main_code_script.py

# Install necessary libraries (in your requirements.txt)
# pillow opencv-python transformers mediapipe diffusers accelerate transformers
# Example install command: pip install pillow opencv-python transformers mediapipe diffusers accelerate transformers
from PIL import Image
import cv2
import mediapipe as mp
import numpy as np
from transformers import pipeline, CLIPImageProcessor, CLIPVisionModelWithProjection, CLIPTextModel, CLIPTextModelWithProjection, AutoTokenizer
from diffusers import StableDiffusionXLInpaintPipeline, DDPMScheduler, AutoencoderKL
import torch
import os
from torchvision import transforms
from typing import List
# from utils_mask import get_mask_location
# from preprocess.humanparsing.run_parsing import Parsing
# from preprocess.openpose.run_openpose import OpenPose
from detectron2.data.detection_utils import convert_PIL_to_numpy, _apply_exif_orientation
from torchvision.transforms.functional import to_pil_image
import apply_net

def pil_to_binary_mask(pil_image, threshold=0):
    np_image = np.array(pil_image)
    grayscale_image = Image.fromarray(np_image).convert("L")
    binary_mask = np.array(grayscale_image) > threshold
    mask = np.zeros(binary_mask.shape, dtype=np.uint8)
    for i in range(binary_mask.shape[0]):
        for j in range(binary_mask.shape[1]):
            if binary_mask[i, j] == True:
                mask[i, j] = 1
    mask = (mask * 255).astype(np.uint8)
    output_mask = Image.fromarray(mask)
    return output_mask # [cite: 60, 61]
base_path = 'yisol/IDM-VTON'
# example_path = os.path.join(os.path.dirname(__file__), 'example')
unet = UNet2DConditionModel.from_pretrained(
    base_path,
    subfolder="unet",
    torch_dtype=torch.float16,
)
unet.requires_grad_(False) # [cite: 61]
tokenizer_one = AutoTokenizer.from_pretrained(
    base_path,
    subfolder="tokenizer",
    revision=None,
    use_fast=False,
) # [cite: 61, 62]
tokenizer_two = AutoTokenizer.from_pretrained(
    base_path,
    subfolder="tokenizer_2",
    revision=None,
    use_fast=False,
) # [cite: 62]
noise_scheduler = DDPMScheduler.from_pretrained(base_path, subfolder="scheduler") # [cite: 62]
text_encoder_one = CLIPTextModel.from_pretrained(
    base_path,
    subfolder="text_encoder",
    torch_dtype=torch.float16,
) # [cite: 62]
text_encoder_two = CLIPTextModelWithProjection.from_pretrained(
    base_path,
    subfolder="text_encoder_2",
    torch_dtype=torch.float16,
) # [cite: 62]
image_encoder = CLIPVisionModelWithProjection.from_pretrained(
    base_path,
    subfolder="image_encoder",
    torch_dtype=torch.float16,
) # [cite: 62, 63]
vae = AutoencoderKL.from_pretrained(base_path,
                                    subfolder="vae",
                                    torch_dtype=torch.float16,
) # [cite: 63]
# "stabilityai/stable-diffusion-xl-base-1.0",
UNet_Encoder = UNet2DConditionModel_ref.from_pretrained(
    base_path,
    subfolder="unet_encoder",
    torch_dtype=torch.float16,
) # [cite: 63]
# parsing_model = Parsing(0)
# openpose_model = OpenPose(0)
UNet_Encoder.requires_grad_(False) # [cite: 63]
image_encoder.requires_grad_(False) # [cite: 63]
vae.requires_grad_(False) # [cite: 63]
unet.requires_grad_(False) # [cite: 63]
text_encoder_one.requires_grad_(False) # [cite: 63]
text_encoder_two.requires_grad_(False) # [cite: 63]
tensor_transfrom = transforms.Compose(
    [
        transforms.ToTensor(),
        transforms.Normalize(,),
    ]
) # [cite: 63, 64]
pipe = TryonPipeline.from_pretrained(
    base_path,
    unet=unet,
    vae=vae,
    feature_extractor=CLIPImageProcessor(),
    text_encoder=text_encoder_one,
    text_encoder_2=text_encoder_two,
    tokenizer=tokenizer_one,
    tokenizer_2=tokenizer_two,
    scheduler=noise_scheduler,
    image_encoder=image_encoder,
    torch_dtype=torch.float16,
) # [cite: 64, 65]
pipe.unet_encoder = UNet_Encoder # [cite: 65]
@spaces.GPU
def start_tryon(dict, garm_img, garment_des, is_checked, is_checked_crop, denoise_steps, seed):
    device = "cuda"
    # openpose_model.preprocessor.body_estimation.model.to(device)
    pipe.to(device)
    pipe.unet_encoder.to(device)
    garm_img = garm_img.convert("RGB").resize((768, 1024))
    human_img_orig = dict["background"].convert("RGB")
    if is_checked_crop:
        width, height = human_img_orig.size
        target_width = int(min(width, height * (3 / 4)))
        target_height = int(min(height, width * (4 / 3)))
        left = (width - target_width) / 2
        top = (height - target_height) / 2
        right = (width + target_width) / 2
        bottom = (height + target_height) / 2
        cropped_img = human_img_orig.crop((left, top, right, bottom))
        crop_size = cropped_img.size
        human_img = cropped_img.resize((384, 512))  # Reduced size for efficiency
    else:
        human_img = human_img_orig.resize((384, 512))  # Reduced size for efficiency
    if is_checked:
        # keypoints = openpose_model(human_img.resize((384, 512)))
        # model_parse, _ = parsing_model(human_img.resize((384, 512)))
        # mask, mask_gray = get_mask_location('hd', "upper_body", model_parse, keypoints)
        # mask = mask.resize((768, 1024))
        # Placeholder for mask generation (replace with your mask logic)
        mask = Image.new('L', (768, 1024), color='white')  # Example: a white mask
        mask_gray = Image.new('RGB', (768, 1024), color='gray') # Example: a gray image
    else:
        mask = pil_to_binary_mask(dict['layers'][0].convert("RGB").resize((768, 1024)))
    mask_gray = (1 - transforms.ToTensor()(mask)) * tensor_transfrom(human_img)
    mask_gray = to_pil_image((mask_gray + 1.0) / 2.0)
    human_img_arg = _apply_exif_orientation(human_img.resize((384, 512)))
    human_img_arg = convert_PIL_to_numpy(human_img_arg, format="BGR")
    args = apply_net.create_argument_parser().parse_args(
        ('show', './configs/densepose_rcnn_R_50_FPN_s1x.yaml', './ckpt/densepose/model_final_162be9.pkl', 'dp_segm',
         '-v', '--opts', 'MODEL.DEVICE', 'cuda'))
    # verbosity = getattr(args, "verbosity", None)
    pose_img = args.func(args, human_img_arg)
    pose_img = pose_img[:, :, ::-1]
    pose_img = Image.fromarray(pose_img).resize((768, 1024))
    with torch.no_grad():
        # Extract the images
        with torch.cuda.amp.autocast():
            with torch.no_grad():
                prompt = "model is wearing " + garment_des
                negative_prompt = "monochrome, lowres, bad anatomy, worst quality, low quality"
                with torch.inference_mode():
                    (
                        prompt_embeds,
                        negative_prompt_embeds,
                        pooled_prompt_embeds,
                        negative_pooled_prompt_embeds,
                    ) = pipe.encode_prompt(
                        prompt,
                        num_images_per_prompt=1,
                        do_classifier_free_guidance=True,
                        negative_prompt=negative_prompt,
                    )
                prompt = "a photo of " + garment_des
                negative_prompt = "monochrome, lowres, bad anatomy, worst quality, low quality"
                if not isinstance(prompt, List):
                    prompt = [prompt] * 1
                if not isinstance(negative_prompt, List):
                    negative_prompt = [negative_prompt] * 1
                with torch.inference_mode():
                    (
                        prompt_embeds_c,
                        _,
                        _,
                        _,
                    ) = pipe.encode_prompt(
                        prompt,
                        num_images_per_prompt=1,
                        do_classifier_free_guidance=False,
                        negative_prompt=negative_prompt,
                    )
                pose_img = tensor_transfrom(pose_img).unsqueeze(0).to(device, torch.float16)
                garm_tensor = tensor_transfrom(garm_img).unsqueeze(0).to(device, torch.float16)
                generator = torch.Generator(device).manual_seed(seed) if seed is not None else None
                images = pipe(
                    prompt_embeds=prompt_embeds.to(device, torch.float16),
                    negative_prompt_embeds=negative_prompt_embeds.to(device, torch.float16),
                    pooled_prompt_embeds=pooled_prompt_embeds.to(device, torch.float16),
                    negative_pooled_prompt_embeds=negative_pooled_prompt_embeds.to(device, torch.float16),
                    num_inference_steps=denoise_steps, # [cite: 78, 79, 80]
                    generator=generator,
                    strength=1.0,
                    pose_img=pose_img.to(device, torch.float16),
                    text_embeds_cloth=prompt_embeds_c.to(device, torch.float16),
                    cloth=garm_tensor.to(device, torch.float16),
                    mask_image=mask,
                    image=human_