import os import cv2 from models import Wav2Lip import torch import audio import numpy as np import subprocess import platform import face_detection def get_smoothened_boxes(boxes, T): for i in range(len(boxes)): if i + T > len(boxes): window = boxes[len(boxes) - T:] else: window = boxes[i : i + T] boxes[i] = np.mean(window, axis=0) return boxes def face_detect(images, device): detector = face_detection.FaceAlignment(face_detection.LandmarksType._2D, flip_input=False, device=device) batch_size = 16 while 1: predictions = [] try: for i in range(0, len(images), batch_size): predictions.extend(detector.get_detections_for_batch(np.array(images[i:i + batch_size]))) except RuntimeError: if batch_size == 1: raise RuntimeError('Image too big to run face detection on GPU. Please use the --resize_factor argument') batch_size //= 2 print('Recovering from OOM error; New batch size: {}'.format(batch_size)) continue break results = [] pady1, pady2, padx1, padx2 = [0, 10, 0, 0] for rect, image in zip(predictions, images): if rect is None: cv2.imwrite('temp/faulty_frame.jpg', image) # check this frame where the face was not detected. raise ValueError('Face not detected! Ensure the video contains a face in all the frames.') y1 = max(0, rect[1] - pady1) y2 = min(image.shape[0], rect[3] + pady2) x1 = max(0, rect[0] - padx1) x2 = min(image.shape[1], rect[2] + padx2) results.append([x1, y1, x2, y2]) boxes = np.array(results) boxes = get_smoothened_boxes(boxes, T=5) results = [[image[y1: y2, x1:x2], (y1, y2, x1, x2)] for image, (x1, y1, x2, y2) in zip(images, boxes)] del detector return results def datagen(frames, mels, batch_size, device): img_size = 96 img_batch, mel_batch, frame_batch, coords_batch = [], [], [], [] #face_det_results = face_detect([frames[0]], device) face_det_results = face_detect(frames, device) for i, m in enumerate(mels): idx = i%len(frames) frame_to_save = frames[idx].copy() face, coords = face_det_results[idx].copy() face = cv2.resize(face, (img_size, img_size)) img_batch.append(face) mel_batch.append(m) frame_batch.append(frame_to_save) coords_batch.append(coords) if len(img_batch) >= batch_size: img_batch, mel_batch = np.asarray(img_batch), np.asarray(mel_batch) img_masked = img_batch.copy() img_masked[:, img_size//2:] = 0 img_batch = np.concatenate((img_masked, img_batch), axis=3) / 255. mel_batch = np.reshape(mel_batch, [len(mel_batch), mel_batch.shape[1], mel_batch.shape[2], 1]) yield img_batch, mel_batch, frame_batch, coords_batch img_batch, mel_batch, frame_batch, coords_batch = [], [], [], [] if len(img_batch) > 0: img_batch, mel_batch = np.asarray(img_batch), np.asarray(mel_batch) img_masked = img_batch.copy() img_masked[:, img_size//2:] = 0 img_batch = np.concatenate((img_masked, img_batch), axis=3) / 255. mel_batch = np.reshape(mel_batch, [len(mel_batch), mel_batch.shape[1], mel_batch.shape[2], 1]) yield img_batch, mel_batch, frame_batch, coords_batch def _load(checkpoint_path, device): if device == 'cuda': checkpoint = torch.load(checkpoint_path) else: checkpoint = torch.load(checkpoint_path, map_location=lambda storage, loc: storage) return checkpoint def load_lip_model(path="checkpoints/wav2lip_gan.pth", device="cuda"): model = Wav2Lip() print("Load checkpoint from: {}".format(path)) checkpoint = _load(path, device) s = checkpoint["state_dict"] new_s = {} for k, v in s.items(): new_s[k.replace('module.', '')] = v model.load_state_dict(new_s) model = model.to(device) return model.eval() def load_input_image_or_video(file_path, resize_factor=2, logging=False): if not os.path.isfile(file_path): print(file_path) raise ValueError('input image or video must be a valid path to video/image file') elif file_path.split('.')[1] in ['jpg', 'png', 'jpeg']: full_frames = [cv2.imread(file_path)] fps = 25 else: video_stream = cv2.VideoCapture(file_path) fps = video_stream.get(cv2.CAP_PROP_FPS) print('Reading video frames...') full_frames = [] while 1: still_reading, frame = video_stream.read() if not still_reading: video_stream.release() break if resize_factor > 1: frame = cv2.resize(frame, (frame.shape[1]//resize_factor, frame.shape[0]//resize_factor)) full_frames.append(frame) #print("Number of frames available for inference: "+str(len(full_frames))) return full_frames, fps def load_input_audio(file_path, fps, results_path): mel_step_size = 16 if not file_path.endswith('.wav'): temp_file = f"{os.path.dirname(results_path)}/temp.wav" command = 'ffmpeg -y -i {} -strict -2 {}'.format(file_path, temp_file) subprocess.call(command, shell=True, stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL) file_path = temp_file wav = audio.load_wav(file_path, sr=16000) mel = audio.melspectrogram(wav) if np.isnan(mel.reshape(-1)).sum() > 0: raise ValueError('Mel contains nan! Using a TTS voice? Add a small epsilon noise to the wav file and try again') mel_chunks = [] mel_idx_multiplier = 80./fps i = 0 while 1: start_idx = int(i * mel_idx_multiplier) if start_idx + mel_step_size > len(mel[0]): mel_chunks.append(mel[:, len(mel[0]) - mel_step_size:]) break mel_chunks.append(mel[:, start_idx : start_idx + mel_step_size]) i += 1 #print("Length of mel chunks: {}".format(len(mel_chunks))) return mel_chunks, file_path def animate_input(frames, audio, audio_file, fps, model, device, results_path): temp_path = f"{os.path.dirname(results_path)}" full_frames = frames[:len(audio)] batch_size = 128 gen = datagen(full_frames.copy(), audio, batch_size, device) for i, (img_batch, mel_batch, frames, coords) in enumerate(gen): if i == 0: temp_video = temp_path + "/result.avi" frame_h, frame_w = full_frames[0].shape[:-1] out = cv2.VideoWriter(temp_video, cv2.VideoWriter_fourcc(*'DIVX'), fps, (frame_w, frame_h)) img_batch = torch.FloatTensor(np.transpose(img_batch, (0, 3, 1, 2))).to(device) mel_batch = torch.FloatTensor(np.transpose(mel_batch, (0, 3, 1, 2))).to(device) with torch.no_grad(): pred = model(mel_batch, img_batch) pred = pred.cpu().numpy().transpose(0, 2, 3, 1) * 255. for p, f, c in zip(pred, frames, coords): y1, y2, x1, x2 = c p = cv2.resize(p.astype(np.uint8), (x2 - x1, y2 - y1)) f[y1:y2, x1:x2] = p out.write(f) out.release() command = 'ffmpeg -y -i {} -i {} -strict -2 -q:v 1 {}'.format(audio_file, temp_video, results_path) subprocess.call(command, shell=platform.system() != 'Windows', stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL)