4.6.6 pde.tools.ffmpeg module

Functions for interacting with FFmpeg

`FFmpegFormat`	defines a FFmpeg format used for storing field data in a video
`formats`	dict of pre-defined `FFmpegFormat` formats
`find_format`	find a defined FFmpegFormat that satisifies the requirements

class FFmpegFormat(pix_fmt_file, pix_fmt_data, channels, bits_per_channel, dtype, codec='ffv1')[source]

Bases: object

defines a FFmpeg format used for storing field data in a video

Note

All pixel formats supported by FFmpeg can be obtained by running ffmpeg -pix_fmts. However, not all pixel formats are supported by all codecs. Supported pixel formats are listed in the output of ffmpeg -h encoder=<ENCODER>, where <ENCODER> is one of the encoders listed in ffmpeg -codecs.

Parameters:

pix_fmt_file (str)
pix_fmt_data (str)
channels (int)
bits_per_channel (int)
dtype (dtype[Any] | None | type[Any] | _SupportsDType[dtype[Any]] | str | tuple[Any, int] | tuple[Any, SupportsIndex | Sequence[SupportsIndex]] | list[Any] | _DTypeDict | tuple[Any, Any])
codec (str)

bits_per_channel: int

number of bits per color channel in this pixel format

Type:: int

property bytes_per_channel: int

number of bytes per color channel

Type:: int

channels: int

number of color channels in this pixel format

Type:: int

codec: str = 'ffv1'

name of the codec that supports this pixel format

Type:: str

data_from_frame(frame_data)[source]

converts data stored in a color channel to normalized data

Parameters:: frame_data (ndarray)

data_to_frame(normalized_data)[source]

converts normalized data to data being stored in a color channel

Parameters:: normalized_data (ndarray)
Return type:: ndarray

dtype: dtype[Any] | None | type[Any] | _SupportsDType[dtype[Any]] | str | tuple[Any, int] | tuple[Any, SupportsIndex | Sequence[SupportsIndex]] | list[Any] | _DTypeDict | tuple[Any, Any]: numpy dtype corresponding to the data of a single channel

property max_value: float | int: maximal value stored in a color channel

pix_fmt_data: str

name of the pixel format used in the frame data

Type:: str

pix_fmt_file: str

name of the pixel format used in the codec

Type:: str

find_format(channels, bits_per_channel=8)[source]

find a defined FFmpegFormat that satisifies the requirements

Parameters:

channels (int) – Minimal number of color channels
bits_per_channel (int) – Minimal number of bits per channel

Returns:

Identifier for a format that satisifies the requirements (but might have more channels or more bits per channel then requested. None is returned if no format can be identified.

Return type:

str

formats = {'gbrap16le': FFmpegFormat(pix_fmt_file='gbrap16le', pix_fmt_data='gbrap16le', channels=4, bits_per_channel=16, dtype=dtype('uint16'), codec='ffv1'), 'gbrp16le': FFmpegFormat(pix_fmt_file='gbrp16le', pix_fmt_data='gbrp16le', channels=3, bits_per_channel=16, dtype=dtype('uint16'), codec='ffv1'), 'gray': FFmpegFormat(pix_fmt_file='gray', pix_fmt_data='gray', channels=1, bits_per_channel=8, dtype=<class 'numpy.uint8'>, codec='ffv1'), 'gray16le': FFmpegFormat(pix_fmt_file='gray16le', pix_fmt_data='gray16le', channels=1, bits_per_channel=16, dtype=dtype('uint16'), codec='ffv1'), 'rgb24': FFmpegFormat(pix_fmt_file='rgb24', pix_fmt_data='rgb24', channels=3, bits_per_channel=8, dtype=<class 'numpy.uint8'>, codec='ffv1'), 'rgb32': FFmpegFormat(pix_fmt_file='rgb32', pix_fmt_data='rgb32', channels=4, bits_per_channel=8, dtype=<class 'numpy.uint8'>, codec='ffv1')}: dict of pre-defined FFmpegFormat formats