app.py

import sys
import os

sys.path.append('./')
os.system("pip install gradio accelerate==0.25.0 torchmetrics==1.2.1 tqdm==4.66.1 fastapi==0.111.0 transformers==4.36.2 diffusers==0.25 einops==0.7.0 bitsandbytes scipy==1.11.1 opencv-python gradio==4.24.0 fvcore cloudpickle omegaconf pycocotools basicsr av onnxruntime==1.16.2 peft==0.11.1 huggingface_hub==0.24.7 --no-deps")
import spaces
from fastapi import FastAPI
app = FastAPI()

from PIL import Image
import gradio as gr
from src.tryon_pipeline import StableDiffusionXLInpaintPipeline as TryonPipeline
from src.unet_hacked_garmnet import UNet2DConditionModel as UNet2DConditionModel_ref
from src.unet_hacked_tryon import UNet2DConditionModel
from transformers import (
    CLIPImageProcessor,
    CLIPVisionModelWithProjection,
    CLIPTextModel,
    CLIPTextModelWithProjection,
)
from diffusers import DDPMScheduler,AutoencoderKL
from typing import List

import torch
import os
from transformers import AutoTokenizer
import numpy as np
from torchvision import transforms


device = 'cuda:0' if torch.cuda.is_available() else 'cpu'

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


base_path = 'yisol/IDM-VTON'

unet = UNet2DConditionModel.from_pretrained(
    base_path,
    subfolder="unet",
    torch_dtype=torch.float16,
)
unet.requires_grad_(False)
tokenizer_one = AutoTokenizer.from_pretrained(
    base_path,
    subfolder="tokenizer",
    revision=None,
    use_fast=False,
)
tokenizer_two = AutoTokenizer.from_pretrained(
    base_path,
    subfolder="tokenizer_2",
    revision=None,
    use_fast=False,
)
noise_scheduler = DDPMScheduler.from_pretrained(base_path, subfolder="scheduler")

text_encoder_one = CLIPTextModel.from_pretrained(
    base_path,
    subfolder="text_encoder",
    torch_dtype=torch.float16,
)
text_encoder_two = CLIPTextModelWithProjection.from_pretrained(
    base_path,
    subfolder="text_encoder_2",
    torch_dtype=torch.float16,
)
image_encoder = CLIPVisionModelWithProjection.from_pretrained(
    base_path,
    subfolder="image_encoder",
    torch_dtype=torch.float16,
    )
vae = AutoencoderKL.from_pretrained(base_path,
                                    subfolder="vae",
                                    torch_dtype=torch.float16,
)

# "stabilityai/stable-diffusion-xl-base-1.0",
UNet_Encoder = UNet2DConditionModel_ref.from_pretrained(
    base_path,
    subfolder="unet_encoder",
    torch_dtype=torch.float16,
)



UNet_Encoder.requires_grad_(False)
image_encoder.requires_grad_(False)
vae.requires_grad_(False)
unet.requires_grad_(False)
text_encoder_one.requires_grad_(False)
text_encoder_two.requires_grad_(False)
tensor_transfrom = transforms.Compose(
            [
                transforms.ToTensor(),
                transforms.Normalize([0.5], [0.5]),
            ]
    )

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,
)
pipe.unet_encoder = UNet_Encoder

@spaces.GPU
def start_tryon(person_img, pose_img, mask_img, cloth_img, garment_des, denoise_steps, seed):
    # Assuming device is set up (e.g., "cuda" or "cpu")
    pipe.to(device)
    pipe.unet_encoder.to(device)

    # Resize and prepare images
    garm_img = cloth_img.convert("RGB").resize((768, 1024))
    human_img = person_img.convert("RGB").resize((768, 1024))
    mask = pil_to_binary_mask(mask_img.convert("RGB").resize((768, 1024)))

    # Prepare pose image (already uploaded)
    pose_img = pose_img.resize((768, 1024))

    
    
    # Embedding generation for prompts
    with torch.no_grad():
        with torch.cuda.amp.autocast():
            # Generate text embeddings for garment description
            prompt = f"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_cloth, 
                   _,
                   _,
                   _,
                )= pipe.encode_prompt(
                   prompt,
                   num_images_per_prompt=1,
                   do_classifier_free_guidance=False,
                   negative_prompt=negative_prompt,
                )

            # Convert images to tensors for processing
            pose_img_tensor = tensor_transfrom(pose_img).unsqueeze(0).to(device, torch.float16)
            garm_tensor = tensor_transfrom(garm_img).unsqueeze(0).to(device, torch.float16)

            # Prepare the generator with optional seed
            generator = torch.Generator(device).manual_seed(seed) if seed is not None else None

            # Generate the virtual try-on output image
            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,
                generator=generator,
                strength=1.0,
                pose_img=pose_img_tensor.to(device, torch.float16),
                text_embeds_cloth=prompt_embeds_cloth.to(device, torch.float16),
                cloth=garm_tensor.to(device, torch.float16),
                mask_image=mask,
                image=human_img,
                height=1024,
                width=768,
                ip_adapter_image=garm_img.resize((768, 1024)),
                guidance_scale=2.0,
            )[0]

    return images[0]


# Gradio interface for the virtual try-on model
image_blocks = gr.Blocks().queue()

with image_blocks as demo:
    gr.Markdown("## SmartLuga")    
    with gr.Row():
        with gr.Column():
            person_img = gr.Image(label='Person Image', sources='upload', type="pil")
            pose_img = gr.Image(label='Pose Image', sources='upload', type="pil")
            mask_img = gr.Image(label='Mask Image', sources='upload', type="pil")

        with gr.Column():
            cloth_img = gr.Image(label='Garment Image', sources='upload', type="pil")
            garment_des = gr.Textbox(placeholder="Description of garment ex) Short Sleeve Round Neck T-shirts", label="Garment Description")

        with gr.Column():
            denoise_steps = gr.Number(label="Denoising Steps", minimum=20, maximum=40, value=30, step=1)
            seed = gr.Number(label="Seed", minimum=-1, maximum=2147483647, step=1, value=42)

        with gr.Column():
            image_out = gr.Image(label="Output Image", elem_id="output-img", show_share_button=False)

    try_button = gr.Button(value="Try-on")
    try_button.click(fn=start_tryon, inputs=[person_img, pose_img, mask_img, cloth_img, garment_des, denoise_steps, seed], outputs=[image_out], api_name='tryon')

image_blocks.launch()