nemo_curator.stages.video.clipping.video_frame_extraction

Module Contents

Classes

Name	Description
`VideoFrameExtractionStage`	Stage that extracts frames from videos into numpy arrays.

Functions

Name	Description
`get_frames_from_ffmpeg`	Fetch resized frames for video.

API

class nemo_curator.stages.video.clipping.video_frame_extraction.VideoFrameExtractionStage(
    output_hw: tuple[int, int] = (27, 48),
    pyncv_batch_size: int = 64,
    decoder_mode: str = 'pynvc',
    verbose: bool = False,
    name: str = 'video_frame_extraction'
)

Dataclass

Bases: ProcessingStage[VideoTask, VideoTask]

Stage that extracts frames from videos into numpy arrays.

This stage handles video frame extraction using either FFmpeg (CPU/GPU) or PyNvCodec, converting video content into standardized frame arrays for downstream processing.

decoder_mode

str = 'pynvc'

name

str = 'video_frame_extraction'

output_hw

tuple[int, int] = (27, 48)

pyncv_batch_size

int = 64

verbose

bool = False

nemo_curator.stages.video.clipping.video_frame_extraction.VideoFrameExtractionStage.__post_init__() -> None

nemo_curator.stages.video.clipping.video_frame_extraction.VideoFrameExtractionStage.inputs() -> tuple[list[str], list[str]]

nemo_curator.stages.video.clipping.video_frame_extraction.VideoFrameExtractionStage.outputs() -> tuple[list[str], list[str]]

nemo_curator.stages.video.clipping.video_frame_extraction.VideoFrameExtractionStage.process(
    task: nemo_curator.tasks.video.VideoTask
) -> nemo_curator.tasks.video.VideoTask

nemo_curator.stages.video.clipping.video_frame_extraction.VideoFrameExtractionStage.setup(
    worker_metadata: nemo_curator.backends.base.WorkerMetadata | None = None
) -> None

Setup method called once before processing begins. Override this method to perform any initialization that should happen once per worker. Args: worker_metadata (WorkerMetadata, optional): Information about the worker (provided by some backends)

nemo_curator.stages.video.clipping.video_frame_extraction.get_frames_from_ffmpeg(
    video_file: pathlib.Path,
    width: int,
    height: int,
    use_gpu: bool = False
) -> numpy.typing.NDArray[numpy.uint8] | None

Fetch resized frames for video.