app.py

import os
import cv2
import numpy as np
import torch
import gradio as gr
import spaces

from glob import glob
from typing import Tuple, Optional

from PIL import Image
from gradio_imageslider import ImageSlider
from transformers import AutoModelForImageSegmentation
from torchvision import transforms

import requests
from io import BytesIO
import zipfile
import random

torch.set_float32_matmul_precision('high')
torch.jit.script = lambda f: f

device = "cuda" if torch.cuda.is_available() else "cpu"

### 이미지 후처리 함수들 ###
def refine_foreground(image, mask, r=90):
    if mask.size != image.size:
        mask = mask.resize(image.size)
    image_np = np.array(image) / 255.0
    mask_np = np.array(mask) / 255.0
    estimated_foreground = FB_blur_fusion_foreground_estimator_2(image_np, mask_np, r=r)
    image_masked = Image.fromarray((estimated_foreground * 255.0).astype(np.uint8))
    return image_masked

def FB_blur_fusion_foreground_estimator_2(image, alpha, r=90):
    alpha = alpha[:, :, None]
    F, blur_B = FB_blur_fusion_foreground_estimator(image, image, image, alpha, r)
    return FB_blur_fusion_foreground_estimator(image, F, blur_B, alpha, r=6)[0]

def FB_blur_fusion_foreground_estimator(image, F, B, alpha, r=90):
    if isinstance(image, Image.Image):
        image = np.array(image) / 255.0
    blurred_alpha = cv2.blur(alpha, (r, r))[:, :, None]
    blurred_FA = cv2.blur(F * alpha, (r, r))
    blurred_F = blurred_FA / (blurred_alpha + 1e-5)
    blurred_B1A = cv2.blur(B * (1 - alpha), (r, r))
    blurred_B = blurred_B1A / ((1 - blurred_alpha) + 1e-5)
    F = blurred_F + alpha * (image - alpha * blurred_F - (1 - alpha) * blurred_B)
    F = np.clip(F, 0, 1)
    return F, blurred_B

class ImagePreprocessor():
    def __init__(self, resolution: Tuple[int, int] = (1024, 1024)) -> None:
        self.transform_image = transforms.Compose([
            transforms.Resize(resolution),
            transforms.ToTensor(),
            transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225]),
        ])
    def proc(self, image: Image.Image) -> torch.Tensor:
        image = self.transform_image(image)
        return image

usage_to_weights_file = {
    'General': 'BiRefNet',
    'General-HR': 'BiRefNet_HR',
    'General-Lite': 'BiRefNet_lite',
    'General-Lite-2K': 'BiRefNet_lite-2K',
    'Matting': 'BiRefNet-matting',
    'Portrait': 'BiRefNet-portrait',
    'DIS': 'BiRefNet-DIS5K',
    'HRSOD': 'BiRefNet-HRSOD',
    'COD': 'BiRefNet-COD',
    'DIS-TR_TEs': 'BiRefNet-DIS5K-TR_TEs',
    'General-legacy': 'BiRefNet-legacy'
}

# 초기 모델 로딩 (기본: General)
birefnet = AutoModelForImageSegmentation.from_pretrained(
    '/'.join(('zhengpeng7', usage_to_weights_file['General'])), 
    trust_remote_code=True
)
birefnet.to(device)
birefnet.eval(); birefnet.half()

@spaces.GPU
def predict(images, resolution, weights_file):
    assert images is not None, 'Images cannot be None.'
    global birefnet
    # 선택된 가중치로 모델 재로딩
    _weights_file = '/'.join(('zhengpeng7', usage_to_weights_file[weights_file] if weights_file is not None else usage_to_weights_file['General']))
    print('Using weights: {}.'.format(_weights_file))
    birefnet = AutoModelForImageSegmentation.from_pretrained(_weights_file, trust_remote_code=True)
    birefnet.to(device)
    birefnet.eval(); birefnet.half()

    try:
        resolution_list = [int(int(reso)//32*32) for reso in resolution.strip().split('x')]
    except:
        if weights_file == 'General-HR':
            resolution_list = [2048, 2048]
        elif weights_file == 'General-Lite-2K':
            resolution_list = [2560, 1440]
        else:
            resolution_list = [1024, 1024]
        print('Invalid resolution input. Automatically changed to default.')

    # 이미지가 단일 객체인지, 리스트(배치)인지 확인
    if isinstance(images, list):
        tab_is_batch = True
    else:
        images = [images]
        tab_is_batch = False

    save_paths = []
    save_dir = 'preds-BiRefNet'
    if tab_is_batch and not os.path.exists(save_dir):
        os.makedirs(save_dir)
    
    outputs = []
    for idx, image_src in enumerate(images):
        if isinstance(image_src, str):
            if os.path.isfile(image_src):
                image_ori = Image.open(image_src)
            else:
                response = requests.get(image_src)
                image_data = BytesIO(response.content)
                image_ori = Image.open(image_data)
        else:
            if isinstance(image_src, np.ndarray):
                image_ori = Image.fromarray(image_src)
            else:
                image_ori = image_src.convert('RGB')
        image = image_ori.convert('RGB')
        preprocessor = ImagePreprocessor(resolution=tuple(resolution_list))
        image_proc = preprocessor.proc(image).unsqueeze(0)
        with torch.no_grad():
            preds = birefnet(image_proc.to(device).half())[-1].sigmoid().cpu()
        pred = preds[0].squeeze()
        pred_pil = transforms.ToPILImage()(pred)
        image_masked = refine_foreground(image, pred_pil)
        image_masked.putalpha(pred_pil.resize(image.size))
        torch.cuda.empty_cache()
        if tab_is_batch:
            file_path = os.path.join(save_dir, "{}.png".format(
                os.path.splitext(os.path.basename(image_src))[0] if isinstance(image_src, str) else f"img_{idx}"
            ))
            image_masked.save(file_path)
            save_paths.append(file_path)
            outputs.append(image_masked)
        else:
            outputs = [image_masked, image_ori]
    
    if tab_is_batch:
        zip_file_path = os.path.join(save_dir, "{}.zip".format(save_dir))
        with zipfile.ZipFile(zip_file_path, 'w') as zipf:
            for file in save_paths:
                zipf.write(file, os.path.basename(file))
        return save_paths, zip_file_path
    else:
        # 반환값을 리스트 형태로 만들어 ImageSlider에서 표시되도록 함.
        return outputs

# 예제 데이터 (이미지, URL, 배치)
examples_image = [[path, "1024x1024", "General"] for path in glob('examples/*')]
examples_text = [[url, "1024x1024", "General"] for url in ["https://hips.hearstapps.com/hmg-prod/images/gettyimages-1229892983-square.jpg"]]
examples_batch = [[file, "1024x1024", "General"] for file in glob('examples/*')]

descriptions = (
    "Upload a picture, our model will extract a highly accurate segmentation of the subject in it.\n"
    "The resolution used in our training was `1024x1024`, which is suggested for good results! "
    "`2048x2048` is suggested for BiRefNet_HR.\n"
    "Our codes can be found at https://github.com/ZhengPeng7/BiRefNet.\n"
    "We also maintain the HF model of BiRefNet at https://huggingface.co/ZhengPeng7/BiRefNet for easier access."
)

# 개선된 UI 스타일 (배경, 컨테이너, 좌측 사이드바, 버튼 애니메이션 등)
css = """
body {
    background: linear-gradient(135deg, #667eea, #764ba2);
    font-family: 'Helvetica Neue', Helvetica, Arial, sans-serif;
    color: #333;
    margin: 0;
    padding: 0;
}
.gradio-container {
    background: rgba(255, 255, 255, 0.95);
    border-radius: 15px;
    padding: 30px 40px;
    box-shadow: 0 8px 30px rgba(0, 0, 0, 0.3);
    margin: 40px auto;
    max-width: 1200px;
}
.gradio-container h1 {
    color: #333;
    text-shadow: 1px 1px 2px rgba(0, 0, 0, 0.2);
}
.fillable { 
    width: 95% !important; 
    max-width: unset !important;
}
#examples_container {
    margin: auto;
    width: 90%;
}
#examples_row {
    justify-content: center;
}
.sidebar {
    background: rgba(255, 255, 255, 0.98);
    border-radius: 10px;
    padding: 20px;
    box-shadow: 0 4px 15px rgba(0, 0, 0, 0.2);
}
button, .btn {
    background: linear-gradient(90deg, #ff8a00, #e52e71);
    border: none;
    color: #fff;
    padding: 12px 24px;
    text-transform: uppercase;
    font-weight: bold;
    letter-spacing: 1px;
    border-radius: 5px;
    cursor: pointer;
    transition: transform 0.2s ease-in-out;
}
button:hover, .btn:hover {
    transform: scale(1.05);
}
"""

title = """
<h1 align="center" style="margin-bottom: 0.2em;">BiRefNet Demo for Subject Extraction</h1>
<p align="center" style="font-size:1.1em; color:#555;">
    Upload an image or provide an image URL to extract the subject with high-precision segmentation.
</p>
"""

with gr.Blocks(css=css, title="BiRefNet Demo") as demo:
    gr.Markdown(title)
    with gr.Tabs():
        with gr.Tab("Image"):
            with gr.Row():
                with gr.Column(scale=1):
                    image_input = gr.Image(type='pil', label='Upload an Image')
                    resolution_input = gr.Textbox(lines=1, placeholder="e.g., 1024x1024", label="Resolution")
                    weights_radio = gr.Radio(list(usage_to_weights_file.keys()), value="General", label="Weights")
                    predict_btn = gr.Button("Predict")
                with gr.Column(scale=2):
                    output_slider = ImageSlider(label="BiRefNet's Prediction", type="pil")
            gr.Examples(examples=examples_image, inputs=[image_input, resolution_input, weights_radio], label="Examples")
        with gr.Tab("Text"):
            with gr.Row():
                with gr.Column(scale=1):
                    image_url = gr.Textbox(label="Paste an Image URL")
                    resolution_input_text = gr.Textbox(lines=1, placeholder="e.g., 1024x1024", label="Resolution")
                    weights_radio_text = gr.Radio(list(usage_to_weights_file.keys()), value="General", label="Weights")
                    predict_btn_text = gr.Button("Predict")
                with gr.Column(scale=2):
                    output_slider_text = ImageSlider(label="BiRefNet's Prediction", type="pil")
            gr.Examples(examples=examples_text, inputs=[image_url, resolution_input_text, weights_radio_text], label="Examples")
        with gr.Tab("Batch"):
            with gr.Row():
                with gr.Column(scale=1):
                    file_input = gr.File(label="Upload Multiple Images", type="filepath", file_count="multiple")
                    resolution_input_batch = gr.Textbox(lines=1, placeholder="e.g., 1024x1024", label="Resolution")
                    weights_radio_batch = gr.Radio(list(usage_to_weights_file.keys()), value="General", label="Weights")
                    predict_btn_batch = gr.Button("Predict")
                with gr.Column(scale=2):
                    output_gallery = gr.Gallery(label="BiRefNet's Predictions").style(grid=[3], height="auto")
                    zip_output = gr.File(label="Download Masked Images")
            gr.Examples(examples=examples_batch, inputs=[file_input, resolution_input_batch, weights_radio_batch], label="Examples")
    with gr.Row():
        gr.Markdown("<p align='center'>Model by <a href='https://huggingface.co/ZhengPeng7/BiRefNet'>ZhengPeng7/BiRefNet</a></p>")

    # 각 탭의 Predict 버튼과 predict 함수 연결
    predict_btn.click(
        fn=predict,
        inputs=[image_input, resolution_input, weights_radio],
        outputs=output_slider
    )
    predict_btn_text.click(
        fn=predict,
        inputs=[image_url, resolution_input_text, weights_radio_text],
        outputs=output_slider_text
    )
    predict_btn_batch.click(
        fn=predict,
        inputs=[file_input, resolution_input_batch, weights_radio_batch],
        outputs=[output_gallery, zip_output]
    )

if __name__ == "__main__":
    demo.launch(share=False, debug=True)