`stages.video.filtering.motion_filter`#

Module Contents#

Classes#

`MotionFilterStage`	Stage for filtering video clips based on motion score.
`MotionVectorDecodeStage`	Stage for decoding motion vector information from video files.

API#

class stages.video.filtering.motion_filter.MotionFilterStage#

Bases: nemo_curator.stages.base.ProcessingStage[nemo_curator.tasks.video.VideoTask, nemo_curator.tasks.video.VideoTask]

Stage for filtering video clips based on motion score.

This class processes video clips through a series of steps including motion score computation and filtering based on thresholds.

global_mean_threshold: float#: 0.00098

inputs() → tuple[list[str], list[str]]#

Define stage input requirements.

Returns (tuple[list[str], list[str]]): Tuple of (required_attributes, required_columns) where: - required_top_level_attributes: List of task attributes that must be present - required_data_attributes: List of attributes within the data that must be present

motion_filter_batch_size: int#: 256

name: str#: ‘motion_filter’

num_gpus_per_worker: float#: 0

outputs() → tuple[list[str], list[str]]#

Define stage output specification.

Returns (tuple[list[str], list[str]]): Tuple of (output_attributes, output_columns) where: - output_top_level_attributes: List of task attributes this stage adds/modifies - output_data_attributes: List of attributes within the data that this stage adds/modifies

per_patch_min_256_threshold: float#: 1e-06

process( task: nemo_curator.tasks.video.VideoTask, ) → nemo_curator.tasks.video.VideoTask#: Process a task and return the result. Args: task (X): Input task to process Returns (Y | list[Y]): - Single task: For 1-to-1 transformations - List of tasks: For 1-to-many transformations (e.g., readers) - None: If the task should be filtered out

score_only: bool#: False

verbose: bool#: False

class stages.video.filtering.motion_filter.MotionVectorDecodeStage#