import os import sys import torch import diffusers import transformers import argparse import peft import copy import cv2 import spaces import gc import tempfile import imageio import threading import gradio as gr import numpy as np from flash_attn import flash_attn_func from peft import LoraConfig from omegaconf import OmegaConf from safetensors.torch import safe_open from PIL import Image, ImageDraw, ImageFilter from huggingface_hub import hf_hub_download from transformers import pipeline from models import HunyuanVideoTransformer3DModel from pipelines import HunyuanVideoImageToVideoPipeline header = """ # DRA-Ctrl Gradio App
arXiv arXiv HuggingFace HuggingFace GitHub Project
""" notice = """ For easier testing, in spatially-aligned image generation tasks, when passing the condition image to `gradio_app`, there's no need to manually input edge maps, depth maps, or other condition images - only the original image is required. The corresponding condition images will be automatically extracted. """ def init_basemodel(): global transformer, scheduler, vae, text_encoder, text_encoder_2, tokenizer, tokenizer_2, image_processor, pipe, current_task pipe = None current_task = None # init models device = "cuda" if torch.cuda.is_available() else "cpu" weight_dtype = torch.bfloat16 transformer = HunyuanVideoTransformer3DModel.from_pretrained('hunyuanvideo-community/HunyuanVideo-I2V', subfolder="transformer", inference_subject_driven=False, low_cpu_mem_usage=True).requires_grad_(False).to(device, dtype=weight_dtype) torch.cuda.empty_cache() gc.collect() scheduler = diffusers.FlowMatchEulerDiscreteScheduler() vae = diffusers.AutoencoderKLHunyuanVideo.from_pretrained('hunyuanvideo-community/HunyuanVideo-I2V', subfolder="vae", low_cpu_mem_usage=True).requires_grad_(False).to(device, dtype=weight_dtype) torch.cuda.empty_cache() gc.collect() text_encoder = transformers.LlavaForConditionalGeneration.from_pretrained('hunyuanvideo-community/HunyuanVideo-I2V', subfolder="text_encoder", low_cpu_mem_usage=True).requires_grad_(False).to(device, dtype=weight_dtype) torch.cuda.empty_cache() gc.collect() text_encoder_2 = transformers.CLIPTextModel.from_pretrained('hunyuanvideo-community/HunyuanVideo-I2V', subfolder="text_encoder_2", low_cpu_mem_usage=True).requires_grad_(False).to(device, dtype=weight_dtype) torch.cuda.empty_cache() gc.collect() tokenizer = transformers.AutoTokenizer.from_pretrained('hunyuanvideo-community/HunyuanVideo-I2V', subfolder="tokenizer") tokenizer_2 = transformers.CLIPTokenizer.from_pretrained('hunyuanvideo-community/HunyuanVideo-I2V', subfolder="tokenizer_2") image_processor = transformers.CLIPImageProcessor.from_pretrained('hunyuanvideo-community/HunyuanVideo-I2V', subfolder="image_processor") vae.enable_tiling() vae.enable_slicing() pipe = HunyuanVideoImageToVideoPipeline( text_encoder=text_encoder, tokenizer=tokenizer, transformer=transformer, vae=vae, scheduler=copy.deepcopy(scheduler), text_encoder_2=text_encoder_2, tokenizer_2=tokenizer_2, image_processor=image_processor, ) # insert LoRA lora_config = LoraConfig( r=16, lora_alpha=16, init_lora_weights="gaussian", target_modules=[ 'attn.to_k', 'attn.to_q', 'attn.to_v', 'attn.to_out.0', 'attn.add_k_proj', 'attn.add_q_proj', 'attn.add_v_proj', 'attn.to_add_out', 'ff.net.0.proj', 'ff.net.2', 'ff_context.net.0.proj', 'ff_context.net.2', 'norm1_context.linear', 'norm1.linear', 'norm.linear', 'proj_mlp', 'proj_out', ] ) transformer.add_adapter(lora_config) # hack LoRA forward def create_hacked_forward(module): if not hasattr(module, 'original_forward'): module.original_forward = module.forward lora_forward = module.forward non_lora_forward = module.base_layer.forward img_sequence_length = int((512 / 8 / 2) ** 2) encoder_sequence_length = 144 + 252 # encoder sequence: 144 img 252 txt num_imgs = 4 num_generated_imgs = 3 def hacked_lora_forward(self, x, *args, **kwargs): if x.shape[1] == img_sequence_length * num_imgs and len(x.shape) > 2: return torch.cat(( lora_forward(x[:, :-img_sequence_length*num_generated_imgs], *args, **kwargs), non_lora_forward(x[:, -img_sequence_length*num_generated_imgs:], *args, **kwargs) ), dim=1) elif x.shape[1] == encoder_sequence_length * 2 or x.shape[1] == encoder_sequence_length: return lora_forward(x, *args, **kwargs) elif x.shape[1] == img_sequence_length * num_imgs + encoder_sequence_length: return torch.cat(( lora_forward(x[:, :(num_imgs - num_generated_imgs)*img_sequence_length], *args, **kwargs), non_lora_forward(x[:, (num_imgs - num_generated_imgs)*img_sequence_length:-encoder_sequence_length], *args, **kwargs), lora_forward(x[:, -encoder_sequence_length:], *args, **kwargs) ), dim=1) elif x.shape[1] == img_sequence_length * num_imgs + encoder_sequence_length * 2: return torch.cat(( lora_forward(x[:, :(num_imgs - num_generated_imgs)*img_sequence_length], *args, **kwargs), non_lora_forward(x[:, (num_imgs - num_generated_imgs)*img_sequence_length:-2*encoder_sequence_length], *args, **kwargs), lora_forward(x[:, -2*encoder_sequence_length:], *args, **kwargs) ), dim=1) elif x.shape[1] == 3072: return non_lora_forward(x, *args, **kwargs) else: raise ValueError( f"hacked_lora_forward receives unexpected sequence length: {x.shape[1]}, input shape: {x.shape}!" ) return hacked_lora_forward.__get__(module, type(module)) for n, m in transformer.named_modules(): if isinstance(m, peft.tuners.lora.layer.Linear): m.forward = create_hacked_forward(m) @spaces.GPU def process_image_and_text(condition_image, target_prompt, condition_image_prompt, task, random_seed, num_steps, inpainting, fill_x1, fill_x2, fill_y1, fill_y2): # set up the model global pipe, current_task, transformer if current_task != task: # load LoRA weights model_root = hf_hub_download( repo_id="Kunbyte/DRA-Ctrl", filename=f"{task}.safetensors", resume_download=True) try: with safe_open(model_root, framework="pt") as f: lora_weights = {} for k in f.keys(): param = f.get_tensor(k) if k.endswith(".weight"): k = k.replace('.weight', '.default.weight') lora_weights[k] = param transformer.load_state_dict(lora_weights, strict=False) except Exception as e: raise ValueError(f'{e}') transformer.requires_grad_(False) # start generation c_txt = None if condition_image_prompt == "" else condition_image_prompt c_img = condition_image.resize((512, 512)) t_txt = target_prompt if task not in ['subject_driven', 'style_transfer']: if task == "canny": def get_canny_edge(img): img_np = np.array(img) img_gray = cv2.cvtColor(img_np, cv2.COLOR_BGR2GRAY) edges = cv2.Canny(img_gray, 100, 200) edges_tmp = Image.fromarray(edges).convert("RGB") edges[edges == 0] = 128 return Image.fromarray(edges).convert("RGB") c_img = get_canny_edge(c_img) elif task == "coloring": c_img = ( c_img.resize((512, 512)) .convert("L") .convert("RGB") ) elif task == "deblurring": blur_radius = 10 c_img = ( c_img.convert("RGB") .filter(ImageFilter.GaussianBlur(blur_radius)) .resize((512, 512)) .convert("RGB") ) elif task == "depth": def get_depth_map(img): from transformers import pipeline depth_pipe = pipeline( task="depth-estimation", model="LiheYoung/depth-anything-small-hf", device="cpu", ) return depth_pipe(img)["depth"].convert("RGB").resize((512, 512)) c_img = get_depth_map(c_img) k = (255 - 128) / 255 b = 128 c_img = c_img.point(lambda x: k * x + b) elif task == "depth_pred": c_img = c_img elif task == "fill": c_img = c_img.resize((512, 512)).convert("RGB") x1, x2 = fill_x1, fill_x2 y1, y2 = fill_y1, fill_y2 mask = Image.new("L", (512, 512), 0) draw = ImageDraw.Draw(mask) draw.rectangle((x1, y1, x2, y2), fill=255) if inpainting: mask = Image.eval(mask, lambda a: 255 - a) c_img = Image.composite( c_img, Image.new("RGB", (512, 512), (255, 255, 255)), mask ) c_img = Image.composite( c_img, Image.new("RGB", (512, 512), (128, 128, 128)), mask ) elif task == "sr": c_img = c_img.resize((int(512 / 4), int(512 / 4))).convert("RGB") c_img = c_img.resize((512, 512)) gen_img = pipe( image=c_img, prompt=[t_txt.strip()], prompt_condition=[c_txt.strip()] if c_txt is not None else None, prompt_2=[t_txt], height=512, width=512, num_frames=5, num_inference_steps=num_steps, guidance_scale=6.0, num_videos_per_prompt=1, generator=torch.Generator(device=pipe.transformer.device).manual_seed(random_seed), output_type='pt', image_embed_interleave=4, frame_gap=48, mixup=True, mixup_num_imgs=2, enhance_tp=task in ['subject_driven'], ).frames output_images = [] for i in range(10): out = gen_img[:, i:i+1, :, :, :] out = out.squeeze(0).squeeze(0).cpu().to(torch.float32).numpy() out = np.transpose(out, (1, 2, 0)) out = (out * 255).astype(np.uint8) out = Image.fromarray(out) output_images.append(out) # video = [np.array(img.convert('RGB')) for img in output_images[1:] + [output_images[0]]] # video = np.stack(video, axis=0) with tempfile.NamedTemporaryFile(suffix=".mp4", delete=False) as f: video_path = f.name imageio.mimsave(video_path, output_images[1:]+[output_images[0]], fps=5) return output_images[0], video_path def get_samples(): sample_list = [ { "task": "subject_driven", "input": "assets/subject_driven_image_generation_dreambench_input.jpg", "target_prompt": "a cat in a chef outfit", "condition_image_prompt": "a cat", "output": "assets/subject_driven_image_generation_dreambench_output.png", "inpainting": False, "fill_x1": None, "fill_x2": None, "fill_y1": None, "fill_y2": None, }, { "task": "subject_driven", "input": "assets/subject_driven_image_generation_input.jpg", "target_prompt": "The woman stands in a snowy forest, captured in a half-portrait", "condition_image_prompt": "Woman in cream knit sweater sits calmly by a crackling fireplace, surrounded by warm candlelight and rustic wooden shelves", "output": "assets/subject_driven_image_generation_output.png", "inpainting": False, "fill_x1": None, "fill_x2": None, "fill_y1": None, "fill_y2": None, }, { "task": "canny", "input": "assets/canny_to_image_input.jpg", "target_prompt": "Mosquito frozen in clear ice cube on sand, glowing sunset casting golden light with misty halo around ice", "condition_image_prompt": "", "output": "assets/canny_to_image_output.png", "inpainting": False, "fill_x1": None, "fill_x2": None, "fill_y1": None, "fill_y2": None, }, { "task": "coloring", "input": "assets/colorization_input.jpg", "target_prompt": "A vibrant young woman with rainbow glasses, yellow eyes, and colorful feather accessory against a bright yellow background", "condition_image_prompt": "", "output": "assets/colorization_output.png", "inpainting": False, "fill_x1": None, "fill_x2": None, "fill_y1": None, "fill_y2": None, }, { "task": "deblurring", "input": "assets/deblurring_input.jpg", "target_prompt": "Vibrant rainbow ball creates dramatic splash in clear water, bubbles swirling against crisp white background", "condition_image_prompt": "", "output": "assets/deblurring_output.png", "inpainting": False, "fill_x1": None, "fill_x2": None, "fill_y1": None, "fill_y2": None, }, { "task": "depth", "input": "assets/depth_to_image_input.jpg", "target_prompt": "Golden-brown cat-shaped bread loaf with closed eyes rests on wooden table, soft kitchen blur in background", "condition_image_prompt": "", "output": "assets/depth_to_image_output.png", "inpainting": False, "fill_x1": None, "fill_x2": None, "fill_y1": None, "fill_y2": None, }, { "task": "depth_pred", "input": "assets/depth_prediction_input.jpg", "target_prompt": "Steaming bowl of ramen with pork slices, soft-boiled egg, greens, and scallions in rich broth on wooden table", "condition_image_prompt": "", "output": "assets/depth_prediction_output.png", "inpainting": False, "fill_x1": None, "fill_x2": None, "fill_y1": None, "fill_y2": None, }, { "task": "fill", "input": "assets/inpainting_input.jpg", "target_prompt": "Mona Lisa dons a medical mask, her enigmatic smile now concealed beneath crisp white fabric", "condition_image_prompt": "", "output": "assets/inpainting_output.png", "inpainting": True, "fill_x1": 170, "fill_x2": 300, "fill_y1": 190, "fill_y2": 290, }, { "task": "fill", "input": "assets/outpainting_input.jpg", "target_prompt": "Her left hand emerges at the frame's lower right, delicately cradling a vibrant red flower against the black void", "condition_image_prompt": "", "output": "assets/outpainting_output.png", "inpainting": False, "fill_x1": 155, "fill_x2": 512, "fill_y1": 0, "fill_y2": 330, }, { "task": "sr", "input": "assets/super_resolution_input.jpg", "target_prompt": "Crispy buffalo wings and golden fries rest on a red-and-white checkered paper lining a gleaming metal tray, with creamy dip", "condition_image_prompt": "", "output": "assets/super_resolution_output.png", "inpainting": False, "fill_x1": None, "fill_x2": None, "fill_y1": None, "fill_y2": None, }, { "task": "style_transfer", "input": "assets/style_transfer_input.png", "target_prompt": "bitmoji style. An orange cat sits quietly on the stone slab. Beside it are the green grasses. With its ears perked up, it looks to one side.", "condition_image_prompt": "An orange cat sits quietly on the stone slab. Beside it are the green grasses. With its ears perked up, it looks to one side.", "output": "assets/style_transfer_output.png", "inpainting": False, "fill_x1": None, "fill_x2": None, "fill_y1": None, "fill_y2": None, }, ] return [ [ sample['task'], Image.open(sample['input']), sample['target_prompt'], sample['condition_image_prompt'], Image.open(sample['output']), sample['inpainting'], sample['fill_x1'], sample['fill_x2'], sample['fill_y1'], sample['fill_y2'], ] for sample in sample_list ] def create_app(): with gr.Blocks() as app: gr.Markdown(header, elem_id="header") gr.Markdown("🚦 To ensure stable model output, we are running the process in a single-threaded serial mode. If your request is queued, please wait patiently for the generation to complete.", elem_id="queue_notice") with gr.Row(equal_height=False): with gr.Column(variant="panel", elem_classes="inputPanel"): condition_image = gr.Image( type="pil", label="Condition Image", width=300, elem_id="input" ) task = gr.Radio( [ ("Subject-driven Image Generation", "subject_driven"), ("Canny-to-Image", "canny"), ("Colorization", "coloring"), ("Deblurring", "deblurring"), ("Depth-to-Image", "depth"), ("Depth Prediction", "depth_pred"), ("In/Out-Painting", "fill"), ("Super-Resolution", "sr"), ("Style Transfer", "style_transfer") ], label="Task Selection", value="subject_driven", interactive=True, elem_id="task_selection" ) gr.Markdown(notice, elem_id="notice") target_prompt = gr.Textbox(lines=2, label="Target Prompt", elem_id="tp") gr.Markdown("**Condition Image Prompt** _(Only required by Subject-driven Image Generation and Style Transfer tasks)_") condition_image_prompt = gr.Textbox(lines=2, label="Condition Image Prompt", elem_id="cp") random_seed = gr.Number(label="Random Seed", precision=0, value=0, elem_id="seed") num_steps = gr.Number(label="Diffusion Inference Steps", precision=0, value=50, elem_id="steps") inpainting = gr.Checkbox(label="Inpainting", value=False, elem_id="inpainting") fill_x1 = gr.Number(label="In/Out-painting Box Left Boundary", precision=0, value=128, elem_id="fill_x1") fill_x2 = gr.Number(label="In/Out-painting Box Right Boundary", precision=0, value=384, elem_id="fill_x2") fill_y1 = gr.Number(label="In/Out-painting Box Top Boundary", precision=0, value=128, elem_id="fill_y1") fill_y2 = gr.Number(label="In/Out-painting Box Bottom Boundary", precision=0, value=384, elem_id="fill_y2") submit_btn = gr.Button("Run", elem_id="submit_btn") with gr.Column(variant="panel", elem_classes="outputPanel"): # output_image = gr.Image(type="pil", elem_id="output") # output_images = gr.Gallery( # label="Output Images", # show_label=True, # elem_id="output_gallery", # columns=1, # rows=10, # object_fit="contain", # height="auto", # ) output_image = gr.Image( type="pil", label="Output Image", elem_id="output_image" ) output_video = gr.Video( label="Output Video", elem_id="output_video" ) with gr.Row(): examples = gr.Examples( examples=get_samples(), inputs=[task, condition_image, target_prompt, condition_image_prompt, output_image, inpainting, fill_x1, fill_x2, fill_y1, fill_y2], label="Examples", ) submit_btn.click( fn=process_image_and_text, inputs=[condition_image, target_prompt, condition_image_prompt, task, random_seed, num_steps, inpainting, fill_x1, fill_x2, fill_y1, fill_y2], outputs=[output_image, output_video], ) return app if __name__ == "__main__": init_basemodel() app = create_app() app.queue(default_concurrency_limit=1) app.launch(debug=True, ssr_mode=False, max_threads=1)