+        int w = min(max(floorDiv(midX + p.kernelW, p.upx), 0), p.inW) - inX;
+        int kernelX = midX + p.kernelW - (inX + 1) * p.upx;
+
+        // Initialize pointers.
+        const T* xp = &p.x[((majorIdx * p.inH + inY) * p.inW + inX) * p.minorDim + minorIdx];
+        const T* kp = &p.k[kernelY * p.kernelW + kernelX];
+        int xpx = p.minorDim;
+        int kpx = -p.upx;
+        int xpy = p.inW * p.minorDim;
+        int kpy = -p.upy * p.kernelW;
+
+        // Inner loop.
+        float v = 0.0f;
+        for (int y = 0; y < h; y++)
+        {
+            for (int x = 0; x < w; x++)
+            {
+                v += (float)(*xp) * (float)(*kp);
+                xp += xpx;
+                kp += kpx;
+            }
+            xp += xpy - w * xpx;
+            kp += kpy - w * kpx;
+        }
+
+        // Store result.
+        p.y[((majorIdx * p.outH + outY) * p.outW + outX) * p.minorDim + minorIdx] = (T)v;
+    }
+}
+
+//------------------------------------------------------------------------
+// Specialized CUDA implementation for small filter kernels.
+
+template <class T, int upx, int upy, int downx, int downy, int kernelW, int kernelH, int tileOutW, int tileOutH>
+static __global__ void UpFirDn2DKernel_small(const UpFirDn2DKernelParams<T> p)
+{
+    //assert(kernelW % upx == 0);
+    //assert(kernelH % upy == 0);
+    const int tileInW = ((tileOutW - 1) * downx + kernelW - 1) / upx + 1;
+    const int tileInH = ((tileOutH - 1) * downy + kernelH - 1) / upy + 1;
+    __shared__ volatile float sk[kernelH][kernelW];
+    __shared__ volatile float sx[tileInH][tileInW];
+
+    // Calculate tile index.
+    int minorIdx = blockIdx.x;
+    int tileOutY = minorIdx / p.minorDim;
+    minorIdx -= tileOutY * p.minorDim;
+    tileOutY *= tileOutH;
+    int tileOutXBase = blockIdx.y * p.loopX * tileOutW;
+    int majorIdxBase = blockIdx.z * p.loopMajor;
+    if (tileOutXBase >= p.outW | tileOutY >= p.outH | majorIdxBase >= p.majorDim)
+        return;
+
+    // Load filter kernel (flipped).
+    for (int tapIdx = threadIdx.x; tapIdx < kernelH * kernelW; tapIdx += blockDim.x)
+    {
+        int ky = tapIdx / kernelW;
+        int kx = tapIdx - ky * kernelW;
+        float v = 0.0f;
+        if (kx < p.kernelW & ky < p.kernelH)
+            v = (float)p.k[(p.kernelH - 1 - ky) * p.kernelW + (p.kernelW - 1 - kx)];
+        sk[ky][kx] = v;
+    }
+
+    // Loop over majorDim and outX.
+    for (int loopMajor = 0, majorIdx = majorIdxBase; loopMajor < p.loopMajor & majorIdx < p.majorDim; loopMajor++, majorIdx++)
+    for (int loopX = 0, tileOutX = tileOutXBase; loopX < p.loopX & tileOutX < p.outW; loopX++, tileOutX += tileOutW)
+    {
+        // Load input pixels.
+        int tileMidX = tileOutX * downx + upx - 1 - p.padx0;
+        int tileMidY = tileOutY * downy + upy - 1 - p.pady0;
+        int tileInX = floorDiv(tileMidX, upx);
+        int tileInY = floorDiv(tileMidY, upy);
+        __syncthreads();
+        for (int inIdx = threadIdx.x; inIdx < tileInH * tileInW; inIdx += blockDim.x)
+        {
+            int relInY = inIdx / tileInW;
+            int relInX = inIdx - relInY * tileInW;
+            int inX = relInX + tileInX;
+            int inY = relInY + tileInY;
+            float v = 0.0f;
+            if (inX >= 0 & inY >= 0 & inX < p.inW & inY < p.inH)
+                v = (float)p.x[((majorIdx * p.inH + inY) * p.inW + inX) * p.minorDim + minorIdx];
+            sx[relInY][relInX] = v;
+        }
+
+        // Loop over output pixels.
+        __syncthreads();
+        for (int outIdx = threadIdx.x; outIdx < tileOutH * tileOutW; outIdx += blockDim.x)
+        {
+            int relOutY = outIdx / tileOutW;
+            int relOutX = outIdx - relOutY * tileOutW;
+            int outX = relOutX + tileOutX;
+            int outY = relOutY + tileOutY;
+
+            // Setup receptive field.
+            int midX = tileMidX + relOutX * downx;
+            int midY = tileMidY + relOutY * downy;
+            int inX = floorDiv(midX, upx);
+            int inY = floorDiv(midY, upy);
+            int relInX = inX - tileInX;
+            int relInY = inY - tileInY;
+            int kernelX = (inX + 1) * upx - midX - 1; // flipped
+            int kernelY = (inY + 1) * upy - midY - 1; // flipped
+
+            // Inner loop.
+            float v = 0.0f;
+            #pragma unroll
+            for (int y = 0; y < kernelH / upy; y++)
+                #pragma unroll
+                for (int x = 0; x < kernelW / upx; x++)
+                    v += sx[relInY + y][relInX + x] * sk[kernelY + y * upy][kernelX + x * upx];
+
+            // Store result.
+            if (outX < p.outW & outY < p.outH)
+                p.y[((majorIdx * p.outH + outY) * p.outW + outX) * p.minorDim + minorIdx] = (T)v;
+        }
+    }
+}
+
+//------------------------------------------------------------------------
+// TensorFlow op.
+
+template <class T>
+struct UpFirDn2DOp : public OpKernel
+{
+    UpFirDn2DKernelParams<T> m_attribs;
+
+    UpFirDn2DOp(OpKernelConstruction* ctx) : OpKernel(ctx)
+    {
+        memset(&m_attribs, 0, sizeof(m_attribs));
+        OP_REQUIRES_OK(ctx, ctx->GetAttr("upx", &m_attribs.upx));
+        OP_REQUIRES_OK(ctx, ctx->GetAttr("upy", &m_attribs.upy));
+        OP_REQUIRES_OK(ctx, ctx->GetAttr("downx", &m_attribs.downx));
+        OP_REQUIRES_OK(ctx, ctx->GetAttr("downy", &m_attribs.downy));
+        OP_REQUIRES_OK(ctx, ctx->GetAttr("padx0", &m_attribs.padx0));
+        OP_REQUIRES_OK(ctx, ctx->GetAttr("padx1", &m_attribs.padx1));
+        OP_REQUIRES_OK(ctx, ctx->GetAttr("pady0", &m_attribs.pady0));
+        OP_REQUIRES_OK(ctx, ctx->GetAttr("pady1", &m_attribs.pady1));
+        OP_REQUIRES(ctx, m_attribs.upx >= 1 && m_attribs.upy >= 1, errors::InvalidArgument("upx and upy must be at least 1x1"));
+        OP_REQUIRES(ctx, m_attribs.downx >= 1 && m_attribs.downy >= 1, errors::InvalidArgument("downx and downy must be at least 1x1"));
+    }
+
+    void Compute(OpKernelContext* ctx)
+    {
+        UpFirDn2DKernelParams<T> p = m_attribs;
+        cudaStream_t stream = ctx->eigen_device<Eigen::GpuDevice>().stream();
+
+        const Tensor& x = ctx->input(0); // [majorDim, inH, inW, minorDim]
+        const Tensor& k = ctx->input(1); // [kernelH, kernelW]
+        p.x = x.flat<T>().data();
+        p.k = k.flat<T>().data();
+        OP_REQUIRES(ctx, x.dims() == 4, errors::InvalidArgument("input must have rank 4"));
+        OP_REQUIRES(ctx, k.dims() == 2, errors::InvalidArgument("kernel must have rank 2"));
+        OP_REQUIRES(ctx, x.NumElements() <= kint32max, errors::InvalidArgument("input too large"));
+        OP_REQUIRES(ctx, k.NumElements() <= kint32max, errors::InvalidArgument("kernel too large"));
+
+        p.majorDim  = (int)x.dim_size(0);
+        p.inH       = (int)x.dim_size(1);
+        p.inW       = (int)x.dim_size(2);
+        p.minorDim  = (int)x.dim_size(3);
+        p.kernelH   = (int)k.dim_size(0);
+        p.kernelW   = (int)k.dim_size(1);
+        OP_REQUIRES(ctx, p.kernelW >= 1 && p.kernelH >= 1, errors::InvalidArgument("kernel must be at least 1x1"));
+
+        p.outW = (p.inW * p.upx + p.padx0 + p.padx1 - p.kernelW + p.downx) / p.downx;
+        p.outH = (p.inH * p.upy + p.pady0 + p.pady1 - p.kernelH + p.downy) / p.downy;
+        OP_REQUIRES(ctx, p.outW >= 1 && p.outH >= 1, errors::InvalidArgument("output must be at least 1x1"));
+
+        Tensor* y = NULL; // [majorDim, outH, outW, minorDim]
+        TensorShape ys;
+        ys.AddDim(p.majorDim);
+        ys.AddDim(p.outH);
+        ys.AddDim(p.outW);
+        ys.AddDim(p.minorDim);
+        OP_REQUIRES_OK(ctx, ctx->allocate_output(0, ys, &y));
+        p.y = y->flat<T>().data();
+        OP_REQUIRES(ctx, y->NumElements() <= kint32max, errors::InvalidArgument("output too large"));
+
+        // Choose CUDA kernel to use.
+        void* cudaKernel = (void*)UpFirDn2DKernel_large<T>;
+        int tileOutW = -1;
+        int tileOutH = -1;
+        if (p.upx == 1 && p.upy == 1 && p.downx == 1 && p.downy == 1 && p.kernelW <= 7 && p.kernelH <= 7) { cudaKernel = (void*)UpFirDn2DKernel_small<T, 1,1, 1,1, 7,7, 64,16>; tileOutW = 64; tileOutH = 16; }
+        if (p.upx == 1 && p.upy == 1 && p.downx == 1 && p.downy == 1 && p.kernelW <= 6 && p.kernelH <= 6) { cudaKernel = (void*)UpFirDn2DKernel_small<T, 1,1, 1,1, 6,6, 64,16>; tileOutW = 64; tileOutH = 16; }
+        if (p.upx == 1 && p.upy == 1 && p.downx == 1 && p.downy == 1 && p.kernelW <= 5 && p.kernelH <= 5) { cudaKernel = (void*)UpFirDn2DKernel_small<T, 1,1, 1,1, 5,5, 64,16>; tileOutW = 64; tileOutH = 16; }
+        if (p.upx == 1 && p.upy == 1 && p.downx == 1 && p.downy == 1 && p.kernelW <= 4 && p.kernelH <= 4) { cudaKernel = (void*)UpFirDn2DKernel_small<T, 1,1, 1,1, 4,4, 64,16>; tileOutW = 64; tileOutH = 16; }
+        if (p.upx == 1 && p.upy == 1 && p.downx == 1 && p.downy == 1 && p.kernelW <= 3 && p.kernelH <= 3) { cudaKernel = (void*)UpFirDn2DKernel_small<T, 1,1, 1,1, 3,3, 64,16>; tileOutW = 64; tileOutH = 16; }
+        if (p.upx == 2 && p.upy == 2 && p.downx == 1 && p.downy == 1 && p.kernelW <= 8 && p.kernelH <= 8) { cudaKernel = (void*)UpFirDn2DKernel_small<T, 2,2, 1,1, 8,8, 64,16>; tileOutW = 64; tileOutH = 16; }
+        if (p.upx == 2 && p.upy == 2 && p.downx == 1 && p.downy == 1 && p.kernelW <= 6 && p.kernelH <= 6) { cudaKernel = (void*)UpFirDn2DKernel_small<T, 2,2, 1,1, 6,6, 64,16>; tileOutW = 64; tileOutH = 16; }
+        if (p.upx == 2 && p.upy == 2 && p.downx == 1 && p.downy == 1 && p.kernelW <= 4 && p.kernelH <= 4) { cudaKernel = (void*)UpFirDn2DKernel_small<T, 2,2, 1,1, 4,4, 64,16>; tileOutW = 64; tileOutH = 16; }
+        if (p.upx == 2 && p.upy == 2 && p.downx == 1 && p.downy == 1 && p.kernelW <= 2 && p.kernelH <= 2) { cudaKernel = (void*)UpFirDn2DKernel_small<T, 2,2, 1,1, 2,2, 64,16>; tileOutW = 64; tileOutH = 16; }
+        if (p.upx == 1 && p.upy == 1 && p.downx == 2 && p.downy == 2 && p.kernelW <= 8 && p.kernelH <= 8) { cudaKernel = (void*)UpFirDn2DKernel_small<T, 1,1, 2,2, 8,8, 32,8>;  tileOutW = 32; tileOutH = 8;  }
+        if (p.upx == 1 && p.upy == 1 && p.downx == 2 && p.downy == 2 && p.kernelW <= 6 && p.kernelH <= 6) { cudaKernel = (void*)UpFirDn2DKernel_small<T, 1,1, 2,2, 6,6, 32,8>;  tileOutW = 32; tileOutH = 8;  }
+        if (p.upx == 1 && p.upy == 1 && p.downx == 2 && p.downy == 2 && p.kernelW <= 4 && p.kernelH <= 4) { cudaKernel = (void*)UpFirDn2DKernel_small<T, 1,1, 2,2, 4,4, 32,8>;  tileOutW = 32; tileOutH = 8;  }
+        if (p.upx == 1 && p.upy == 1 && p.downx == 2 && p.downy == 2 && p.kernelW <= 2 && p.kernelH <= 2) { cudaKernel = (void*)UpFirDn2DKernel_small<T, 1,1, 2,2, 2,2, 32,8>;  tileOutW = 32; tileOutH = 8;  }
+
+        // Choose launch params.
+        dim3 blockSize;
+        dim3 gridSize;
+        if (tileOutW > 0 && tileOutH > 0) // small
+        {
+            p.loopMajor = (p.majorDim - 1) / 16384 + 1;
+            p.loopX = 1;
+            blockSize = dim3(32 * 8, 1, 1);
+            gridSize = dim3(((p.outH - 1) / tileOutH + 1) * p.minorDim, (p.outW - 1) / (p.loopX * tileOutW) + 1, (p.majorDim - 1) / p.loopMajor + 1);
+        }
+        else // large
+        {
+            p.loopMajor = (p.majorDim - 1) / 16384 + 1;
+            p.loopX = 4;
+            blockSize = dim3(4, 32, 1);
+            gridSize = dim3((p.outH * p.minorDim - 1) / blockSize.x + 1, (p.outW - 1) / (p.loopX * blockSize.y) + 1, (p.majorDim - 1) / p.loopMajor + 1);
+        }
+
+        // Launch CUDA kernel.
+        void* args[] = {&p};
+        OP_CHECK_CUDA_ERROR(ctx, cudaLaunchKernel(cudaKernel, gridSize, blockSize, args, 0, stream));
+    }
+};
+
+REGISTER_OP("UpFirDn2D")
+    .Input      ("x: T")
+    .Input      ("k: T")
+    .Output     ("y: T")
+    .Attr       ("T: {float, half}")
+    .Attr       ("upx: int = 1")
+    .Attr       ("upy: int = 1")
+    .Attr       ("downx: int = 1")
+    .Attr       ("downy: int = 1")
+    .Attr       ("padx0: int = 0")
+    .Attr       ("padx1: int = 0")
+    .Attr       ("pady0: int = 0")
+    .Attr       ("pady1: int = 0");
+REGISTER_KERNEL_BUILDER(Name("UpFirDn2D").Device(DEVICE_GPU).TypeConstraint<float>("T"), UpFirDn2DOp<float>);
+REGISTER_KERNEL_BUILDER(Name("UpFirDn2D").Device(DEVICE_GPU).TypeConstraint<Eigen::half>("T"), UpFirDn2DOp<Eigen::half>);
+
+//------------------------------------------------------------------------

stylegan_human/dnnlib/tflib/ops/upfirdn_2d.py

@@ -0,0 +1,366 @@
+# Copyright (c) SenseTime Research. All rights reserved.
+
+# Copyright (c) 2019, NVIDIA Corporation. All rights reserved.
+#
+# This work is made available under the Nvidia Source Code License-NC.
+# To view a copy of this license, visit
+# https://nvlabs.github.io/stylegan2/license.html
+
+"""Custom TensorFlow ops for efficient resampling of 2D images."""
+
+import os
+import numpy as np
+import tensorflow as tf
+from .. import custom_ops
+
+def _get_plugin():
+    return custom_ops.get_plugin(os.path.splitext(__file__)[0] + '.cu')
+
+#----------------------------------------------------------------------------
+
+def upfirdn_2d(x, k, upx=1, upy=1, downx=1, downy=1, padx0=0, padx1=0, pady0=0, pady1=0, impl='cuda'):
+    r"""Pad, upsample, FIR filter, and downsample a batch of 2D images.
+
+    Accepts a batch of 2D images of the shape `[majorDim, inH, inW, minorDim]`
+    and performs the following operations for each image, batched across
+    `majorDim` and `minorDim`:
+
+    1. Pad the image with zeros by the specified number of pixels on each side
+       (`padx0`, `padx1`, `pady0`, `pady1`). Specifying a negative value
+       corresponds to cropping the image.
+
+    2. Upsample the image by inserting the zeros after each pixel (`upx`, `upy`).
+
+    3. Convolve the image with the specified 2D FIR filter (`k`), shrinking the
+       image so that the footprint of all output pixels lies within the input image.
+
+    4. Downsample the image by throwing away pixels (`downx`, `downy`).
+
+    This sequence of operations bears close resemblance to scipy.signal.upfirdn().
+    The fused op is considerably more efficient than performing the same calculation
+    using standard TensorFlow ops. It supports gradients of arbitrary order.
+
+    Args:
+        x:      Input tensor of the shape `[majorDim, inH, inW, minorDim]`.
+        k:      2D FIR filter of the shape `[firH, firW]`.
+        upx:    Integer upsampling factor along the X-axis (default: 1).
+        upy:    Integer upsampling factor along the Y-axis (default: 1).
+        downx:  Integer downsampling factor along the X-axis (default: 1).
+        downy:  Integer downsampling factor along the Y-axis (default: 1).
+        padx0:  Number of pixels to pad on the left side (default: 0).
+        padx1:  Number of pixels to pad on the right side (default: 0).
+        pady0:  Number of pixels to pad on the top side (default: 0).
+        pady1:  Number of pixels to pad on the bottom side (default: 0).
+        impl:   Name of the implementation to use. Can be `"ref"` or `"cuda"` (default).
+
+    Returns:
+        Tensor of the shape `[majorDim, outH, outW, minorDim]`, and same datatype as `x`.
+    """
+
+    impl_dict = {
+        'ref':  _upfirdn_2d_ref,
+        'cuda': _upfirdn_2d_cuda,
+    }
+    return impl_dict[impl](x=x, k=k, upx=upx, upy=upy, downx=downx, downy=downy, padx0=padx0, padx1=padx1, pady0=pady0, pady1=pady1)
+
+#----------------------------------------------------------------------------
+
+def _upfirdn_2d_ref(x, k, upx, upy, downx, downy, padx0, padx1, pady0, pady1):
+    """Slow reference implementation of `upfirdn_2d()` using standard TensorFlow ops."""
+
+    x = tf.convert_to_tensor(x)
+    k = np.asarray(k, dtype=np.float32)
+    assert x.shape.rank == 4
+    inH = x.shape[1].value
+    inW = x.shape[2].value
+    minorDim = _shape(x, 3)
+    kernelH, kernelW = k.shape
+    assert inW >= 1 and inH >= 1
+    assert kernelW >= 1 and kernelH >= 1
+    assert isinstance(upx, int) and isinstance(upy, int)
+    assert isinstance(downx, int) and isinstance(downy, int)
+    assert isinstance(padx0, int) and isinstance(padx1, int)
+    assert isinstance(pady0, int) and isinstance(pady1, int)
+
+    # Upsample (insert zeros).
+    x = tf.reshape(x, [-1, inH, 1, inW, 1, minorDim])
+    x = tf.pad(x, [[0, 0], [0, 0], [0, upy - 1], [0, 0], [0, upx - 1], [0, 0]])
+    x = tf.reshape(x, [-1, inH * upy, inW * upx, minorDim])
+
+    # Pad (crop if negative).
+    x = tf.pad(x, [[0, 0], [max(pady0, 0), max(pady1, 0)], [max(padx0, 0), max(padx1, 0)], [0, 0]])
+    x = x[:, max(-pady0, 0) : x.shape[1].value - max(-pady1, 0), max(-padx0, 0) : x.shape[2].value - max(-padx1, 0), :]
+
+    # Convolve with filter.
+    x = tf.transpose(x, [0, 3, 1, 2])
+    x = tf.reshape(x, [-1, 1, inH * upy + pady0 + pady1, inW * upx + padx0 + padx1])
+    w = tf.constant(k[::-1, ::-1, np.newaxis, np.newaxis], dtype=x.dtype)
+    x = tf.nn.conv2d(x, w, strides=[1,1,1,1], padding='VALID', data_format='NCHW')
+    x = tf.reshape(x, [-1, minorDim, inH * upy + pady0 + pady1 - kernelH + 1, inW * upx + padx0 + padx1 - kernelW + 1])
+    x = tf.transpose(x, [0, 2, 3, 1])
+
+    # Downsample (throw away pixels).
+    return x[:, ::downy, ::downx, :]
+
+#----------------------------------------------------------------------------
+
+def _upfirdn_2d_cuda(x, k, upx, upy, downx, downy, padx0, padx1, pady0, pady1):
+    """Fast CUDA implementation of `upfirdn_2d()` using custom ops."""
+
+    x = tf.convert_to_tensor(x)
+    k = np.asarray(k, dtype=np.float32)
+    majorDim, inH, inW, minorDim = x.shape.as_list()
+    kernelH, kernelW = k.shape
+    assert inW >= 1 and inH >= 1
+    assert kernelW >= 1 and kernelH >= 1
+    assert isinstance(upx, int) and isinstance(upy, int)
+    assert isinstance(downx, int) and isinstance(downy, int)
+    assert isinstance(padx0, int) and isinstance(padx1, int)
+    assert isinstance(pady0, int) and isinstance(pady1, int)
+
+    outW = (inW * upx + padx0 + padx1 - kernelW) // downx + 1
+    outH = (inH * upy + pady0 + pady1 - kernelH) // downy + 1
+    assert outW >= 1 and outH >= 1
+
+    kc = tf.constant(k, dtype=x.dtype)
+    gkc = tf.constant(k[::-1, ::-1], dtype=x.dtype)
+    gpadx0 = kernelW - padx0 - 1
+    gpady0 = kernelH - pady0 - 1
+    gpadx1 = inW * upx - outW * downx + padx0 - upx + 1
+    gpady1 = inH * upy - outH * downy + pady0 - upy + 1
+
+    @tf.custom_gradient
+    def func(x):
+        y = _get_plugin().up_fir_dn2d(x=x, k=kc, upx=upx, upy=upy, downx=downx, downy=downy, padx0=padx0, padx1=padx1, pady0=pady0, pady1=pady1)
+        y.set_shape([majorDim, outH, outW, minorDim])
+        @tf.custom_gradient
+        def grad(dy):
+            dx = _get_plugin().up_fir_dn2d(x=dy, k=gkc, upx=downx, upy=downy, downx=upx, downy=upy, padx0=gpadx0, padx1=gpadx1, pady0=gpady0, pady1=gpady1)
+            dx.set_shape([majorDim, inH, inW, minorDim])
+            return dx, func
+        return y, grad
+    return func(x)
+
+#----------------------------------------------------------------------------
+
+def filter_2d(x, k, gain=1, data_format='NCHW', impl='cuda'):
+    r"""Filter a batch of 2D images with the given FIR filter.
+
+    Accepts a batch of 2D images of the shape `[N, C, H, W]` or `[N, H, W, C]`
+    and filters each image with the given filter. The filter is normalized so that
+    if the input pixels are constant, they will be scaled by the specified `gain`.
+    Pixels outside the image are assumed to be zero.
+
+    Args:
+        x:            Input tensor of the shape `[N, C, H, W]` or `[N, H, W, C]`.
+        k:            FIR filter of the shape `[firH, firW]` or `[firN]` (separable).
+        gain:         Scaling factor for signal magnitude (default: 1.0).
+        data_format:  `'NCHW'` or `'NHWC'` (default: `'NCHW'`).
+        impl:         Name of the implementation to use. Can be `"ref"` or `"cuda"` (default).
+
+    Returns:
+        Tensor of the same shape and datatype as `x`.
+    """
+
+    k = _setup_kernel(k) * gain
+    p = k.shape[0] - 1
+    return _simple_upfirdn_2d(x, k, pad0=(p+1)//2, pad1=p//2, data_format=data_format, impl=impl)
+
+#----------------------------------------------------------------------------
+
+def upsample_2d(x, k=None, factor=2, gain=1, data_format='NCHW', impl='cuda'):
+    r"""Upsample a batch of 2D images with the given filter.
+
+    Accepts a batch of 2D images of the shape `[N, C, H, W]` or `[N, H, W, C]`
+    and upsamples each image with the given filter. The filter is normalized so that
+    if the input pixels are constant, they will be scaled by the specified `gain`.
+    Pixels outside the image are assumed to be zero, and the filter is padded with
+    zeros so that its shape is a multiple of the upsampling factor.
+
+    Args:
+        x:            Input tensor of the shape `[N, C, H, W]` or `[N, H, W, C]`.
+        k:            FIR filter of the shape `[firH, firW]` or `[firN]` (separable).
+                      The default is `[1] * factor`, which corresponds to nearest-neighbor
+                      upsampling.
+        factor:       Integer upsampling factor (default: 2).
+        gain:         Scaling factor for signal magnitude (default: 1.0).
+        data_format:  `'NCHW'` or `'NHWC'` (default: `'NCHW'`).
+        impl:         Name of the implementation to use. Can be `"ref"` or `"cuda"` (default).
+
+    Returns:
+        Tensor of the shape `[N, C, H * factor, W * factor]` or
+        `[N, H * factor, W * factor, C]`, and same datatype as `x`.
+    """
+
+    assert isinstance(factor, int) and factor >= 1
+    if k is None:
+        k = [1] * factor
+    k = _setup_kernel(k) * (gain * (factor ** 2))
+    p = k.shape[0] - factor
+    return _simple_upfirdn_2d(x, k, up=factor, pad0=(p+1)//2+factor-1, pad1=p//2, data_format=data_format, impl=impl)
+
+#----------------------------------------------------------------------------
+
+def downsample_2d(x, k=None, factor=2, gain=1, data_format='NCHW', impl='cuda'):
+    r"""Downsample a batch of 2D images with the given filter.
+
+    Accepts a batch of 2D images of the shape `[N, C, H, W]` or `[N, H, W, C]`
+    and downsamples each image with the given filter. The filter is normalized so that
+    if the input pixels are constant, they will be scaled by the specified `gain`.
+    Pixels outside the image are assumed to be zero, and the filter is padded with
+    zeros so that its shape is a multiple of the downsampling factor.
+
+    Args:
+        x:            Input tensor of the shape `[N, C, H, W]` or `[N, H, W, C]`.
+        k:            FIR filter of the shape `[firH, firW]` or `[firN]` (separable).
+                      The default is `[1] * factor`, which corresponds to average pooling.
+        factor:       Integer downsampling factor (default: 2).
+        gain:         Scaling factor for signal magnitude (default: 1.0).
+        data_format:  `'NCHW'` or `'NHWC'` (default: `'NCHW'`).
+        impl:         Name of the implementation to use. Can be `"ref"` or `"cuda"` (default).
+
+    Returns:
+        Tensor of the shape `[N, C, H // factor, W // factor]` or
+        `[N, H // factor, W // factor, C]`, and same datatype as `x`.
+    """
+
+    assert isinstance(factor, int) and factor >= 1
+    if k is None:
+        k = [1] * factor
+    k = _setup_kernel(k) * gain
+    p = k.shape[0] - factor
+    return _simple_upfirdn_2d(x, k, down=factor, pad0=(p+1)//2, pad1=p//2, data_format=data_format, impl=impl)
+
+#----------------------------------------------------------------------------
+
+def upsample_conv_2d(x, w, k=None, factor=2, gain=1, data_format='NCHW', impl='cuda'):
+    r"""Fused `upsample_2d()` followed by `tf.nn.conv2d()`.
+
+    Padding is performed only once at the beginning, not between the operations.
+    The fused op is considerably more efficient than performing the same calculation
+    using standard TensorFlow ops. It supports gradients of arbitrary order.
+
+    Args:
+        x:            Input tensor of the shape `[N, C, H, W]` or `[N, H, W, C]`.
+        w:            Weight tensor of the shape `[filterH, filterW, inChannels, outChannels]`.
+                      Grouped convolution can be performed by `inChannels = x.shape[0] // numGroups`.
+        k:            FIR filter of the shape `[firH, firW]` or `[firN]` (separable).
+                      The default is `[1] * factor`, which corresponds to nearest-neighbor
+                      upsampling.
+        factor:       Integer upsampling factor (default: 2).
+        gain:         Scaling factor for signal magnitude (default: 1.0).
+        data_format:  `'NCHW'` or `'NHWC'` (default: `'NCHW'`).
+        impl:         Name of the implementation to use. Can be `"ref"` or `"cuda"` (default).
+
+    Returns:
+        Tensor of the shape `[N, C, H * factor, W * factor]` or
+        `[N, H * factor, W * factor, C]`, and same datatype as `x`.
+    """
+
+    assert isinstance(factor, int) and factor >= 1
+
+    # Check weight shape.
+    w = tf.convert_to_tensor(w)
+    assert w.shape.rank == 4
+    convH = w.shape[0].value
+    convW = w.shape[1].value
+    inC = _shape(w, 2)
+    outC = _shape(w, 3)
+    assert convW == convH
+
+    # Setup filter kernel.
+    if k is None:
+        k = [1] * factor
+    k = _setup_kernel(k) * (gain * (factor ** 2))
+    p = (k.shape[0] - factor) - (convW - 1)
+
+    # Determine data dimensions.
+    if data_format == 'NCHW':
+        stride = [1, 1, factor, factor]
+        output_shape = [_shape(x, 0), outC, (_shape(x, 2) - 1) * factor + convH, (_shape(x, 3) - 1) * factor + convW]
+        num_groups = _shape(x, 1) // inC
+    else:
+        stride = [1, factor, factor, 1]
+        output_shape = [_shape(x, 0), (_shape(x, 1) - 1) * factor + convH, (_shape(x, 2) - 1) * factor + convW, outC]
+        num_groups = _shape(x, 3) // inC
+
+    # Transpose weights.
+    w = tf.reshape(w, [convH, convW, inC, num_groups, -1])
+    w = tf.transpose(w[::-1, ::-1], [0, 1, 4, 3, 2])
+    w = tf.reshape(w, [convH, convW, -1, num_groups * inC])
+
+    # Execute.
+    x = tf.nn.conv2d_transpose(x, w, output_shape=output_shape, strides=stride, padding='VALID', data_format=data_format)
+    return _simple_upfirdn_2d(x, k, pad0=(p+1)//2+factor-1, pad1=p//2+1, data_format=data_format, impl=impl)
+
+#----------------------------------------------------------------------------
+
+def conv_downsample_2d(x, w, k=None, factor=2, gain=1, data_format='NCHW', impl='cuda'):
+    r"""Fused `tf.nn.conv2d()` followed by `downsample_2d()`.
+
+    Padding is performed only once at the beginning, not between the operations.
+    The fused op is considerably more efficient than performing the same calculation
+    using standard TensorFlow ops. It supports gradients of arbitrary order.
+
+    Args:
+        x:            Input tensor of the shape `[N, C, H, W]` or `[N, H, W, C]`.
+        w:            Weight tensor of the shape `[filterH, filterW, inChannels, outChannels]`.
+                      Grouped convolution can be performed by `inChannels = x.shape[0] // numGroups`.
+        k:            FIR filter of the shape `[firH, firW]` or `[firN]` (separable).
+                      The default is `[1] * factor`, which corresponds to average pooling.
+        factor:       Integer downsampling factor (default: 2).
+        gain:         Scaling factor for signal magnitude (default: 1.0).
+        data_format:  `'NCHW'` or `'NHWC'` (default: `'NCHW'`).
+        impl:         Name of the implementation to use. Can be `"ref"` or `"cuda"` (default).
+
+    Returns:
+        Tensor of the shape `[N, C, H // factor, W // factor]` or
+        `[N, H // factor, W // factor, C]`, and same datatype as `x`.
+    """
+
+    assert isinstance(factor, int) and factor >= 1
+    w = tf.convert_to_tensor(w)
+    convH, convW, _inC, _outC = w.shape.as_list()
+    assert convW == convH
+    if k is None:
+        k = [1] * factor
+    k = _setup_kernel(k) * gain
+    p = (k.shape[0] - factor) + (convW - 1)
+    if data_format == 'NCHW':
+        s = [1, 1, factor, factor]
+    else:
+        s = [1, factor, factor, 1]
+    x = _simple_upfirdn_2d(x, k, pad0=(p+1)//2, pad1=p//2, data_format=data_format, impl=impl)
+    return tf.nn.conv2d(x, w, strides=s, padding='VALID', data_format=data_format)
+
+#----------------------------------------------------------------------------
+# Internal helper funcs.
+
+def _shape(tf_expr, dim_idx):
+    if tf_expr.shape.rank is not None:
+        dim = tf_expr.shape[dim_idx].value
+        if dim is not None:
+            return dim
+    return tf.shape(tf_expr)[dim_idx]
+
+def _setup_kernel(k):
+    k = np.asarray(k, dtype=np.float32)
+    if k.ndim == 1:
+        k = np.outer(k, k)
+    k /= np.sum(k)
+    assert k.ndim == 2
+    assert k.shape[0] == k.shape[1]
+    return k
+
+def _simple_upfirdn_2d(x, k, up=1, down=1, pad0=0, pad1=0, data_format='NCHW', impl='cuda'):
+    assert data_format in ['NCHW', 'NHWC']
+    assert x.shape.rank == 4
+    y = x
+    if data_format == 'NCHW':
+        y = tf.reshape(y, [-1, _shape(y, 2), _shape(y, 3), 1])
+    y = upfirdn_2d(y, k, upx=up, upy=up, downx=down, downy=down, padx0=pad0, padx1=pad1, pady0=pad0, pady1=pad1, impl=impl)
+    if data_format == 'NCHW':
+        y = tf.reshape(y, [-1, _shape(x, 1), _shape(y, 1), _shape(y, 2)])
+    return y
+
+#----------------------------------------------------------------------------

stylegan_human/dnnlib/tflib/optimizer.py

@@ -0,0 +1,338 @@
+# Copyright (c) SenseTime Research. All rights reserved.
+
+# Copyright (c) 2019, NVIDIA Corporation. All rights reserved.
+#
+# This work is made available under the Nvidia Source Code License-NC.
+# To view a copy of this license, visit
+# https://nvlabs.github.io/stylegan2/license.html
+
+"""Helper wrapper for a Tensorflow optimizer."""
+
+import numpy as np
+import tensorflow as tf
+
+from collections import OrderedDict
+from typing import List, Union
+
+from . import autosummary
+from . import tfutil
+from .. import util
+
+from .tfutil import TfExpression, TfExpressionEx
+
+try:
+    # TensorFlow 1.13
+    from tensorflow.python.ops import nccl_ops
+except:
+    # Older TensorFlow versions
+    import tensorflow.contrib.nccl as nccl_ops
+
+class Optimizer:
+    """A Wrapper for tf.train.Optimizer.
+
+    Automatically takes care of:
+    - Gradient averaging for multi-GPU training.
+    - Gradient accumulation for arbitrarily large minibatches.
+    - Dynamic loss scaling and typecasts for FP16 training.
+    - Ignoring corrupted gradients that contain NaNs/Infs.
+    - Reporting statistics.
+    - Well-chosen default settings.
+    """
+
+    def __init__(self,
+        name:                   str             = "Train",                  # Name string that will appear in TensorFlow graph.
+        tf_optimizer:           str             = "tf.train.AdamOptimizer", # Underlying optimizer class.
+        learning_rate:          TfExpressionEx  = 0.001,                    # Learning rate. Can vary over time.
+        minibatch_multiplier:   TfExpressionEx  = None,                     # Treat N consecutive minibatches as one by accumulating gradients.
+        share:                  "Optimizer"     = None,                     # Share internal state with a previously created optimizer?
+        use_loss_scaling:       bool            = False,                    # Enable dynamic loss scaling for robust mixed-precision training?
+        loss_scaling_init:      float           = 64.0,                     # Log2 of initial loss scaling factor.
+        loss_scaling_inc:       float           = 0.0005,                   # Log2 of per-minibatch loss scaling increment when there is no overflow.
+        loss_scaling_dec:       float           = 1.0,                      # Log2 of per-minibatch loss scaling decrement when there is an overflow.
+        report_mem_usage:       bool            = False,                    # Report fine-grained memory usage statistics in TensorBoard?
+        **kwargs):
+
+        # Public fields.
+        self.name                   = name
+        self.learning_rate          = learning_rate
+        self.minibatch_multiplier   = minibatch_multiplier
+        self.id                     = self.name.replace("/", ".")
+        self.scope                  = tf.get_default_graph().unique_name(self.id)
+        self.optimizer_class        = util.get_obj_by_name(tf_optimizer)
+        self.optimizer_kwargs       = dict(kwargs)
+        self.use_loss_scaling       = use_loss_scaling
+        self.loss_scaling_init      = loss_scaling_init
+        self.loss_scaling_inc       = loss_scaling_inc
+        self.loss_scaling_dec       = loss_scaling_dec
+
+        # Private fields.
+        self._updates_applied       = False
+        self._devices               = OrderedDict() # device_name => EasyDict()
+        self._shared_optimizers     = OrderedDict() # device_name => optimizer_class
+        self._gradient_shapes       = None          # [shape, ...]
+        self._report_mem_usage      = report_mem_usage
+
+        # Validate arguments.
+        assert callable(self.optimizer_class)
+
+        # Share internal state if requested.
+        if share is not None:
+            assert isinstance(share, Optimizer)
+            assert self.optimizer_class is share.optimizer_class
+            assert self.learning_rate is share.learning_rate
+            assert self.optimizer_kwargs == share.optimizer_kwargs
+            self._shared_optimizers = share._shared_optimizers # pylint: disable=protected-access
+
+    def _get_device(self, device_name: str):
+        """Get internal state for the given TensorFlow device."""
+        tfutil.assert_tf_initialized()
+        if device_name in self._devices:
+            return self._devices[device_name]
+
+        # Initialize fields.
+        device = util.EasyDict()
+        device.name             = device_name
+        device.optimizer        = None          # Underlying optimizer:     optimizer_class
+        device.loss_scaling_var = None          # Log2 of loss scaling:     tf.Variable
+        device.grad_raw         = OrderedDict() # Raw gradients:            var => [grad, ...]
+        device.grad_clean       = OrderedDict() # Clean gradients:          var => grad
+        device.grad_acc_vars    = OrderedDict() # Accumulation sums:        var => tf.Variable
+        device.grad_acc_count   = None          # Accumulation counter:     tf.Variable
+        device.grad_acc         = OrderedDict() # Accumulated gradients:    var => grad
+
+        # Setup TensorFlow objects.
+        with tfutil.absolute_name_scope(self.scope + "/Devices"), tf.device(device_name), tf.control_dependencies(None):
+            if device_name not in self._shared_optimizers:
+                optimizer_name = self.scope.replace("/", "_") + "_opt%d" % len(self._shared_optimizers)
+                self._shared_optimizers[device_name] = self.optimizer_class(name=optimizer_name, learning_rate=self.learning_rate, **self.optimizer_kwargs)
+            device.optimizer = self._shared_optimizers[device_name]
+            if self.use_loss_scaling:
+                device.loss_scaling_var = tf.Variable(np.float32(self.loss_scaling_init), trainable=False, name="loss_scaling_var")
+
+        # Register device.
+        self._devices[device_name] = device
+        return device
+
+    def register_gradients(self, loss: TfExpression, trainable_vars: Union[List, dict]) -> None:
+        """Register the gradients of the given loss function with respect to the given variables.
+        Intended to be called once per GPU."""
+        tfutil.assert_tf_initialized()
+        assert not self._updates_applied
+        device = self._get_device(loss.device)
+
+        # Validate trainables.
+        if isinstance(trainable_vars, dict):
+            trainable_vars = list(trainable_vars.values())  # allow passing in Network.trainables as vars
+        assert isinstance(trainable_vars, list) and len(trainable_vars) >= 1
+        assert all(tfutil.is_tf_expression(expr) for expr in trainable_vars + [loss])
+        assert all(var.device == device.name for var in trainable_vars)
+
+        # Validate shapes.
+        if self._gradient_shapes is None:
+            self._gradient_shapes = [var.shape.as_list() for var in trainable_vars]
+        assert len(trainable_vars) == len(self._gradient_shapes)
+        assert all(var.shape.as_list() == var_shape for var, var_shape in zip(trainable_vars, self._gradient_shapes))
+
+        # Report memory usage if requested.
+        deps = []
+        if self._report_mem_usage:
+            self._report_mem_usage = False
+            try:
+                with tf.name_scope(self.id + '_mem'), tf.device(device.name), tf.control_dependencies([loss]):
+                    deps.append(autosummary.autosummary(self.id + "/mem_usage_gb", tf.contrib.memory_stats.BytesInUse() / 2**30))
+            except tf.errors.NotFoundError:
+                pass
+
+        # Compute gradients.
+        with tf.name_scope(self.id + "_grad"), tf.device(device.name), tf.control_dependencies(deps):
+            loss = self.apply_loss_scaling(tf.cast(loss, tf.float32))
+            gate = tf.train.Optimizer.GATE_NONE  # disable gating to reduce memory usage
+            grad_list = device.optimizer.compute_gradients(loss=loss, var_list=trainable_vars, gate_gradients=gate)
+
+        # Register gradients.
+        for grad, var in grad_list:
+            if var not in device.grad_raw:
+                device.grad_raw[var] = []
+            device.grad_raw[var].append(grad)
+
+    def apply_updates(self, allow_no_op: bool = False) -> tf.Operation:
+        """Construct training op to update the registered variables based on their gradients."""
+        tfutil.assert_tf_initialized()
+        assert not self._updates_applied
+        self._updates_applied = True
+        all_ops = []
+
+        # Check for no-op.
+        if allow_no_op and len(self._devices) == 0:
+            with tfutil.absolute_name_scope(self.scope):
+                return tf.no_op(name='TrainingOp')
+
+        # Clean up gradients.
+        for device_idx, device in enumerate(self._devices.values()):
+            with tfutil.absolute_name_scope(self.scope + "/Clean%d" % device_idx), tf.device(device.name):
+                for var, grad in device.grad_raw.items():
+
+                    # Filter out disconnected gradients and convert to float32.
+                    grad = [g for g in grad if g is not None]
+                    grad = [tf.cast(g, tf.float32) for g in grad]
+
+                    # Sum within the device.
+                    if len(grad) == 0:
+                        grad = tf.zeros(var.shape)  # No gradients => zero.
+                    elif len(grad) == 1:
+                        grad = grad[0]              # Single gradient => use as is.
+                    else:
+                        grad = tf.add_n(grad)       # Multiple gradients => sum.
+
+                    # Scale as needed.
+                    scale = 1.0 / len(device.grad_raw[var]) / len(self._devices)
+                    scale = tf.constant(scale, dtype=tf.float32, name="scale")
+                    if self.minibatch_multiplier is not None:
+                        scale /= tf.cast(self.minibatch_multiplier, tf.float32)
+                    scale = self.undo_loss_scaling(scale)
+                    device.grad_clean[var] = grad * scale
+
+        # Sum gradients across devices.
+        if len(self._devices) > 1:
+            with tfutil.absolute_name_scope(self.scope + "/Broadcast"), tf.device(None):
+                for all_vars in zip(*[device.grad_clean.keys() for device in self._devices.values()]):
+                    if len(all_vars) > 0 and all(dim > 0 for dim in all_vars[0].shape.as_list()): # NCCL does not support zero-sized tensors.
+                        all_grads = [device.grad_clean[var] for device, var in zip(self._devices.values(), all_vars)]
+                        all_grads = nccl_ops.all_sum(all_grads)
+                        for device, var, grad in zip(self._devices.values(), all_vars, all_grads):
+                            device.grad_clean[var] = grad
+
+        # Apply updates separately on each device.
+        for device_idx, device in enumerate(self._devices.values()):
+            with tfutil.absolute_name_scope(self.scope + "/Apply%d" % device_idx), tf.device(device.name):
+                # pylint: disable=cell-var-from-loop
+
+                # Accumulate gradients over time.
+                if self.minibatch_multiplier is None:
+                    acc_ok = tf.constant(True, name='acc_ok')
+                    device.grad_acc = OrderedDict(device.grad_clean)
+                else:
+                    # Create variables.
+                    with tf.control_dependencies(None):
+                        for var in device.grad_clean.keys():
+                            device.grad_acc_vars[var] = tf.Variable(tf.zeros(var.shape), trainable=False, name="grad_acc_var")
+                        device.grad_acc_count = tf.Variable(tf.zeros([]), trainable=False, name="grad_acc_count")
+
+                    # Track counter.
+                    count_cur = device.grad_acc_count + 1.0
+                    count_inc_op = lambda: tf.assign(device.grad_acc_count, count_cur)
+                    count_reset_op = lambda: tf.assign(device.grad_acc_count, tf.zeros([]))
+                    acc_ok = (count_cur >= tf.cast(self.minibatch_multiplier, tf.float32))
+                    all_ops.append(tf.cond(acc_ok, count_reset_op, count_inc_op))
+
+                    # Track gradients.
+                    for var, grad in device.grad_clean.items():
+                        acc_var = device.grad_acc_vars[var]
+                        acc_cur = acc_var + grad
+                        device.grad_acc[var] = acc_cur
+                        with tf.control_dependencies([acc_cur]):
+                            acc_inc_op = lambda: tf.assign(acc_var, acc_cur)
+                            acc_reset_op = lambda: tf.assign(acc_var, tf.zeros(var.shape))
+                            all_ops.append(tf.cond(acc_ok, acc_reset_op, acc_inc_op))
+
+                # No overflow => apply gradients.
+                all_ok = tf.reduce_all(tf.stack([acc_ok] + [tf.reduce_all(tf.is_finite(g)) for g in device.grad_acc.values()]))
+                apply_op = lambda: device.optimizer.apply_gradients([(tf.cast(grad, var.dtype), var) for var, grad in device.grad_acc.items()])
+                all_ops.append(tf.cond(all_ok, apply_op, tf.no_op))
+
+                # Adjust loss scaling.
+                if self.use_loss_scaling:
+                    ls_inc_op = lambda: tf.assign_add(device.loss_scaling_var, self.loss_scaling_inc)
+                    ls_dec_op = lambda: tf.assign_sub(device.loss_scaling_var, self.loss_scaling_dec)
+                    ls_update_op = lambda: tf.group(tf.cond(all_ok, ls_inc_op, ls_dec_op))
+                    all_ops.append(tf.cond(acc_ok, ls_update_op, tf.no_op))
+
+                # Last device => report statistics.
+                if device_idx == len(self._devices) - 1:
+                    all_ops.append(autosummary.autosummary(self.id + "/learning_rate", self.learning_rate))
+                    all_ops.append(autosummary.autosummary(self.id + "/overflow_frequency", tf.where(all_ok, 0, 1), condition=acc_ok))
+                    if self.use_loss_scaling:
+                        all_ops.append(autosummary.autosummary(self.id + "/loss_scaling_log2", device.loss_scaling_var))
+
+        # Initialize variables.
+        self.reset_optimizer_state()
+        if self.use_loss_scaling:
+            tfutil.init_uninitialized_vars([device.loss_scaling_var for device in self._devices.values()])
+        if self.minibatch_multiplier is not None:
+            tfutil.run([var.initializer for device in self._devices.values() for var in list(device.grad_acc_vars.values()) + [device.grad_acc_count]])
+
+        # Group everything into a single op.
+        with tfutil.absolute_name_scope(self.scope):
+            return tf.group(*all_ops, name="TrainingOp")
+
+    def reset_optimizer_state(self) -> None:
+        """Reset internal state of the underlying optimizer."""
+        tfutil.assert_tf_initialized()
+        tfutil.run([var.initializer for device in self._devices.values() for var in device.optimizer.variables()])
+
+    def get_loss_scaling_var(self, device: str) -> Union[tf.Variable, None]:
+        """Get or create variable representing log2 of the current dynamic loss scaling factor."""
+        return self._get_device(device).loss_scaling_var
+
+    def apply_loss_scaling(self, value: TfExpression) -> TfExpression:
+        """Apply dynamic loss scaling for the given expression."""
+        assert tfutil.is_tf_expression(value)
+        if not self.use_loss_scaling:
+            return value
+        return value * tfutil.exp2(self.get_loss_scaling_var(value.device))
+
+    def undo_loss_scaling(self, value: TfExpression) -> TfExpression:
+        """Undo the effect of dynamic loss scaling for the given expression."""
+        assert tfutil.is_tf_expression(value)
+        if not self.use_loss_scaling:
+            return value
+        return value * tfutil.exp2(-self.get_loss_scaling_var(value.device)) # pylint: disable=invalid-unary-operand-type
+
+
+class SimpleAdam:
+    """Simplified version of tf.train.AdamOptimizer that behaves identically when used with dnnlib.tflib.Optimizer."""
+
+    def __init__(self, name="Adam", learning_rate=0.001, beta1=0.9, beta2=0.999, epsilon=1e-8):
+        self.name = name
+        self.learning_rate = learning_rate
+        self.beta1 = beta1
+        self.beta2 = beta2
+        self.epsilon = epsilon
+        self.all_state_vars = []
+
+    def variables(self):
+        return self.all_state_vars
+
+    def compute_gradients(self, loss, var_list, gate_gradients=tf.train.Optimizer.GATE_NONE):
+        assert gate_gradients == tf.train.Optimizer.GATE_NONE
+        return list(zip(tf.gradients(loss, var_list), var_list))
+
+    def apply_gradients(self, grads_and_vars):
+        with tf.name_scope(self.name):
+            state_vars = []
+            update_ops = []
+
+            # Adjust learning rate to deal with startup bias.
+            with tf.control_dependencies(None):
+                b1pow_var = tf.Variable(dtype=tf.float32, initial_value=1, trainable=False)
+                b2pow_var = tf.Variable(dtype=tf.float32, initial_value=1, trainable=False)
+                state_vars += [b1pow_var, b2pow_var]
+            b1pow_new = b1pow_var * self.beta1
+            b2pow_new = b2pow_var * self.beta2
+            update_ops += [tf.assign(b1pow_var, b1pow_new), tf.assign(b2pow_var, b2pow_new)]
+            lr_new = self.learning_rate * tf.sqrt(1 - b2pow_new) / (1 - b1pow_new)
+
+            # Construct ops to update each variable.
+            for grad, var in grads_and_vars:
+                with tf.control_dependencies(None):
+                    m_var = tf.Variable(dtype=tf.float32, initial_value=tf.zeros_like(var), trainable=False)
+                    v_var = tf.Variable(dtype=tf.float32, initial_value=tf.zeros_like(var), trainable=False)
+                    state_vars += [m_var, v_var]
+                m_new = self.beta1 * m_var + (1 - self.beta1) * grad
+                v_new = self.beta2 * v_var + (1 - self.beta2) * tf.square(grad)
+                var_delta = lr_new * m_new / (tf.sqrt(v_new) + self.epsilon)
+                update_ops += [tf.assign(m_var, m_new), tf.assign(v_var, v_new), tf.assign_sub(var, var_delta)]
+
+            # Group everything together.
+            self.all_state_vars += state_vars
+            return tf.group(*update_ops)

stylegan_human/dnnlib/tflib/tfutil.py

@@ -0,0 +1,254 @@
+# Copyright (c) SenseTime Research. All rights reserved.
+
+# Copyright (c) 2019, NVIDIA Corporation. All rights reserved.
+#
+# This work is made available under the Nvidia Source Code License-NC.
+# To view a copy of this license, visit
+# https://nvlabs.github.io/stylegan2/license.html
+
+"""Miscellaneous helper utils for Tensorflow."""
+
+import os
+import numpy as np
+import tensorflow as tf
+
+# Silence deprecation warnings from TensorFlow 1.13 onwards
+import logging
+logging.getLogger('tensorflow').setLevel(logging.ERROR)
+import tensorflow.contrib   # requires TensorFlow 1.x!
+tf.contrib = tensorflow.contrib
+
+from typing import Any, Iterable, List, Union
+
+TfExpression = Union[tf.Tensor, tf.Variable, tf.Operation]
+"""A type that represents a valid Tensorflow expression."""
+
+TfExpressionEx = Union[TfExpression, int, float, np.ndarray]
+"""A type that can be converted to a valid Tensorflow expression."""
+
+
+def run(*args, **kwargs) -> Any:
+    """Run the specified ops in the default session."""
+    assert_tf_initialized()
+    return tf.get_default_session().run(*args, **kwargs)
+
+
+def is_tf_expression(x: Any) -> bool:
+    """Check whether the input is a valid Tensorflow expression, i.e., Tensorflow Tensor, Variable, or Operation."""
+    return isinstance(x, (tf.Tensor, tf.Variable, tf.Operation))
+
+
+def shape_to_list(shape: Iterable[tf.Dimension]) -> List[Union[int, None]]:
+    """Convert a Tensorflow shape to a list of ints. Retained for backwards compatibility -- use TensorShape.as_list() in new code."""
+    return [dim.value for dim in shape]
+
+
+def flatten(x: TfExpressionEx) -> TfExpression:
+    """Shortcut function for flattening a tensor."""
+    with tf.name_scope("Flatten"):
+        return tf.reshape(x, [-1])
+
+
+def log2(x: TfExpressionEx) -> TfExpression:
+    """Logarithm in base 2."""
+    with tf.name_scope("Log2"):
+        return tf.log(x) * np.float32(1.0 / np.log(2.0))
+
+
+def exp2(x: TfExpressionEx) -> TfExpression:
+    """Exponent in base 2."""
+    with tf.name_scope("Exp2"):
+        return tf.exp(x * np.float32(np.log(2.0)))
+
+
+def lerp(a: TfExpressionEx, b: TfExpressionEx, t: TfExpressionEx) -> TfExpressionEx:
+    """Linear interpolation."""
+    with tf.name_scope("Lerp"):
+        return a + (b - a) * t
+
+
+def lerp_clip(a: TfExpressionEx, b: TfExpressionEx, t: TfExpressionEx) -> TfExpression:
+    """Linear interpolation with clip."""
+    with tf.name_scope("LerpClip"):
+        return a + (b - a) * tf.clip_by_value(t, 0.0, 1.0)
+
+
+def absolute_name_scope(scope: str) -> tf.name_scope:
+    """Forcefully enter the specified name scope, ignoring any surrounding scopes."""
+    return tf.name_scope(scope + "/")
+
+
+def absolute_variable_scope(scope: str, **kwargs) -> tf.variable_scope:
+    """Forcefully enter the specified variable scope, ignoring any surrounding scopes."""
+    return tf.variable_scope(tf.VariableScope(name=scope, **kwargs), auxiliary_name_scope=False)
+
+
+def _sanitize_tf_config(config_dict: dict = None) -> dict:
+    # Defaults.
+    cfg = dict()
+    cfg["rnd.np_random_seed"]               = None      # Random seed for NumPy. None = keep as is.
+    cfg["rnd.tf_random_seed"]               = "auto"    # Random seed for TensorFlow. 'auto' = derive from NumPy random state. None = keep as is.
+    cfg["env.TF_CPP_MIN_LOG_LEVEL"]         = "1"       # 0 = Print all available debug info from TensorFlow. 1 = Print warnings and errors, but disable debug info.
+    cfg["graph_options.place_pruned_graph"] = True      # False = Check that all ops are available on the designated device. True = Skip the check for ops that are not used.
+    cfg["gpu_options.allow_growth"]         = True      # False = Allocate all GPU memory at the beginning. True = Allocate only as much GPU memory as needed.
+
+    # Remove defaults for environment variables that are already set.
+    for key in list(cfg):
+        fields = key.split(".")
+        if fields[0] == "env":
+            assert len(fields) == 2
+            if fields[1] in os.environ:
+                del cfg[key]
+
+    # User overrides.
+    if config_dict is not None:
+        cfg.update(config_dict)
+    return cfg
+
+
+def init_tf(config_dict: dict = None) -> None:
+    """Initialize TensorFlow session using good default settings."""
+    # Skip if already initialized.
+    if tf.get_default_session() is not None:
+        return
+
+    # Setup config dict and random seeds.
+    cfg = _sanitize_tf_config(config_dict)
+    np_random_seed = cfg["rnd.np_random_seed"]
+    if np_random_seed is not None:
+        np.random.seed(np_random_seed)
+    tf_random_seed = cfg["rnd.tf_random_seed"]
+    if tf_random_seed == "auto":
+        tf_random_seed = np.random.randint(1 << 31)
+    if tf_random_seed is not None:
+        tf.set_random_seed(tf_random_seed)
+
+    # Setup environment variables.
+    for key, value in cfg.items():
+        fields = key.split(".")
+        if fields[0] == "env":
+            assert len(fields) == 2
+            os.environ[fields[1]] = str(value)
+
+    # Create default TensorFlow session.
+    create_session(cfg, force_as_default=True)
+
+
+def assert_tf_initialized():
+    """Check that TensorFlow session has been initialized."""
+    if tf.get_default_session() is None:
+        raise RuntimeError("No default TensorFlow session found. Please call dnnlib.tflib.init_tf().")
+
+
+def create_session(config_dict: dict = None, force_as_default: bool = False) -> tf.Session:
+    """Create tf.Session based on config dict."""
+    # Setup TensorFlow config proto.
+    cfg = _sanitize_tf_config(config_dict)
+    config_proto = tf.ConfigProto()
+    for key, value in cfg.items():
+        fields = key.split(".")
+        if fields[0] not in ["rnd", "env"]:
+            obj = config_proto
+            for field in fields[:-1]:
+                obj = getattr(obj, field)
+            setattr(obj, fields[-1], value)
+
+    # Create session.
+    session = tf.Session(config=config_proto)
+    if force_as_default:
+        # pylint: disable=protected-access
+        session._default_session = session.as_default()
+        session._default_session.enforce_nesting = False
+        session._default_session.__enter__()
+    return session
+
+
+def init_uninitialized_vars(target_vars: List[tf.Variable] = None) -> None:
+    """Initialize all tf.Variables that have not already been initialized.
+
+    Equivalent to the following, but more efficient and does not bloat the tf graph:
+    tf.variables_initializer(tf.report_uninitialized_variables()).run()
+    """
+    assert_tf_initialized()
+    if target_vars is None:
+        target_vars = tf.global_variables()
+
+    test_vars = []
+    test_ops = []
+
+    with tf.control_dependencies(None):  # ignore surrounding control_dependencies
+        for var in target_vars:
+            assert is_tf_expression(var)
+
+            try:
+                tf.get_default_graph().get_tensor_by_name(var.name.replace(":0", "/IsVariableInitialized:0"))
+            except KeyError:
+                # Op does not exist => variable may be uninitialized.
+                test_vars.append(var)
+
+                with absolute_name_scope(var.name.split(":")[0]):
+                    test_ops.append(tf.is_variable_initialized(var))
+
+    init_vars = [var for var, inited in zip(test_vars, run(test_ops)) if not inited]
+    run([var.initializer for var in init_vars])
+
+
+def set_vars(var_to_value_dict: dict) -> None:
+    """Set the values of given tf.Variables.
+
+    Equivalent to the following, but more efficient and does not bloat the tf graph:
+    tflib.run([tf.assign(var, value) for var, value in var_to_value_dict.items()]
+    """
+    assert_tf_initialized()
+    ops = []
+    feed_dict = {}
+
+    for var, value in var_to_value_dict.items():
+        assert is_tf_expression(var)
+
+        try:
+            setter = tf.get_default_graph().get_tensor_by_name(var.name.replace(":0", "/setter:0"))  # look for existing op
+        except KeyError:
+            with absolute_name_scope(var.name.split(":")[0]):
+                with tf.control_dependencies(None):  # ignore surrounding control_dependencies
+                    setter = tf.assign(var, tf.placeholder(var.dtype, var.shape, "new_value"), name="setter")  # create new setter
+
+        ops.append(setter)
+        feed_dict[setter.op.inputs[1]] = value
+
+    run(ops, feed_dict)
+
+
+def create_var_with_large_initial_value(initial_value: np.ndarray, *args, **kwargs):
+    """Create tf.Variable with large initial value without bloating the tf graph."""
+    assert_tf_initialized()
+    assert isinstance(initial_value, np.ndarray)
+    zeros = tf.zeros(initial_value.shape, initial_value.dtype)
+    var = tf.Variable(zeros, *args, **kwargs)
+    set_vars({var: initial_value})
+    return var
+
+
+def convert_images_from_uint8(images, drange=[-1,1], nhwc_to_nchw=False):
+    """Convert a minibatch of images from uint8 to float32 with configurable dynamic range.
+    Can be used as an input transformation for Network.run().
+    """
+    images = tf.cast(images, tf.float32)
+    if nhwc_to_nchw:
+        images = tf.transpose(images, [0, 3, 1, 2])
+    return images * ((drange[1] - drange[0]) / 255) + drange[0]
+
+
+def convert_images_to_uint8(images, drange=[-1,1], nchw_to_nhwc=False, shrink=1):
+    """Convert a minibatch of images from float32 to uint8 with configurable dynamic range.
+    Can be used as an output transformation for Network.run().
+    """
+    images = tf.cast(images, tf.float32)
+    if shrink > 1:
+        ksize = [1, 1, shrink, shrink]
+        images = tf.nn.avg_pool(images, ksize=ksize, strides=ksize, padding="VALID", data_format="NCHW")
+    if nchw_to_nhwc:
+        images = tf.transpose(images, [0, 2, 3, 1])
+    scale = 255 / (drange[1] - drange[0])
+    images = images * scale + (0.5 - drange[0] * scale)
+    return tf.saturate_cast(images, tf.uint8)

stylegan_human/dnnlib/util.py

@@ -0,0 +1,479 @@
+# Copyright (c) SenseTime Research. All rights reserved.
+# Copyright (c) 2021, NVIDIA CORPORATION.  All rights reserved.
+#
+# NVIDIA CORPORATION and its licensors retain all intellectual property
+# and proprietary rights in and to this software, related documentation
+# and any modifications thereto.  Any use, reproduction, disclosure or
+# distribution of this software and related documentation without an express
+# license agreement from NVIDIA CORPORATION is strictly prohibited.
+
+"""Miscellaneous utility classes and functions."""
+
+import ctypes
+import fnmatch
+import importlib
+import inspect
+import numpy as np
+import os
+import shutil
+import sys
+import types
+import io
+import pickle
+import re
+import requests
+import html
+import hashlib
+import glob
+import tempfile
+import urllib
+import urllib.request
+import uuid
+
+from distutils.util import strtobool
+from typing import Any, List, Tuple, Union
+
+
+# Util classes
+# ------------------------------------------------------------------------------------------
+
+
+class EasyDict(dict):
+    """Convenience class that behaves like a dict but allows access with the attribute syntax."""
+
+    def __getattr__(self, name: str) -> Any:
+        try:
+            return self[name]
+        except KeyError:
+            raise AttributeError(name)
+
+    def __setattr__(self, name: str, value: Any) -> None:
+        self[name] = value
+
+    def __delattr__(self, name: str) -> None:
+        del self[name]
+
+
+class Logger(object):
+    """Redirect stderr to stdout, optionally print stdout to a file, and optionally force flushing on both stdout and the file."""
+
+    def __init__(self, file_name: str = None, file_mode: str = "w", should_flush: bool = True):
+        self.file = None
+
+        if file_name is not None:
+            self.file = open(file_name, file_mode)
+
+        self.should_flush = should_flush
+        self.stdout = sys.stdout
+        self.stderr = sys.stderr
+
+        sys.stdout = self
+        sys.stderr = self
+
+    def __enter__(self) -> "Logger":
+        return self
+
+    def __exit__(self, exc_type: Any, exc_value: Any, traceback: Any) -> None:
+        self.close()
+
+    def write(self, text: Union[str, bytes]) -> None:
+        """Write text to stdout (and a file) and optionally flush."""
+        if isinstance(text, bytes):
+            text = text.decode()
+        if len(text) == 0: # workaround for a bug in VSCode debugger: sys.stdout.write(''); sys.stdout.flush() => crash
+            return
+
+        if self.file is not None:
+            self.file.write(text)
+
+        self.stdout.write(text)
+
+        if self.should_flush:
+            self.flush()
+
+    def flush(self) -> None:
+        """Flush written text to both stdout and a file, if open."""
+        if self.file is not None:
+            self.file.flush()
+
+        self.stdout.flush()
+
+    def close(self) -> None:
+        """Flush, close possible files, and remove stdout/stderr mirroring."""
+        self.flush()
+
+        # if using multiple loggers, prevent closing in wrong order
+        if sys.stdout is self:
+            sys.stdout = self.stdout
+        if sys.stderr is self:
+            sys.stderr = self.stderr
+
+        if self.file is not None:
+            self.file.close()
+            self.file = None
+
+
+# Cache directories
+# ------------------------------------------------------------------------------------------
+
+_dnnlib_cache_dir = None
+
+def set_cache_dir(path: str) -> None:
+    global _dnnlib_cache_dir
+    _dnnlib_cache_dir = path
+
+def make_cache_dir_path(*paths: str) -> str:
+    if _dnnlib_cache_dir is not None:
+        return os.path.join(_dnnlib_cache_dir, *paths)
+    if 'DNNLIB_CACHE_DIR' in os.environ:
+        return os.path.join(os.environ['DNNLIB_CACHE_DIR'], *paths)
+    if 'HOME' in os.environ:
+        return os.path.join(os.environ['HOME'], '.cache', 'dnnlib', *paths)
+    if 'USERPROFILE' in os.environ:
+        return os.path.join(os.environ['USERPROFILE'], '.cache', 'dnnlib', *paths)
+    return os.path.join(tempfile.gettempdir(), '.cache', 'dnnlib', *paths)
+
+# Small util functions
+# ------------------------------------------------------------------------------------------
+
+
+def format_time(seconds: Union[int, float]) -> str:
+    """Convert the seconds to human readable string with days, hours, minutes and seconds."""
+    s = int(np.rint(seconds))
+
+    if s < 60:
+        return "{0}s".format(s)
+    elif s < 60 * 60:
+        return "{0}m {1:02}s".format(s // 60, s % 60)
+    elif s < 24 * 60 * 60:
+        return "{0}h {1:02}m {2:02}s".format(s // (60 * 60), (s // 60) % 60, s % 60)
+    else:
+        return "{0}d {1:02}h {2:02}m".format(s // (24 * 60 * 60), (s // (60 * 60)) % 24, (s // 60) % 60)
+
+
+def ask_yes_no(question: str) -> bool:
+    """Ask the user the question until the user inputs a valid answer."""
+    while True:
+        try:
+            print("{0} [y/n]".format(question))
+            return strtobool(input().lower())
+        except ValueError:
+            pass
+
+
+def tuple_product(t: Tuple) -> Any:
+    """Calculate the product of the tuple elements."""
+    result = 1
+
+    for v in t:
+        result *= v
+
+    return result
+
+
+_str_to_ctype = {
+    "uint8": ctypes.c_ubyte,
+    "uint16": ctypes.c_uint16,
+    "uint32": ctypes.c_uint32,
+    "uint64": ctypes.c_uint64,
+    "int8": ctypes.c_byte,
+    "int16": ctypes.c_int16,
+    "int32": ctypes.c_int32,
+    "int64": ctypes.c_int64,
+    "float32": ctypes.c_float,
+    "float64": ctypes.c_double
+}
+
+
+def get_dtype_and_ctype(type_obj: Any) -> Tuple[np.dtype, Any]:
+    """Given a type name string (or an object having a __name__ attribute), return matching Numpy and ctypes types that have the same size in bytes."""
+    type_str = None
+
+    if isinstance(type_obj, str):
+        type_str = type_obj
+    elif hasattr(type_obj, "__name__"):
+        type_str = type_obj.__name__
+    elif hasattr(type_obj, "name"):
+        type_str = type_obj.name
+    else:
+        raise RuntimeError("Cannot infer type name from input")
+
+    assert type_str in _str_to_ctype.keys()
+
+    my_dtype = np.dtype(type_str)
+    my_ctype = _str_to_ctype[type_str]
+
+    assert my_dtype.itemsize == ctypes.sizeof(my_ctype)
+
+    return my_dtype, my_ctype
+
+
+def is_pickleable(obj: Any) -> bool:
+    try:
+        with io.BytesIO() as stream:
+            pickle.dump(obj, stream)
+        return True
+    except:
+        return False
+
+
+# Functionality to import modules/objects by name, and call functions by name
+# ------------------------------------------------------------------------------------------
+
+def get_module_from_obj_name(obj_name: str) -> Tuple[types.ModuleType, str]:
+    """Searches for the underlying module behind the name to some python object.
+    Returns the module and the object name (original name with module part removed)."""
+
+    # allow convenience shorthands, substitute them by full names
+    obj_name = re.sub("^np.", "numpy.", obj_name)
+    obj_name = re.sub("^tf.", "tensorflow.", obj_name)
+
+    # list alternatives for (module_name, local_obj_name)
+    parts = obj_name.split(".")
+    name_pairs = [(".".join(parts[:i]), ".".join(parts[i:])) for i in range(len(parts), 0, -1)]
+
+    # try each alternative in turn
+    for module_name, local_obj_name in name_pairs:
+        try:
+            module = importlib.import_module(module_name) # may raise ImportError
+            get_obj_from_module(module, local_obj_name) # may raise AttributeError
+            return module, local_obj_name
+        except:
+            pass
+
+    # maybe some of the modules themselves contain errors?
+    for module_name, _local_obj_name in name_pairs:
+        try:
+            importlib.import_module(module_name) # may raise ImportError
+        except ImportError:
+            if not str(sys.exc_info()[1]).startswith("No module named '" + module_name + "'"):
+                raise
+
+    # maybe the requested attribute is missing?
+    for module_name, local_obj_name in name_pairs:
+        try:
+            module = importlib.import_module(module_name) # may raise ImportError
+            get_obj_from_module(module, local_obj_name) # may raise AttributeError
+        except ImportError:
+            pass
+
+    # we are out of luck, but we have no idea why
+    raise ImportError(obj_name)
+
+
+def get_obj_from_module(module: types.ModuleType, obj_name: str) -> Any:
+    """Traverses the object name and returns the last (rightmost) python object."""
+    if obj_name == '':
+        return module
+    obj = module
+    for part in obj_name.split("."):
+        obj = getattr(obj, part)
+    return obj
+
+
+def get_obj_by_name(name: str) -> Any:
+    """Finds the python object with the given name."""
+    module, obj_name = get_module_from_obj_name(name)
+    return get_obj_from_module(module, obj_name)
+
+
+def call_func_by_name(*args, func_name: str = None, **kwargs) -> Any:
+    """Finds the python object with the given name and calls it as a function."""
+    assert func_name is not None
+    # print('func_name: ', func_name) #'training.dataset.ImageFolderDataset'
+    func_obj = get_obj_by_name(func_name) 
+    assert callable(func_obj)
+    return func_obj(*args, **kwargs)
+
+
+def construct_class_by_name(*args, class_name: str = None, **kwargs) -> Any:
+    """Finds the python class with the given name and constructs it with the given arguments."""
+    return call_func_by_name(*args, func_name=class_name, **kwargs)
+
+
+def get_module_dir_by_obj_name(obj_name: str) -> str:
+    """Get the directory path of the module containing the given object name."""
+    module, _ = get_module_from_obj_name(obj_name)
+    return os.path.dirname(inspect.getfile(module))
+
+
+def is_top_level_function(obj: Any) -> bool:
+    """Determine whether the given object is a top-level function, i.e., defined at module scope using 'def'."""
+    return callable(obj) and obj.__name__ in sys.modules[obj.__module__].__dict__
+
+
+def get_top_level_function_name(obj: Any) -> str:
+    """Return the fully-qualified name of a top-level function."""
+    assert is_top_level_function(obj)
+    module = obj.__module__
+    if module == '__main__':
+        module = os.path.splitext(os.path.basename(sys.modules[module].__file__))[0]
+    return module + "." + obj.__name__
+
+
+# File system helpers
+# ------------------------------------------------------------------------------------------
+
+def list_dir_recursively_with_ignore(dir_path: str, ignores: List[str] = None, add_base_to_relative: bool = False) -> List[Tuple[str, str]]:
+    """List all files recursively in a given directory while ignoring given file and directory names.
+    Returns list of tuples containing both absolute and relative paths."""
+    assert os.path.isdir(dir_path)
+    base_name = os.path.basename(os.path.normpath(dir_path))
+
+    if ignores is None:
+        ignores = []
+
+    result = []
+
+    for root, dirs, files in os.walk(dir_path, topdown=True):
+        for ignore_ in ignores:
+            dirs_to_remove = [d for d in dirs if fnmatch.fnmatch(d, ignore_)]
+
+            # dirs need to be edited in-place
+            for d in dirs_to_remove:
+                dirs.remove(d)
+
+            files = [f for f in files if not fnmatch.fnmatch(f, ignore_)]
+
+        absolute_paths = [os.path.join(root, f) for f in files]
+        relative_paths = [os.path.relpath(p, dir_path) for p in absolute_paths]
+
+        if add_base_to_relative:
+            relative_paths = [os.path.join(base_name, p) for p in relative_paths]
+
+        assert len(absolute_paths) == len(relative_paths)
+        result += zip(absolute_paths, relative_paths)
+
+    return result
+
+
+def copy_files_and_create_dirs(files: List[Tuple[str, str]]) -> None:
+    """Takes in a list of tuples of (src, dst) paths and copies files.
+    Will create all necessary directories."""
+    for file in files:
+        target_dir_name = os.path.dirname(file[1])
+
+        # will create all intermediate-level directories
+        if not os.path.exists(target_dir_name):
+            os.makedirs(target_dir_name)
+
+        shutil.copyfile(file[0], file[1])
+
+
+# URL helpers
+# ------------------------------------------------------------------------------------------
+
+def is_url(obj: Any, allow_file_urls: bool = False) -> bool:
+    """Determine whether the given object is a valid URL string."""
+    if not isinstance(obj, str) or not "://" in obj:
+        return False
+    if allow_file_urls and obj.startswith('file://'):
+        return True
+    try:
+        res = requests.compat.urlparse(obj)
+        if not res.scheme or not res.netloc or not "." in res.netloc:
+            return False
+        res = requests.compat.urlparse(requests.compat.urljoin(obj, "/"))
+        if not res.scheme or not res.netloc or not "." in res.netloc:
+            return False
+    except:
+        return False
+    return True
+
+
+def open_url(url: str, cache_dir: str = None, num_attempts: int = 10, verbose: bool = True, return_filename: bool = False, cache: bool = True) -> Any:
+    """Download the given URL and return a binary-mode file object to access the data."""
+    assert num_attempts >= 1
+    assert not (return_filename and (not cache))
+
+    # Doesn't look like an URL scheme so interpret it as a local filename.
+    if not re.match('^[a-z]+://', url):
+        return url if return_filename else open(url, "rb")
+
+    # Handle file URLs.  This code handles unusual file:// patterns that
+    # arise on Windows:
+    #
+    # file:///c:/foo.txt
+    #
+    # which would translate to a local '/c:/foo.txt' filename that's
+    # invalid.  Drop the forward slash for such pathnames.
+    #
+    # If you touch this code path, you should test it on both Linux and
+    # Windows.
+    #
+    # Some internet resources suggest using urllib.request.url2pathname() but
+    # but that converts forward slashes to backslashes and this causes
+    # its own set of problems.
+    if url.startswith('file://'):
+        filename = urllib.parse.urlparse(url).path
+        if re.match(r'^/[a-zA-Z]:', filename):
+            filename = filename[1:]
+        return filename if return_filename else open(filename, "rb")
+
+    assert is_url(url)
+
+    # Lookup from cache.
+    if cache_dir is None:
+        cache_dir = make_cache_dir_path('downloads')
+
+    url_md5 = hashlib.md5(url.encode("utf-8")).hexdigest()
+    if cache:
+        cache_files = glob.glob(os.path.join(cache_dir, url_md5 + "_*"))
+        if len(cache_files) == 1:
+            filename = cache_files[0]
+            return filename if return_filename else open(filename, "rb")
+
+    # Download.
+    url_name = None
+    url_data = None
+    with requests.Session() as session:
+        if verbose:
+            print("Downloading %s ..." % url, end="", flush=True)
+        for attempts_left in reversed(range(num_attempts)):
+            try:
+                with session.get(url) as res:
+                    res.raise_for_status()
+                    if len(res.content) == 0:
+                        raise IOError("No data received")
+
+                    if len(res.content) < 8192:
+                        content_str = res.content.decode("utf-8")
+                        if "download_warning" in res.headers.get("Set-Cookie", ""):
+                            links = [html.unescape(link) for link in content_str.split('"') if "export=download" in link]
+                            if len(links) == 1:
+                                url = requests.compat.urljoin(url, links[0])
+                                raise IOError("Google Drive virus checker nag")
+                        if "Google Drive - Quota exceeded" in content_str:
+                            raise IOError("Google Drive download quota exceeded -- please try again later")
+
+                    match = re.search(r'filename="([^"]*)"', res.headers.get("Content-Disposition", ""))
+                    url_name = match[1] if match else url
+                    url_data = res.content
+                    if verbose:
+                        print(" done")
+                    break
+            except KeyboardInterrupt:
+                raise
+            except:
+                if not attempts_left:
+                    if verbose:
+                        print(" failed")
+                    raise
+                if verbose:
+                    print(".", end="", flush=True)
+
+    # Save to cache.
+    if cache:
+        safe_name = re.sub(r"[^0-9a-zA-Z-._]", "_", url_name)
+        cache_file = os.path.join(cache_dir, url_md5 + "_" + safe_name)
+        temp_file = os.path.join(cache_dir, "tmp_" + uuid.uuid4().hex + "_" + url_md5 + "_" + safe_name)
+        os.makedirs(cache_dir, exist_ok=True)
+        with open(temp_file, "wb") as f:
+            f.write(url_data)
+        os.replace(temp_file, cache_file) # atomic
+        if return_filename:
+            return cache_file
+
+    # Return data as file object.
+    assert not return_filename
+    return io.BytesIO(url_data)

stylegan_human/docs/Dataset.md

@@ -0,0 +1,74 @@
+# SHHQ Dataset
+<img src="../img/preview_samples1.png" width="96%" height="96%">
+
+## Overview
+SHHQ is a dataset with high-quality full-body human images in a resolution of 1024 × 512.
+Since we need to follow a rigorous legal review in our institute, we can not release all of the data at once.
+
+For now, SHHQ-1.0 with 40K images is released! More data will be released in the later versions.
+
+
+## Data Sources
+Images are collected in two main ways: 
+1) From the Internet. 
+We developed a crawler tool with an official API, mainly downloading images from Flickr, Pixabay and Pexels. So you need to meet all the following licenses when using the dataset: CC0, [Pixabay License](https://pixabay.com/service/license/), and [Pexels Licenses](https://www.pexels.com/license/).
+2) From the data providers. 
+We purchased images from databases of individual photographers, modeling agencies and other suppliers.
+Images were reviewed by our legal team prior to purchase to ensure permission for use in research.
+
+### Note: 
+The composition of SHHQ-1.0: 
+
+1) Images obtained from the above sources.
+2) Processed 9991 DeepFashion [[1]](#1) images (retain only full body images).
+3) 1940 African images from the InFashAI [[2]](#2) dataset to increase data diversity.
+
+## Data License
+We are aware of privacy concerns and seriously treat the license and privacy issues. All released data will be ensured under the license of CC0 and free for research use. Also, persons in the dataset are anonymised without additional private or sensitive metadata.
+
+## Agreement
+The SHHQ is available for non-commercial research purposes only. 
+
+You agree not to reproduce, duplicate, copy, sell, trade, resell or exploit any portion of the images and any portion of the derived data for commercial purposes. 
+
+You agree NOT to further copy, publish or distribute any portion of SHHQ to any third party for any purpose. Except, for internal use at a single site within the same organization it is allowed to make copies of the dataset.
+
+Shanghai AI Lab reserves the right to terminate your access to the SHHQ at any time.
+
+## Dataset Preview
+For those interested in our dataset, we provide a preview version with 100 images randomly sampled from SHHQ-1.0: [SHHQ-1.0_samples](https://drive.google.com/file/d/1tnNFfmFtzRbYL3qEnNXQ_ShaN9YV5tI5/view?usp=sharing).
+
+In SHHQ-1.0, we provide aligned raw images along with machine-calculated segmentation masks. Later we are planning to release manually annotated human-parsing version of these 40,000 images. Please stay tuned. 
+ 
+> We also provide script [bg_white.py](../bg_white.py) to whiten the background of the raw image using its segmentation mask.
+
+If you want to access the full SHHQ-1.0, please read the following instructions.
+
+## Model trained using SHHQ-1.0
+
+| Structure | 1024x512 | Metric | Scores |  512x256 | Metric | Scores | 
+| --------- |:----------:| :----------:| :----------:|  :-----: |  :-----: |  :-----: | 
+| StyleGAN1 | to be released | - | - | to be released | - | - |
+| StyleGAN2 | [SHHQ-1.0_sg2_1024.pkl](https://drive.google.com/file/d/1PuvE72xpc69Zq4y58dohuKbG9dFnnjEX/view?usp=sharing) | fid50k_full | 3.56 | [SHHQ-1.0_sg2_512.pkl](https://drive.google.com/file/d/170t2FRWxR8_TG3_y0nVtDBogLPOClnyf/view?usp=sharing) | fid50k_full | 3.68 |
+| StyleGAN3 | to be released | - | - |to be released | - | - |
+
+
+## Download Instructions
+Please download the SHHQ Dataset Release Agreement from [link](./SHHQ_Dataset_Release_Agreement.pdf).
+Read it carefully, complete and sign it appropriately. 
+
+Please send the completed form to Jianglin Fu ([email protected]) and Shikai Li ([email protected]), and cc to Wayne Wu ([email protected]) using institutional email address. The email Subject Title is "SHHQ Dataset Release Agreement". We will verify your request and contact you with the dataset link and password to unzip the image data.
+
+Note:
+
+1. We are currently facing large incoming applications, and we need to carefully verify all the applicants, please be patient, and we will reply to you as soon as possible.
+
+2. The signature in the agreement should be hand-written.
+
+## References
+<a id="1">[1]</a> 
+Liu, Ziwei and Luo, Ping and Qiu, Shi and Wang, Xiaogang and Tang, Xiaoou. DeepFashion: Powering Robust Clothes Recognition and Retrieval with Rich Annotations. CVPR (2016)
+
+<a id="2">[2]</a> 
+Hacheme, Gilles and Sayouti, Noureini. Neural fashion image captioning: Accounting for data diversity. arXiv preprint arXiv:2106.12154 (2021)
+

stylegan_human/edit.py

@@ -0,0 +1,194 @@
+# Copyright (c) SenseTime Research. All rights reserved.
+
+import os
+import sys
+import torch
+import numpy as np
+sys.path.append(".")
+from torch_utils.models import Generator
+import click
+import cv2
+from typing import List, Optional
+import subprocess
+import legacy
+from edit.edit_helper import conv_warper, decoder, encoder_ifg, encoder_ss, encoder_sefa 
+
+
+"""
+Edit generated images with different SOTA methods. 
+ Notes:
+ 1. We provide some latent directions in the folder, you can play around with them.
+ 2. ''upper_length'' and ''bottom_length'' of ''attr_name'' are available for demo.
+ 3. Layers to control and editing strength are set in edit/edit_config.py.
+ 
+Examples:
+
+\b
+# Editing with InterfaceGAN, StyleSpace, and Sefa
+python edit.py --network pretrained_models/stylegan_human_v2_1024.pkl --attr_name upper_length \\
+    --seeds 61531,61570,61571,61610 --outdir outputs/edit_results
+
+
+# Editing using inverted latent code
+python edit.py ---network outputs/pti/checkpoints/model_test.pkl --attr_name upper_length  \\
+    --outdir outputs/edit_results --real True --real_w_path outputs/pti/embeddings/test/PTI/test/0.pt --real_img_path aligned_image/test.png
+
+"""
+
+
+
+@click.command()
+@click.pass_context
+@click.option('--network', 'ckpt_path', help='Network pickle filename', required=True)
+@click.option('--attr_name', help='choose one of the attr: upper_length or bottom_length', type=str, required=True)
+@click.option('--trunc', 'truncation', type=float, help='Truncation psi', default=0.8, show_default=True)
+@click.option('--gen_video', type=bool, default=True, help='If want to generate video')
+@click.option('--combine', type=bool, default=True,  help='If want to combine different editing results in the same frame')
+@click.option('--seeds', type=legacy.num_range, help='List of random seeds')
+@click.option('--outdir', help='Where to save the output images', type=str, required=True, default='outputs/editing', metavar='DIR')
+@click.option('--real', type=bool, help='True for editing real image', default=False)
+@click.option('--real_w_path',  help='Path of latent code for real image')
+@click.option('--real_img_path',  help='Path of real image, this just concat real image with inverted and edited results together')
+
+
+
+def main(
+    ctx: click.Context,
+    ckpt_path: str,
+    attr_name: str,
+    truncation: float,
+    gen_video: bool,
+    combine: bool,
+    seeds: Optional[List[int]],
+    outdir: str,
+    real: str,
+    real_w_path: str,
+    real_img_path: str
+):
+    ## convert pkl to pth
+    # if not os.path.exists(ckpt_path.replace('.pkl','.pth')):
+    legacy.convert(ckpt_path, ckpt_path.replace('.pkl','.pth'), G_only=real)
+    ckpt_path = ckpt_path.replace('.pkl','.pth') 
+    print("start...", flush=True)
+    config = {"latent" : 512, "n_mlp" : 8, "channel_multiplier": 2}
+    generator = Generator(
+            size = 1024,
+            style_dim=config["latent"],
+            n_mlp=config["n_mlp"],
+            channel_multiplier=config["channel_multiplier"]
+        )
+    
+    generator.load_state_dict(torch.load(ckpt_path)['g_ema'])
+    generator.eval().cuda()
+
+    with torch.no_grad():
+        mean_path = os.path.join('edit','mean_latent.pkl')
+        if not os.path.exists(mean_path):
+            mean_n = 3000
+            mean_latent = generator.mean_latent(mean_n).detach()
+            legacy.save_obj(mean_latent, mean_path)
+        else:
+            mean_latent = legacy.load_pkl(mean_path).cuda() 
+        finals = []
+
+        ## -- selected sample seeds -- ##
+        # seeds = [60948,60965,61174,61210,61511,61598,61610] #bottom -> long
+        #         [60941,61064,61103,61313,61531,61570,61571] # bottom -> short
+        #         [60941,60965,61064,61103,6117461210,61531,61570,61571,61610] # upper --> long
+        #         [60948,61313,61511,61598] # upper --> short
+        if real: seeds = [0]
+
+        for t in seeds:
+            if real: # now assume process single real image only
+                if real_img_path:
+                    real_image = cv2.imread(real_img_path)
+                    real_image = cv2.cvtColor(real_image, cv2.COLOR_BGR2RGB)
+                    import torchvision.transforms as transforms
+                    transform = transforms.Compose( # normalize to (-1, 1)
+                        [transforms.ToTensor(),
+                        transforms.Normalize(mean=(.5,.5,.5), std=(.5,.5,.5))]
+                    )
+                    real_image = transform(real_image).unsqueeze(0).cuda()                
+
+                test_input = torch.load(real_w_path)
+                output, _ = generator(test_input, False, truncation=1,input_is_latent=True, real=True)
+
+            else: # generate image from random seeds
+                test_input = torch.from_numpy(np.random.RandomState(t).randn(1, 512)).float().cuda()  # torch.Size([1, 512])
+                output, _ = generator([test_input], False, truncation=truncation, truncation_latent=mean_latent, real=real)
+            
+            # interfacegan
+            style_space, latent, noise = encoder_ifg(generator, test_input, attr_name, truncation, mean_latent,real=real)
+            image1 = decoder(generator, style_space, latent, noise)
+            # stylespace
+            style_space, latent, noise = encoder_ss(generator, test_input, attr_name, truncation, mean_latent,real=real)
+            image2 = decoder(generator, style_space, latent, noise)
+            # sefa
+            latent, noise = encoder_sefa(generator, test_input, attr_name, truncation, mean_latent,real=real)
+            image3, _ = generator([latent], noise=noise, input_is_latent=True)
+            if real_img_path:
+                final = torch.cat((real_image, output, image1, image2, image3), 3)
+            else:
+                final = torch.cat((output, image1, image2, image3), 3)
+
+            # legacy.visual(output, f'{outdir}/{attr_name}_{t:05d}_raw.jpg')
+            # legacy.visual(image1, f'{outdir}/{attr_name}_{t:05d}_ifg.jpg')
+            # legacy.visual(image2, f'{outdir}/{attr_name}_{t:05d}_ss.jpg')
+            # legacy.visual(image3, f'{outdir}/{attr_name}_{t:05d}_sefa.jpg')
+
+            if gen_video:
+                total_step = 90
+                if real:
+                    video_ifg_path = f"{outdir}/video/ifg_{attr_name}_{real_w_path.split('/')[-2]}/"
+                    video_ss_path = f"{outdir}/video/ss_{attr_name}_{real_w_path.split('/')[-2]}/"
+                    video_sefa_path = f"{outdir}/video/ss_{attr_name}_{real_w_path.split('/')[-2]}/"
+                else:
+                    video_ifg_path = f"{outdir}/video/ifg_{attr_name}_{t:05d}/"
+                    video_ss_path = f"{outdir}/video/ss_{attr_name}_{t:05d}/"
+                    video_sefa_path = f"{outdir}/video/ss_{attr_name}_{t:05d}/"
+                video_comb_path = f"{outdir}/video/tmp"    
+
+                if combine:
+                    if not os.path.exists(video_comb_path):
+                        os.makedirs(video_comb_path)
+                else:
+                    if not os.path.exists(video_ifg_path):
+                        os.makedirs(video_ifg_path)
+                    if not os.path.exists(video_ss_path):
+                        os.makedirs(video_ss_path)
+                    if not os.path.exists(video_sefa_path):
+                        os.makedirs(video_sefa_path)
+                for i in range(total_step):
+                    style_space, latent, noise = encoder_ifg(generator, test_input, attr_name, truncation, mean_latent, step=i, total=total_step,real=real)
+                    image1 = decoder(generator, style_space, latent, noise)
+                    style_space, latent, noise = encoder_ss(generator, test_input, attr_name, truncation, mean_latent, step=i, total=total_step,real=real)
+                    image2 = decoder(generator, style_space, latent, noise)
+                    latent, noise = encoder_sefa(generator, test_input, attr_name, truncation, mean_latent, step=i, total=total_step,real=real)
+                    image3, _ = generator([latent], noise=noise, input_is_latent=True)
+                    if combine:
+                        if real_img_path:
+                            comb_img = torch.cat((real_image, output, image1, image2, image3), 3)
+                        else:
+                            comb_img = torch.cat((output, image1, image2, image3), 3)
+                        legacy.visual(comb_img, os.path.join(video_comb_path, f'{i:05d}.jpg'))
+                    else:
+                        legacy.visual(image1, os.path.join(video_ifg_path, f'{i:05d}.jpg'))
+                        legacy.visual(image2, os.path.join(video_ss_path, f'{i:05d}.jpg'))
+                if combine:
+                    cmd=f"ffmpeg -hide_banner -loglevel error -y -r 30 -i {video_comb_path}/%05d.jpg -vcodec libx264 -pix_fmt yuv420p {video_ifg_path.replace('ifg_', '')[:-1] + '.mp4'}"
+                    subprocess.call(cmd, shell=True)
+                else:
+                    cmd=f"ffmpeg -hide_banner -loglevel error -y -r 30 -i {video_ifg_path}/%05d.jpg -vcodec libx264 -pix_fmt yuv420p {video_ifg_path[:-1] + '.mp4'}"
+                    subprocess.call(cmd, shell=True)
+                    cmd=f"ffmpeg -hide_banner -loglevel error -y -r 30 -i {video_ss_path}/%05d.jpg -vcodec libx264 -pix_fmt yuv420p {video_ss_path[:-1] + '.mp4'}"
+                    subprocess.call(cmd, shell=True)
+
+        # interfacegan, stylespace, sefa
+        finals.append(final)
+
+        final = torch.cat(finals, 2)
+        legacy.visual(final, os.path.join(outdir,'final.jpg'))
+
+
+if __name__ == "__main__":
+    main()

stylegan_human/edit/__init__.py

@@ -0,0 +1,3 @@
+# Copyright (c) SenseTime Research. All rights reserved.
+
+# empty

stylegan_human/edit/edit_config.py

@@ -0,0 +1,16 @@
+# Copyright (c) SenseTime Research. All rights reserved.
+
+attr_dict = dict(
+    interface_gan={ # strength
+        'upper_length': [-1], # strength: negative for shorter, positive for longer
+        'bottom_length': [1]
+    },
+    stylespace={ # layer, strength, threshold
+        'upper_length': [5, -5, 0.0028],  # strength: negative for shorter, positive for longer
+        'bottom_length': [3, 5, 0.003] 
+    },
+    sefa={ # layer, strength
+        'upper_length': [[4, 5, 6, 7], 5], #-5 # strength: negative for longer, positive for shorter
+        'bottom_length': [[4, 5, 6, 7], 5]
+    }
+)

stylegan_human/edit/edit_helper.py

@@ -0,0 +1,215 @@
+# Copyright (c) SenseTime Research. All rights reserved.
+
+from legacy import save_obj, load_pkl
+import torch
+from torch.nn import functional as F
+import pandas as pd
+from .edit_config import attr_dict
+import os
+
+def conv_warper(layer, input, style, noise):
+    # the conv should change
+    conv = layer.conv
+    batch, in_channel, height, width = input.shape
+
+    style = style.view(batch, 1, in_channel, 1, 1)
+    weight = conv.scale * conv.weight * style
+
+    if conv.demodulate:
+        demod = torch.rsqrt(weight.pow(2).sum([2, 3, 4]) + 1e-8)
+        weight = weight * demod.view(batch, conv.out_channel, 1, 1, 1)
+
+    weight = weight.view(
+        batch * conv.out_channel, in_channel, conv.kernel_size, conv.kernel_size
+    )
+
+    if conv.upsample:
+        input = input.view(1, batch * in_channel, height, width)
+        weight = weight.view(
+            batch, conv.out_channel, in_channel, conv.kernel_size, conv.kernel_size
+        )
+        weight = weight.transpose(1, 2).reshape(
+            batch * in_channel, conv.out_channel, conv.kernel_size, conv.kernel_size
+        )
+        out = F.conv_transpose2d(input, weight, padding=0, stride=2, groups=batch)
+        _, _, height, width = out.shape
+        out = out.view(batch, conv.out_channel, height, width)
+        out = conv.blur(out)
+
+    elif conv.downsample:
+        input = conv.blur(input)
+        _, _, height, width = input.shape
+        input = input.view(1, batch * in_channel, height, width)
+        out = F.conv2d(input, weight, padding=0, stride=2, groups=batch)
+        _, _, height, width = out.shape
+        out = out.view(batch, conv.out_channel, height, width)
+
+    else:
+        input = input.view(1, batch * in_channel, height, width)
+        out = F.conv2d(input, weight, padding=conv.padding, groups=batch)
+        _, _, height, width = out.shape
+        out = out.view(batch, conv.out_channel, height, width)
+        
+    out = layer.noise(out, noise=noise)
+    out = layer.activate(out)
+    
+    return out
+
+def decoder(G, style_space, latent, noise):
+    # an decoder warper for G
+    out = G.input(latent)
+    out = conv_warper(G.conv1, out, style_space[0], noise[0])
+    skip = G.to_rgb1(out, latent[:, 1])
+
+    i = 1
+    for conv1, conv2, noise1, noise2, to_rgb in zip(
+        G.convs[::2], G.convs[1::2], noise[1::2], noise[2::2], G.to_rgbs
+    ):
+        out = conv_warper(conv1, out, style_space[i], noise=noise1)
+        out = conv_warper(conv2, out, style_space[i+1], noise=noise2)
+        skip = to_rgb(out, latent[:, i + 2], skip)
+        i += 2
+    image = skip
+
+    return image
+
+def encoder_ifg(G, noise, attr_name, truncation=1, truncation_latent=None, 
+                  latent_dir='latent_direction/ss/',
+                  step=0, total=0, real=False):
+    if not real:
+        styles = [noise]
+        styles = [G.style(s) for s in styles]
+    style_space = []
+    
+    if truncation<1:
+        if not real: 
+            style_t = []
+            for style in styles:
+                style_t.append(truncation_latent + truncation * (style - truncation_latent))
+            styles = style_t
+        else: # styles are latent (tensor: 1,18,512), for real PTI output
+            truncation_latent = truncation_latent.repeat(18,1).unsqueeze(0) # (1,512) --> (1,18,512)
+            styles = torch.add(truncation_latent,torch.mul(torch.sub(noise,truncation_latent),truncation))
+
+
+    noise = [getattr(G.noises, 'noise_{}'.format(i)) for i in range(G.num_layers)]
+    if not real:
+        inject_index = G.n_latent
+        latent = styles[0].unsqueeze(1).repeat(1, inject_index, 1)
+    else: latent=styles
+
+    style_space.append(G.conv1.conv.modulation(latent[:, 0]))
+    i = 1
+    for conv1, conv2, noise1, noise2, to_rgb in zip(
+        G.convs[::2], G.convs[1::2], noise[1::2], noise[2::2], G.to_rgbs
+    ):
+        style_space.append(conv1.conv.modulation(latent[:, i]))
+        style_space.append(conv2.conv.modulation(latent[:, i+1]))
+        i += 2
+
+    # get layer, strength by dict
+    strength = attr_dict['interface_gan'][attr_name][0]
+
+    if step != 0 and total != 0:
+        strength = step / total * strength
+    for i in range(15):
+        style_vect = load_pkl(os.path.join(latent_dir, '{}/style_vect_mean_{}.pkl'.format(attr_name, i)))
+        style_vect = torch.from_numpy(style_vect).to(latent.device).float()
+        style_space[i] += style_vect * strength
+        
+    return style_space, latent, noise
+
+def encoder_ss(G, noise, attr_name, truncation=1, truncation_latent=None, 
+               statics_dir="latent_direction/ss_statics",
+               latent_dir="latent_direction/ss/",
+               step=0, total=0,real=False):
+    if not real:
+        styles = [noise]
+        styles = [G.style(s) for s in styles]
+    style_space = []
+
+    if truncation<1:
+        if not real:
+            style_t = []
+            for style in styles:
+                style_t.append(
+                    truncation_latent + truncation * (style - truncation_latent)
+                )
+            styles = style_t
+        else: # styles are latent (tensor: 1,18,512), for real PTI output
+            truncation_latent = truncation_latent.repeat(18,1).unsqueeze(0) # (1,512) --> (1,18,512)
+            styles = torch.add(truncation_latent,torch.mul(torch.sub(noise,truncation_latent),truncation))
+
+    noise = [getattr(G.noises, 'noise_{}'.format(i)) for i in range(G.num_layers)]
+    
+    if not real:
+        inject_index = G.n_latent
+        latent = styles[0].unsqueeze(1).repeat(1, inject_index, 1)
+    else: latent = styles
+
+    style_space.append(G.conv1.conv.modulation(latent[:, 0]))
+    i = 1
+    for conv1, conv2, noise1, noise2, to_rgb in zip(
+        G.convs[::2], G.convs[1::2], noise[1::2], noise[2::2], G.to_rgbs
+    ):
+        style_space.append(conv1.conv.modulation(latent[:, i]))
+        style_space.append(conv2.conv.modulation(latent[:, i+1]))
+        i += 2
+    # get threshold, layer, strength by dict
+    layer, strength, threshold = attr_dict['stylespace'][attr_name] 
+
+    statis_dir = os.path.join(statics_dir, "{}_statis/{}".format(attr_name, layer))
+    statis_csv_path = os.path.join(statis_dir, "statis.csv")
+    statis_df = pd.read_csv(statis_csv_path)
+    statis_df = statis_df.sort_values(by='channel', ascending=True)
+    ch_mask = statis_df['strength'].values
+    ch_mask = torch.from_numpy(ch_mask).to(latent.device).float()
+    ch_mask = (ch_mask.abs()>threshold).float()
+    style_vect = load_pkl(os.path.join(latent_dir, '{}/style_vect_mean_{}.pkl'.format(attr_name, layer)))
+    style_vect = torch.from_numpy(style_vect).to(latent.device).float()
+
+    style_vect = style_vect * ch_mask
+
+    if step != 0 and total != 0:
+        strength = step / total * strength
+
+    style_space[layer] += style_vect * strength
+        
+    return style_space, latent, noise
+
+def encoder_sefa(G, noise, attr_name, truncation=1, truncation_latent=None, 
+                  latent_dir='latent_direction/sefa/',
+                  step=0, total=0, real=False):
+    if not real: 
+        styles = [noise]
+        styles = [G.style(s) for s in styles]
+    
+    if truncation<1:
+        if not real:
+            style_t = []
+            for style in styles:
+                style_t.append(
+                    truncation_latent + truncation * (style - truncation_latent)
+                )
+            styles = style_t
+        else:
+            truncation_latent = truncation_latent.repeat(18,1).unsqueeze(0) # (1,512) --> (1,18,512)
+            styles = torch.add(truncation_latent,torch.mul(torch.sub(noise,truncation_latent),truncation))
+
+
+    noise = [getattr(G.noises, 'noise_{}'.format(i)) for i in range(G.num_layers)]
+    if not real:
+        inject_index = G.n_latent
+        latent = styles[0].unsqueeze(1).repeat(1, inject_index, 1)
+    else: latent = styles
+    
+    layer, strength = attr_dict['sefa'][attr_name] 
+
+    sefa_vect = torch.load(os.path.join(latent_dir, '{}.pt'.format(attr_name))).to(latent.device).float()
+    if step != 0 and total != 0:
+        strength = step / total * strength
+    for l in layer:
+        latent[:, l, :] += (sefa_vect * strength * 2)
+        
+
+    return latent, noise

stylegan_human/environment.yml

@@ -0,0 +1,30 @@
+name: stylehuman
+channels:
+  - pytorch
+  - nvidia
+dependencies:
+  - python == 3.8
+  - pip
+  - numpy>=1.20
+  - click>=8.0
+  - pillow=8.3.1
+  - scipy=1.7.1
+  - pytorch=1.9.1
+  - cudatoolkit=11.1
+  - requests=2.26.0
+  - tqdm=4.62.2
+  - ninja=1.10.2
+  - matplotlib=3.4.2
+  - imageio=2.9.0
+  - pip:
+    - imgui==1.3.0
+    - glfw==2.2.0
+    - pyopengl==3.1.5
+    - imageio-ffmpeg==0.4.3
+    - lpips==0.1.4
+    - pyspng
+    - dlib
+    - opencv-python
+    - pandas
+    - moviepy
+    - imutils