app.py

import glob
import os
from copy import deepcopy

import gradio as gr
import numpy as np
import PIL
import spaces
import torch
import yaml
from huggingface_hub import hf_hub_download
from PIL import Image
from safetensors.torch import load_file
from torchvision.transforms import ToPILImage, ToTensor
from transformers import AutoModelForImageSegmentation
from utils import extract_object, get_model_from_config, resize_and_center_crop

huggingface_token = os.getenv("HUGGINGFACE_TOKEN")

ASPECT_RATIOS = {
    str(512 / 2048): (512, 2048),
    str(1024 / 1024): (1024, 1024),
    str(2048 / 512): (2048, 512),
    str(896 / 1152): (896, 1152),
    str(1152 / 896): (1152, 896),
    str(512 / 1920): (512, 1920),
    str(640 / 1536): (640, 1536),
    str(768 / 1280): (768, 1280),
    str(1280 / 768): (1280, 768),
    str(1536 / 640): (1536, 640),
    str(1920 / 512): (1920, 512),
}

MODEL_PATH = hf_hub_download("jasperai/LBM_relighting", "model.safetensors", token=huggingface_token)
CONFIG_PATH = hf_hub_download("jasperai/LBM_relighting", "config.yaml", token=huggingface_token)

with open(CONFIG_PATH, "r") as f:
    config = yaml.safe_load(f)
model = get_model_from_config(**config)
sd = load_file(MODEL_PATH)
model.load_state_dict(sd, strict=True)
model.to("cuda").to(torch.bfloat16)
birefnet = AutoModelForImageSegmentation.from_pretrained("ZhengPeng7/BiRefNet", trust_remote_code=True).cuda()
image_size = (1024, 1024)

@spaces.GPU
def evaluate(fg_image: PIL.Image.Image, bg_image: PIL.Image.Image, num_sampling_steps: int = 4):
    ori_h_bg, ori_w_bg = fg_image.size
    ar_bg = ori_h_bg / ori_w_bg
    closest_ar_bg = min(ASPECT_RATIOS, key=lambda x: abs(float(x) - ar_bg))
    dimensions_bg = ASPECT_RATIOS[closest_ar_bg]

    _, fg_mask = extract_object(birefnet, deepcopy(fg_image))

    fg_image = resize_and_center_crop(fg_image, dimensions_bg[0], dimensions_bg[1])
    fg_mask = resize_and_center_crop(fg_mask, dimensions_bg[0], dimensions_bg[1])
    bg_image = resize_and_center_crop(bg_image, dimensions_bg[0], dimensions_bg[1])

    img_pasted = Image.composite(fg_image, bg_image, fg_mask)
    img_pasted_tensor = ToTensor()(img_pasted).unsqueeze(0) * 2 - 1
    batch = {"source_image": img_pasted_tensor.cuda().to(torch.bfloat16)}

    z_source = model.vae.encode(batch[model.source_key])
    output_image = model.sample(z=z_source, num_steps=num_sampling_steps, conditioner_inputs=batch, max_samples=1).clamp(-1, 1)
    output_image = (output_image[0].float().cpu() + 1) / 2
    output_image = ToPILImage()(output_image)
    output_image = Image.composite(output_image, bg_image, fg_mask)
    output_image.resize((ori_h_bg, ori_w_bg))

    return (np.array(img_pasted), np.array(output_image))

with gr.Blocks() as app:
    gr.HTML("""
    <style>
    body::before {
        content: "";
        display: block;
        height: 320px;
        background-color: var(--body-background-fill);
    }
    button[aria-label="Fullscreen"], button[aria-label="Fullscreen"]:hover {
        display: none !important;
        visibility: hidden !important;
        opacity: 0 !important;
        pointer-events: none !important;
    }
    button[aria-label="Share"], button[aria-label="Share"]:hover {
        display: none !important;
    }
    button[aria-label="Download"] {
        transform: scale(3);
        transform-origin: top right;
        margin: 0 !important;
        padding: 6px !important;
    }
    </style>
    """)

    gr.Markdown("# Ndrysho Sfondin")
    gr.Markdown("Zëvendëso sfondin e fotove me rindriçim të avancuar nga inteligjenca artificiale.")

    with gr.Row():
        with gr.Column():
            with gr.Row():
                fg_image = gr.Image(type="pil", label="Imazhi Kryesor", image_mode="RGB", height=360)
                bg_image = gr.Image(type="pil", label="Sfondi i Ri", image_mode="RGB", height=360)

            with gr.Row():
                submit_button = gr.Button("Rindriço")
            with gr.Row():
                num_inference_steps = gr.Slider(minimum=1, maximum=4, value=4, step=1, visible=False)

            bg_gallery = gr.Gallery(object_fit="contain", visible=False)

        with gr.Column():
            output_slider = gr.ImageSlider(label="Para / Pas", type="numpy")
            output_slider.upload(fn=evaluate, inputs=[fg_image, bg_image, num_inference_steps], outputs=[output_slider])

    submit_button.click(evaluate, inputs=[fg_image, bg_image, num_inference_steps], outputs=[output_slider], show_progress="full", show_api=False)

    def bg_gallery_selected(gal, evt: gr.SelectData):
        return gal[evt.index][0]

    bg_gallery.select(bg_gallery_selected, inputs=bg_gallery, outputs=bg_image)

if __name__ == "__main__":
    app.launch(share=True)