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

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


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

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

### image_proc.py
def refine_foreground(image, mask, r=90):
    if mask.size != image.size:
        mask = mask.resize(image.size)
    image = np.array(image) / 255.0
    mask = np.array(mask) / 255.0
    estimated_foreground = FB_blur_fusion_foreground_estimator_2(image, mask, 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):
    # Thanks to the source: https://github.com/Photoroom/fast-foreground-estimation
    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-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'
}

birefnet = AutoModelForImageSegmentation.from_pretrained('zhengpeng7/BiRefNet_lite', trust_remote_code=True)
birefnet.to(device)
birefnet.eval()


@spaces.GPU
def predict(images):
    assert (images is not None), 'AssertionError: images cannot be None.'

    global birefnet
    # Load BiRefNet with chosen weights
    _weights_file = 'zhengpeng7/BiRefNet_lite'
    print('Using weights: {}.'.format(_weights_file))
    birefnet = AutoModelForImageSegmentation.from_pretrained(_weights_file, trust_remote_code=True)
    birefnet.to(device)
    birefnet.eval()

    #try:
    #    resolution = [int(int(reso)//32*32) for reso in resolution.strip().split('x')]
    #except:
    #    resolution = (1024, 1024) if weights_file not in ['General-Lite-2K'] else (2560, 1440)
    #    print('Invalid resolution input. Automatically changed to 1024x1024 or 2K.')

    if isinstance(images, list):
        # For tab_batch
        save_paths = []
        save_dir = 'preds-BiRefNet'
        if not os.path.exists(save_dir):
            os.makedirs(save_dir)
        tab_is_batch = True
    else:
        images = [images]
        tab_is_batch = False

    for idx_image, 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:
            image_ori = Image.fromarray(image_src)

        image = image_ori.convert('RGB')
        # Preprocess the image
        image_preprocessor = ImagePreprocessor() #(resolution=tuple(resolution))
        image_proc = image_preprocessor.proc(image)
        image_proc = image_proc.unsqueeze(0)

        # Prediction
        with torch.no_grad():
            preds = birefnet(image_proc.to(device))[-1].sigmoid().cpu()
        pred = preds[0].squeeze()

        # Show Results
        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:
            save_file_path = os.path.join(save_dir, "{}.png".format(os.path.splitext(os.path.basename(image_src))[0]))
            image_masked.save(save_file_path)
            save_paths.append(save_file_path)

    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:
        return (image_masked, image_ori)


examples = [[_] for _ in glob('examples/*')][:]
# Add the option of resolution in a text box.
for idx_example, example in enumerate(examples):
    examples[idx_example].append('1024x1024')
examples.append(examples[-1].copy())
examples[-1][1] = '512x512'

examples_url = [
    ['https://hips.hearstapps.com/hmg-prod/images/gettyimages-1229892983-square.jpg'],
]
for idx_example_url, example_url in enumerate(examples_url):
    examples_url[idx_example_url].append('1024x1024')

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`, thus the suggested resolution to obtain good results!\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.')

tab_image = gr.Interface(
    fn=predict,
    inputs=[
        gr.Image(label='Upload an image'),
        #gr.Textbox(lines=1, placeholder="Type the resolution (`WxH`) you want, e.g., `1024x1024`.", label="Resolution"),
        #gr.Radio(list(usage_to_weights_file.keys()), value='General', label="Weights", info="Choose the weights you want.")
    ],
    outputs=ImageSlider(label="BiRefNet's prediction", type="pil"),
    #examples=examples,
    api_name="image",
    description=descriptions,
)

demo = gr.TabbedInterface(
    [tab_image],
    ['image'],
    title="BiRefNet demo for subject extraction.",
)

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