def generate_align_hash(self, videos: list) -> {str: Align}:
align_hash = {}
for path in videos:
video_name = get_file_name(path)
align_path = os.path.join(self.aligns_path, video_name) + '.align'
align_hash[video_name] = align_from_file(align_path,
self.max_string)
return align_hash
def extract_video_data(path: str,
detector,
predictor,
verbose: bool = True) -> Optional[np.ndarray]:
if verbose: print('\n{}'.format(path))
video_data = skvideo.io.vread(path)
video_data_len = len(video_data)
# if video_data_len != env.FRAME_COUNT:
# print(ERROR_LOG + 'Wrong number of frames: {}'.format(video_data_len))
# return None
mouth_data = []
bar = ShadyBar(
get_file_name(path) + '.npy',
max=video_data_len,
suffix='%(percent)d%% [%(elapsed_td)s]') if verbose else None
countOfNone = 1
for i, f in enumerate(video_data):
c = extract_mouth_on_frame(f, detector, predictor, i)
if c is None: return None
mouth_data.append(c)
# if c is None:
# countOfNone+=1
# else:
# while countOfNone>=1:
# mouth_data.append(c)
# countOfNone -=1
if verbose and bar: bar.next()
mouth_data = np.array(mouth_data)
if verbose and bar: bar.finish()
return mouth_data
def get_data_from_path(self,
path: str) -> (np.ndarray, str, np.ndarray, int):
align = self.align_hash[get_file_name(path)]
return get_video_data_from_file(
path), align.sentence, align.labels, align.length
def extract_to_npy(videos_path: str, output_path: str, predictor_path: str,
pattern: str, first_video: int, last_video: int):
print('\nEXTRACT\n')
print('Searching for files in: {}'.format(videos_path))
print('Using predictor at: {}'.format(predictor_path))
print('Match for: {}\n'.format(pattern))
videos_failed = []
had_errors = False
videos_failed_path = os.path.join(output_path, 'videos_failed.log')
if is_file(videos_failed_path):
with open(videos_failed_path, 'r+') as f:
for line in f.readlines():
videos_failed.append(line.rstrip())
last_group_dir_name = ''
video_count_per_group = 0
detector = dlib.get_frontal_face_detector()
predictor = dlib.shape_predictor(predictor_path)
for file_path in get_files_in_dir(videos_path, pattern):
group_dir_name = os.path.basename(os.path.dirname(file_path))
video_file_name = get_file_name(file_path)
video_target_dir = os.path.join(output_path, group_dir_name)
video_target_path = os.path.join(video_target_dir,
video_file_name) + '.npy'
if video_target_path in videos_failed:
print(Fore.RED +
'Video {} is probably corrupted and was ignored'.format(
video_file_name))
continue
if group_dir_name == last_group_dir_name:
video_count_per_group += 1
else:
video_count_per_group = 0
last_group_dir_name = group_dir_name
if video_count_per_group < first_video or video_count_per_group >= last_video:
continue
if is_file(video_target_path):
print(Style.DIM + Fore.CYAN + 'Video {} is already at: {}'.format(
video_file_name, video_target_path))
continue
make_dir_if_not_exists(video_target_dir)
if not video_to_frames(file_path, video_target_path, detector,
predictor):
had_errors = True
if video_count_per_group > 0:
video_count_per_group -= 1
else:
video_count_per_group = 0
with open(videos_failed_path, 'a+') as f:
f.write(video_target_path + '\n')
if had_errors:
print(Fore.YELLOW + '\nExtraction finished with some errors')
else:
print(Fore.GREEN + '\nExtraction finished!')