import os os.environ["CUDA_VISIBLE_DEVICES"] = "1" import sys sys.path.append(".") sys.path.append("..") import time import argparse import cv2 import scipy import torch import numpy as np import face_alignment as FAN from PIL import Image from tqdm import tqdm from facenet_pytorch import MTCNN from torchvision import transforms from models.deep3d_module.util.preprocess import align_img from models.deep3d_module.util.load_mats import load_lm3d device = 'cuda' BFM_PATH = '/raid/xjd/workspace/opensource/Live3DPortrait/models/deep3d_module/BFM' mtcnn = MTCNN(keep_all=True, device=device) fa = FAN.FaceAlignment(FAN.LandmarksType.TWO_D, flip_input=False) lm3d_std = load_lm3d(BFM_PATH) ''' This script is used to preprocess a video ''' # convert 68 landmarks to 5 landmarks def extract_5p(lm): lm_idx = np.array([31, 37, 40, 43, 46, 49, 55]) - 1 lm5p = np.stack([lm[lm_idx[0], :], np.mean(lm[lm_idx[[1, 2]], :], 0), np.mean( lm[lm_idx[[3, 4]], :], 0), lm[lm_idx[5], :], lm[lm_idx[6], :]], axis=0) lm5p = lm5p[[1, 2, 0, 3, 4], :] return lm5p def extract_images(path, out_path, fps=25): # with same resolution print(f'[INFO] ===== extract images from {path} to {out_path} =====') cmd = f'ffmpeg -i {path} -vf fps={fps} -s 512x512 -qmin 1 -q:v 1 -start_number 0 {os.path.join(out_path, "%d.jpg")}' os.system(cmd) print(f'[INFO] ===== extracted images =====') def align(img, landmarks68, output_size=1500): enable_padding = True # transform_size = int(512 * 4096 / 1500) transform_size = 4096 # Parse landmarks. # pylint: disable=unused-variable lm = np.array(landmarks68) lm_chin = lm[0 : 17] # left-right lm_eyebrow_left = lm[17 : 22] # left-right lm_eyebrow_right = lm[22 : 27] # left-right lm_nose = lm[27 : 31] # top-down lm_nostrils = lm[31 : 36] # top-down lm_eye_left = lm[36 : 42] # left-clockwise lm_eye_right = lm[42 : 48] # left-clockwise lm_mouth_outer = lm[48 : 60] # left-clockwise lm_mouth_inner = lm[60 : 68] # left-clockwise # Calculate auxiliary vectors. eye_left = np.mean(lm_eye_left, axis=0) eye_right = np.mean(lm_eye_right, axis=0) eye_avg = (eye_left + eye_right) * 0.5 eye_to_eye = eye_right - eye_left mouth_left = lm_mouth_outer[0] mouth_right = lm_mouth_outer[6] mouth_avg = (mouth_left + mouth_right) * 0.5 eye_to_mouth = mouth_avg - eye_avg # Choose oriented crop rectangle. x = eye_to_eye - np.flipud(eye_to_mouth) * [-1, 1] x /= np.hypot(*x) x *= max(np.hypot(*eye_to_eye) * 2.0, np.hypot(*eye_to_mouth) * 1.8) y = np.flipud(x) * [-1, 1] q_scale = 1.8 x = q_scale * x y = q_scale * y c = eye_avg + eye_to_mouth * 0.1 quad = np.stack([c - x - y, c - x + y, c + x + y, c + x - y]) qsize = np.hypot(*x) * 2 import time # Shrink. # start_time = time.time() shrink = int(np.floor(qsize / output_size * 0.5)) if shrink > 1: rsize = (int(np.rint(float(img.size[0]) / shrink)), int(np.rint(float(img.size[1]) / shrink))) img = img.resize(rsize, Image.ANTIALIAS) quad /= shrink qsize /= shrink # print("shrink--- %s seconds ---" % (time.time() - start_time)) # Crop. # start_time = time.time() border = max(int(np.rint(qsize * 0.1)), 3) crop = (int(np.floor(min(quad[:,0]))), int(np.floor(min(quad[:,1]))), int(np.ceil(max(quad[:,0]))), int(np.ceil(max(quad[:,1])))) crop = (max(crop[0] - border, 0), max(crop[1] - border, 0), min(crop[2] + border, img.size[0]), min(crop[3] + border, img.size[1])) if crop[2] - crop[0] < img.size[0] or crop[3] - crop[1] < img.size[1]: img = img.crop(crop) quad -= crop[0:2] # print("crop--- %s seconds ---" % (time.time() - start_time)) # Pad. # start_time = time.time() pad = (int(np.floor(min(quad[:,0]))), int(np.floor(min(quad[:,1]))), int(np.ceil(max(quad[:,0]))), int(np.ceil(max(quad[:,1])))) pad = (max(-pad[0] + border, 0), max(-pad[1] + border, 0), max(pad[2] - img.size[0] + border, 0), max(pad[3] - img.size[1] + border, 0)) if enable_padding and max(pad) > border - 4: pad = np.maximum(pad, int(np.rint(qsize * 0.3))) img = np.pad(np.float32(img), ((pad[1], pad[3]), (pad[0], pad[2]), (0, 0)), 'reflect') h, w, _ = img.shape y, x, _ = np.ogrid[:h, :w, :1] mask = np.maximum(1.0 - np.minimum(np.float32(x) / pad[0], np.float32(w-1-x) / pad[2]), 1.0 - np.minimum(np.float32(y) / pad[1], np.float32(h-1-y) / pad[3])) low_res = cv2.resize(img, (0,0), fx=0.1, fy=0.1, interpolation = cv2.INTER_AREA) blur = qsize * 0.02*0.1 low_res = scipy.ndimage.gaussian_filter(low_res, [blur, blur, 0]) low_res = cv2.resize(low_res, (img.shape[1], img.shape[0]), interpolation = cv2.INTER_LANCZOS4) img += (low_res - img) * np.clip(mask * 3.0 + 1.0, 0.0, 1.0) median = cv2.resize(img, (0,0), fx=0.1, fy=0.1, interpolation = cv2.INTER_AREA) median = np.median(median, axis=(0,1)) img += (median - img) * np.clip(mask, 0.0, 1.0) img = Image.fromarray(np.uint8(np.clip(np.rint(img), 0, 255)), 'RGB') quad += pad[:2] # print("pad--- %s seconds ---" % (time.time() - start_time)) # Transform. # start_time = time.time() img = img.transform((transform_size, transform_size), Image.QUAD, (quad + 0.5).flatten(), Image.BILINEAR) if output_size < transform_size: img = img.resize((output_size, output_size), Image.ANTIALIAS) # print("transform--- %s seconds ---" % (time.time() - start_time)) return img def preprocess_single_image(image_path, output_path, eye_align=False): image = Image.open(image_path).convert('RGB') if eye_align: # Detect face using face_alignment landmarks = fa.get_landmarks_from_image(image_path) img = align(image, landmarks[0]) # center crop transform = transforms.Compose([ transforms.CenterCrop(1024), # 700, FFHQ in eg3d process need their own crop param # transforms.Resize(512), ]) image = transform(img) # Detect face using MTCNN # s = time.time() # image_tensor = transforms.ToTensor()(image).unsqueeze(0).to(device=device) boxes, probs, points = mtcnn.detect(image, landmarks=True) # about 2~3.5s # print(f'[INFO] ===== MTCNN detect time: {time.time() - s} =====') _,H = image.size lm = points if boxes is None or boxes[0] is None or len(boxes) == 0: print(f'[INFO] ===== MTCNN detect failed =====') return lm = lm.reshape([-1, 2]) lm = lm[:5, :] # only use 5 landmarks,select the largest prob face lm[:, -1] = H - 1 - lm[:, -1] target_size = 1024. rescale_factor = 300 center_crop_size = 700 output_size = 512 # s = time.time() _, im_high, _, _, = align_img(image, lm, lm3d_std, target_size=target_size, rescale_factor=rescale_factor) # about 0.02s # print(f'[INFO] ===== align_img time: {time.time() - s} =====') left = int(im_high.size[0]/2 - center_crop_size/2) upper = int(im_high.size[1]/2 - center_crop_size/2) right = left + center_crop_size lower = upper + center_crop_size im_cropped = im_high.crop((left, upper, right,lower)) im_cropped = im_cropped.resize((output_size, output_size), resample=Image.LANCZOS) basename = os.path.basename(image_path) im_cropped.save(os.path.join(output_path, basename)) def preprocess_video(video_path, output_path): ori_imgs_dir = os.path.join(output_path, 'ori_imgs') # two mode: crop and crop_align crop_imgs_dir = os.path.join(output_path, 'crop_imgs') crop_align_imgs_dir = os.path.join(output_path, 'crop_align_imgs') if not os.path.exists(ori_imgs_dir): os.makedirs(ori_imgs_dir) if not os.path.exists(crop_imgs_dir): os.makedirs(crop_imgs_dir) if not os.path.exists(crop_align_imgs_dir): os.makedirs(crop_align_imgs_dir) # extract frames from video extract_images(video_path, ori_imgs_dir) # preprocess images for img_name in tqdm(os.listdir(ori_imgs_dir)): img_path = os.path.join(ori_imgs_dir, img_name) preprocess_single_image(img_path, crop_imgs_dir, eye_align=False) # preprocess_single_image(img_path, crop_align_imgs_dir, eye_align=True) def main(): parser = argparse.ArgumentParser(description='Preprocess video.') parser.add_argument('--dataset_path', type=str, default="/NAS/xjd/datasets/CelebVHQ/35666/", help='Path to the image to be preprocessed.') parser.add_argument('--output_path', type=str, default="/NAS/xjd/datasets/CelebVHQ/preprocess", help='Path to save the preprocessed image.') args = parser.parse_args() for video_name in os.listdir(args.dataset_path): video_path = os.path.join(args.dataset_path, video_name) output_path = os.path.join(args.output_path, video_name[:-4]) # remove .mp4 if not os.path.exists(output_path): os.makedirs(output_path) preprocess_video(video_path, output_path) if __name__ == "__main__": main()