gnes.preprocessor.video.ffmpeg module

class gnes.preprocessor.video.ffmpeg.FFmpegPreprocessor(frame_size='192:168', frame_rate=10, frame_num=-1, duplicate_rm=True, use_phash_weight=False, phash_thresh=5, *args, **kwargs)

Bases: gnes.preprocessor.base.BaseVideoPreprocessor

apply(doc)

Return type:None

duplicate_rm_hash(images)

Return type:List[ndarray]

static pic_weight(images)

Return type:List[float]

train(*args, **kwargs)

Train the model, need to be overrided

class gnes.preprocessor.video.ffmpeg.FFmpegVideoSegmentor(frame_size='192:168', frame_rate=10, frame_num=-1, segment_method='cut_by_frame', segment_interval=-1, segment_num=3, max_frames_per_doc=-1, use_image_input=False, splitter='__split__', *args, **kwargs)

Bases: gnes.preprocessor.base.BaseVideoPreprocessor

apply(doc)

Return type:None

train(*args, **kwargs)

Train the model, need to be overrided

class gnes.preprocessor.video.ffmpeg.GifChunkPreprocessor(uniform_doc_weight=True, *args, **kwargs)

Bases: gnes.preprocessor.base.RawChunkPreprocessor, gnes.preprocessor.base.BaseVideoPreprocessor

train(*args, **kwargs)

Train the model, need to be overrided