TokenFlow-master/preprocess.py

from transformers import CLIPTextModel, CLIPTokenizer, logging
from diffusers import AutoencoderKL, UNet2DConditionModel, DDIMScheduler
logging.set_verbosity_error()

import os
from tqdm import tqdm, trange
import torch
import torch.nn as nn
import argparse
import json
from torchvision.io import write_video
from pathlib import Path
from util import *
import torchvision.transforms as T


def get_timesteps(scheduler, num_inference_steps, strength, device):
    init_timestep = min(int(num_inference_steps * strength), num_inference_steps)
    t_start = max(num_inference_steps - init_timestep, 0)
    timesteps = scheduler.timesteps[t_start:]
    return timesteps, num_inference_steps - t_start


class Preprocess(nn.Module):
    def __init__(self, device, opt, hf_key=None):
        super().__init__()
        self.device = device
        self.sd_version = opt.sd_version
        self.use_depth = False
        print(f'[INFO] loading stable diffusion...')
        if hf_key is not None:
            print(f'[INFO] using hugging face custom model key: {hf_key}')
            model_key = hf_key
        elif self.sd_version == '2.1':
            model_key = "stabilityai/stable-diffusion-2-1-base"
        elif self.sd_version == '2.0':
            model_key = "stabilityai/stable-diffusion-2-base"
        elif self.sd_version == '1.5' or self.sd_version == 'ControlNet':
            model_key = "runwayml/stable-diffusion-v1-5"
        elif self.sd_version == 'depth':
            model_key = "stabilityai/stable-diffusion-2-depth"
        else:
            raise ValueError(f'Stable-diffusion version {self.sd_version} not supported.')
        self.model_key = model_key
        self.vae = AutoencoderKL.from_pretrained(model_key, subfolder="vae", revision="fp16",
                                                 torch_dtype=torch.float16).to(self.device)
        self.tokenizer = CLIPTokenizer.from_pretrained(model_key, subfolder="tokenizer")
        self.text_encoder = CLIPTextModel.from_pretrained(model_key, subfolder="text_encoder", revision="fp16",
                                                          torch_dtype=torch.float16).to(self.device)
        self.unet = UNet2DConditionModel.from_pretrained(model_key, subfolder="unet", revision="fp16",
                                                         torch_dtype=torch.float16).to(self.device)
        self.paths, self.frames, self.latents = self.get_data(opt.data_path, opt.n_frames)
        
        if self.sd_version == 'ControlNet':
            from diffusers import ControlNetModel, StableDiffusionControlNetPipeline
            controlnet = ControlNetModel.from_pretrained("lllyasviel/sd-controlnet-canny", torch_dtype=torch.float16).to(self.device)
            control_pipe = StableDiffusionControlNetPipeline.from_pretrained(
                "runwayml/stable-diffusion-v1-5", controlnet=controlnet, torch_dtype=torch.float16
            ).to(self.device)
            self.unet = control_pipe.unet
            self.controlnet = control_pipe.controlnet
            self.canny_cond = self.get_canny_cond()
        elif self.sd_version == 'depth':
            self.depth_maps = self.prepare_depth_maps()
        self.scheduler = DDIMScheduler.from_pretrained(model_key, subfolder="scheduler")
        print(f'[INFO] loaded stable diffusion!')
        
    @torch.no_grad()   
    def prepare_depth_maps(self, model_type='DPT_Large', device='cuda'):
        depth_maps = []
        midas = torch.hub.load("intel-isl/MiDaS", model_type)
        midas.to(device)
        midas.eval()
        midas_transforms = torch.hub.load("intel-isl/MiDaS", "transforms")
        if model_type == "DPT_Large" or model_type == "DPT_Hybrid":
            transform = midas_transforms.dpt_transform
        else:
            transform = midas_transforms.small_transform
        for i in range(len(self.paths)):
            img = cv2.imread(self.paths[i])
            img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
            latent_h = img.shape[0] // 8
            latent_w = img.shape[1] // 8
            input_batch = transform(img).to(device)
            prediction = midas(input_batch)
            depth_map = torch.nn.functional.interpolate(
                prediction.unsqueeze(1),
                size=(latent_h, latent_w),
                mode="bicubic",
                align_corners=False,
            )
            depth_min = torch.amin(depth_map, dim=[1, 2, 3], keepdim=True)
            depth_max = torch.amax(depth_map, dim=[1, 2, 3], keepdim=True)
            depth_map = 2.0 * (depth_map - depth_min) / (depth_max - depth_min) - 1.0
            depth_maps.append(depth_map)
        return torch.cat(depth_maps).to(self.device).to(torch.float16)
    
    @torch.no_grad()
    def get_canny_cond(self):
        canny_cond = []
        for image in self.frames.cpu().permute(0, 2, 3, 1):
            image = np.uint8(np.array(255 * image))
            low_threshold = 100
            high_threshold = 200
            image = cv2.Canny(image, low_threshold, high_threshold)
            image = image[:, :, None]
            image = np.concatenate([image, image, image], axis=2)
            image = torch.from_numpy((image.astype(np.float32) / 255.0))
            canny_cond.append(image)
        canny_cond = torch.stack(canny_cond).permute(0, 3, 1, 2).to(self.device).to(torch.float16)
        return canny_cond
    
    def controlnet_pred(self, latent_model_input, t, text_embed_input, controlnet_cond):
        down_block_res_samples, mid_block_res_sample = self.controlnet(
            latent_model_input,
            t,
            encoder_hidden_states=text_embed_input,
            controlnet_cond=controlnet_cond,
            conditioning_scale=1,
            return_dict=False,
        )
        noise_pred = self.unet(
            latent_model_input,
            t,
            encoder_hidden_states=text_embed_input,
            cross_attention_kwargs={},
            down_block_additional_residuals=down_block_res_samples,
            mid_block_additional_residual=mid_block_res_sample,
            return_dict=False,
        )[0]
        return noise_pred
    
    @torch.no_grad()
    def get_text_embeds(self, prompt, negative_prompt, device="cuda"):
        text_input = self.tokenizer(prompt, padding='max_length', max_length=self.tokenizer.model_max_length,
                                    truncation=True, return_tensors='pt')
        text_embeddings = self.text_encoder(text_input.input_ids.to(device))[0]
        uncond_input = self.tokenizer(negative_prompt, padding='max_length', max_length=self.tokenizer.model_max_length,
                                      return_tensors='pt')
        uncond_embeddings = self.text_encoder(uncond_input.input_ids.to(device))[0]
        text_embeddings = torch.cat([uncond_embeddings, text_embeddings])
        return text_embeddings

    @torch.no_grad()
    def decode_latents(self, latents):
        decoded = []
        batch_size = 8
        for b in range(0, latents.shape[0], batch_size):
                latents_batch = 1 / 0.18215 * latents[b:b + batch_size]
                imgs = self.vae.decode(latents_batch).sample
                imgs = (imgs / 2 + 0.5).clamp(0, 1)
                decoded.append(imgs)
        return torch.cat(decoded)

    @torch.no_grad()
    def encode_imgs(self, imgs, batch_size=10, deterministic=True):
        imgs = 2 * imgs - 1
        latents = []
        for i in range(0, len(imgs), batch_size):
            posterior = self.vae.encode(imgs[i:i + batch_size]).latent_dist
            latent = posterior.mean if deterministic else posterior.sample()
            latents.append(latent * 0.18215)
        latents = torch.cat(latents)
        return latents

    def get_data(self, frames_path, n_frames):
        paths = [f"{frames_path}/{i:05d}.png" for i in range(n_frames)]
        print(f"Checking paths: {paths}")
        frames = []
        for path in paths:
            if os.path.exists(path):
                frames.append(Image.open(path).convert('RGB'))
            else:
                print(f"File not found: {path}")
        if not frames:
            raise FileNotFoundError(f"No frames found in {frames_path}")
        if frames[0].size[0] == frames[0].size[1]:
            frames = [frame.resize((512, 512), resample=Image.Resampling.LANCZOS) for frame in frames]
        frames = torch.stack([T.ToTensor()(frame) for frame in frames]).to(torch.float16).to(self.device)
        latents = self.encode_imgs(frames, deterministic=True).to(torch.float16).to(self.device)
        return paths, frames, latents

    @torch.no_grad()
    def ddim_inversion(self, cond, latent_frames, save_path, batch_size, save_latents=True, timesteps_to_save=None):
        timesteps = reversed(self.scheduler.timesteps)
        timesteps_to_save = timesteps_to_save if timesteps_to_save is not None else timesteps
        for i, t in enumerate(tqdm(timesteps)):
            for b in range(0, latent_frames.shape[0], batch_size):
                x_batch = latent_frames[b:b + batch_size]
                model_input = x_batch
                cond_batch = cond.repeat(x_batch.shape[0], 1, 1)
                if self.sd_version == 'depth':
                    depth_maps = torch.cat([self.depth_maps[b: b + batch_size]])
                    model_input = torch.cat([x_batch, depth_maps],dim=1)
                alpha_prod_t = self.scheduler.alphas_cumprod[t]
                alpha_prod_t_prev = (
                    self.scheduler.alphas_cumprod[timesteps[i - 1]]
                    if i > 0 else self.scheduler.final_alpha_cumprod
                )
                mu = alpha_prod_t ** 0.5
                mu_prev = alpha_prod_t_prev ** 0.5
                sigma = (1 - alpha_prod_t) ** 0.5
                sigma_prev = (1 - alpha_prod_t_prev) ** 0.5
                eps = self.unet(model_input, t, encoder_hidden_states=cond_batch).sample if self.sd_version != 'ControlNet' \
                    else self.controlnet_pred(x_batch, t, cond_batch, torch.cat([self.canny_cond[b: b + batch_size]]))
                pred_x0 = (x_batch - sigma_prev * eps) / mu_prev
                latent_frames[b:b + batch_size] = mu * pred_x0 + sigma * eps
            if save_latents and t in timesteps_to_save:
                torch.save(latent_frames, os.path.join(save_path, 'latents', f'noisy_latents_{t}.pt'))
        torch.save(latent_frames, os.path.join(save_path, 'latents', f'noisy_latents_{t}.pt'))
        return latent_frames

    @torch.no_grad()
    def ddim_sample(self, x, cond, batch_size):
        timesteps = self.scheduler.timesteps
        for i, t in enumerate(tqdm(timesteps)):
            for b in range(0, x.shape[0], batch_size):
                x_batch = x[b:b + batch_size]
                model_input = x_batch
                cond_batch = cond.repeat(x_batch.shape[0], 1, 1)
                if self.sd_version == 'depth':
                    depth_maps = torch.cat([self.depth_maps[b: b + batch_size]])
                    model_input = torch.cat([x_batch, depth_maps],dim=1)
                alpha_prod_t = self.scheduler.alphas_cumprod[t]
                alpha_prod_t_prev = (
                    self.scheduler.alphas_cumprod[timesteps[i + 1]]
                    if i < len(timesteps) - 1
                    else self.scheduler.final_alpha_cumprod
                )
                mu = alpha_prod_t ** 0.5
                sigma = (1 - alpha_prod_t) ** 0.5
                mu_prev = alpha_prod_t_prev ** 0.5
                sigma_prev = (1 - alpha_prod_t_prev) ** 0.5
                eps = self.unet(model_input, t, encoder_hidden_states=cond_batch).sample if self.sd_version != 'ControlNet' \
                    else self.controlnet_pred(x_batch, t, cond_batch, torch.cat([self.canny_cond[b: b + batch_size]]))
                pred_x0 = (x_batch - sigma * eps) / mu
                x[b:b + batch_size] = mu_prev * pred_x0 + sigma_prev * eps
        return x

    @torch.no_grad()
    def extract_latents(self, 
                        num_steps,
                        save_path,
                        batch_size,
                        timesteps_to_save,
                        inversion_prompt=''):
        self.scheduler.set_timesteps(num_steps)
        cond = self.get_text_embeds(inversion_prompt, "")[1].unsqueeze(0)
        latent_frames = self.latents
        inverted_x = self.ddim_inversion(cond,
                                         latent_frames,
                                         save_path,
                                         batch_size=batch_size,
                                         save_latents=True,
                                         timesteps_to_save=timesteps_to_save)
        latent_reconstruction = self.ddim_sample(inverted_x, cond, batch_size=batch_size)
        rgb_reconstruction = self.decode_latents(latent_reconstruction)
        return rgb_reconstruction


def save_video_frames(video_path, output_dir='data', img_size=(512,512)):
    if not os.path.exists(video_path):
        raise FileNotFoundError(f"Video file not found: {video_path}")
    video, _, _ = read_video(video_path, output_format="TCHW")
    print(f"Extracted {len(video)} frames from video")
    if len(video) == 0:
        raise ValueError("No frames extracted from video")
    if video_path.endswith('.mov'):
        video = T.functional.rotate(video, -90)
    os.makedirs(output_dir, exist_ok=True)
    for i in range(len(video)):
        ind = str(i).zfill(5)
        image = T.ToPILImage()(video[i])
        image_resized = image.resize(img_size, resample=Image.Resampling.LANCZOS)
        save_path = os.path.join(output_dir, f"{ind}.png")
        image_resized.save(save_path)
        print(f"Saved frame: {save_path}")


def prep(opt, video_info):
    video_filename = video_info['video']
    inversion_prompt = video_info['prompt']
    video_name = Path(video_filename).stem
    video_path = os.path.join('/home/wangjuntong/video_editing_dataset/all_sourse', video_filename)
    print(f"Video path: {video_path}")
    if not os.path.exists(video_path):
        raise FileNotFoundError(f"Video file not found: {video_path}")
    frames_dir = os.path.join('data', video_name)
    os.makedirs(frames_dir, exist_ok=True)
    save_video_frames(video_path, output_dir=frames_dir, img_size=(opt.W, opt.H))
    print(f"Frames saved to: {frames_dir}")
    print(f"Files in directory: {os.listdir(frames_dir)}")
    opt.data_path = frames_dir
    save_path = os.path.join(opt.save_dir, video_name, f'sd_{opt.sd_version}', f'steps_{opt.steps}', f'nframes_{opt.n_frames}')
    os.makedirs(save_path, exist_ok=True)
    os.makedirs(os.path.join(save_path, 'latents'), exist_ok=True)
    with open(os.path.join(save_path, 'inversion_prompt.txt'), 'w') as f:
        f.write(inversion_prompt)
    add_dict_to_yaml_file(os.path.join(opt.save_dir, 'inversion_prompts.yaml'), video_name, inversion_prompt)
    if opt.sd_version == '2.1':
        model_key = "stabilityai/stable-diffusion-2-1-base"
    elif opt.sd_version == '2.0':
        model_key = "stabilityai/stable-diffusion-2-base"
    elif opt.sd_version == '1.5' or opt.sd_version == 'ControlNet':
        model_key = "runwayml/stable-diffusion-v1-5"
    elif opt.sd_version == 'depth':
        model_key = "stabilityai/stable-diffusion-2-depth"
    toy_scheduler = DDIMScheduler.from_pretrained(model_key, subfolder="scheduler")
    toy_scheduler.set_timesteps(opt.save_steps)
    timesteps_to_save, num_inference_steps = get_timesteps(toy_scheduler, num_inference_steps=opt.save_steps, strength=1.0, device=device)
    model = Preprocess(device, opt)
    recon_frames = model.extract_latents(
        num_steps=opt.steps,
        save_path=save_path,
        batch_size=opt.batch_size,
        timesteps_to_save=timesteps_to_save,
        inversion_prompt=inversion_prompt,
    )
    frames_path = os.path.join(save_path, 'frames')
    os.makedirs(frames_path, exist_ok=True)
    for i, frame in enumerate(recon_frames):
        T.ToPILImage()(frame).save(os.path.join(frames_path, f'{i:05d}.png'))
    frames = (recon_frames * 255).to(torch.uint8).cpu().permute(0, 2, 3, 1)
    write_video(os.path.join(save_path, f'inverted.mp4'), frames, fps=10)


if __name__ == "__main__":
    device = 'cuda'
    parser = argparse.ArgumentParser()
    parser.add_argument('--jsonl_path', type=str, required=True, help='Path to the JSONL file containing video information')
    parser.add_argument('--save_dir', type=str, default='./outputs', help='Directory to save the outputs')
    parser.add_argument('--sd_version', type=str, default='2.1', choices=['1.5', '2.0', '2.1', 'ControlNet', 'depth'], help='Stable diffusion version')
    parser.add_argument('--steps', type=int, default=500, help='Number of steps')
    parser.add_argument('--batch_size', type=int, default=40, help='Batch size')
    parser.add_argument('--save_steps', type=int, default=50, help='Save steps')
    parser.add_argument('--n_frames', type=int, default=40, help='Number of frames')
    parser.add_argument('--H', type=int, default=512, help='Height of the frames')
    parser.add_argument('--W', type=int, default=512, help='Width of the frames')
    opt = parser.parse_args()
    with open(opt.jsonl_path, 'r') as f:
        lines = f.readlines()
    video_infos = [json.loads(line) for line in lines]
    seed_everything(1)
    for video_info in video_infos:
        prep(opt, video_info)