Files
lerobot/src/lerobot/datasets/pyav_utils.py
T
Caroline Pascal 3dd19d043e feat(depth maps): adding support for depth in LeRobot (#3644)
* feat(depth): add depth quantization helpers and tests

* feat(video): add ffv1 to supported codecs

* feat(depth): persist depth metadata

* feat(depth): extend quantization tools to better fit the encoding/decoding pipeline

* feat(depth): plumb DepthEncoderConfig through LeRobotDataset and DatasetWriter

* feat(depth): wire StreamingVideoEncoder + writer to depth encoder

* feat(depth): wire DatasetReader to decode_depth_frames

* feat(cameras/realsense): expose async depth in metric meters

* feat(features): route 2D camera shapes to observation.depth.<key>

* feat(robots/so_follower): emit + populate depth keys when use_depth

* feat(record): plumb DepthEncoderConfig through lerobot-record

* feat(viz): render depth observations as rr.DepthImage in Viridis

* feat(depth maps writer): adding support for raw depth maps recording with image writer

* chore(format): format code

* feat(depth shape): ensuring depth maps shape is always including the channel

* feat(is_depth): simplifying is_depth nested name + legacy support

* fix(stop_event): fixing stop_event race condition in camera classes

* fix(plumbing): fixing missing parts in the depth maps pipeline

* chore(typos): fixing typos

* test(fix): fixing exisiting tests to still work with latest features

* tests(depth): adding new tests for depth integration validation

* feat(pix_fmt channels): use PyAv to check get pixel formats number of channels

* feat(refactor): refactor DepthEncoderConfig quantization pipeline, so that the methods do not live in the config class. Add pixel format - channels validation.Move the default pixel format for depth in the config file.

* fix(pre-commit): fixing mutable defautl value

* fix(info): fixing info metadata update when is_depth_map was set

* tests(typos): fixing typos in tests

* fix(realsense): fixing typo in realsense serial number

* fix(normalization): restricting 255 normalization to non depth/uint8 images only

* fix(typo): fixing typo

* fix(TIFF): add missing quantization and cleanup for TIFF files

* feat(batched dequantization): optimizing dequantize_depth for torch based batched dequantization

* feat(tools): adding depth support in LeRobotDataset edition tools

* test(aggregate): extending aggregation tests to depth frames

* test(cleaning): cleaning up tests

* fix(from_video_info): fixing early validation issue in from_video_info

* fix(typo): fixing typo

* fix(is_depth): adding missing doctrings and is_depth arguments in video decoding functions

Co-authored-by: Wensi (Vince) Ai <59036629+wensi-ai@users.noreply.github.com>

* fix(depth units): fixing depth units output for the realsense cameras

* feat(output unit): adding support for output unit specification at dataset reading/training time

Co-authored-by: Wensi (Vince) Ai <59036629+wensi-ai@users.noreply.github.com>

* test(depth): cleaning up depth tests

* test(depth encoding): updating and cleaning video/depth encoding tests

* chore(format): formatting code

* docs(depth): improving depth maps docs

* test(fix): fixing depth tests

* test(dataset tools): adding missing tests for new dataset edition tools features

* chore(format): formatting code

* fix(pyav check): fixing PyAV option validation for integer codec options by normalizing
numeric values before calling `is_integer()`

Co-authored-by: Wensi (Vince) Ai <59036629+wensi-ai@users.noreply.github.com>

* docs(mermaid): fixing mermaid diagram

* fix(rebase): rebase follow up corrections

* feat(dataset tools): adding missing docstrings and features for depth fill support in dataset edition tools

* docs(docstring): updating docstrings

* docs(dataset tools): updating docs

* fix(save images): fixing image saving in dataset tools

* fix(update video info): fixing update video info logic to match the recording and editing use cases

* test(reencode): fixing reencoding monkeypatch

* fix(review): add Claude review

* chore(format): format code

* fix(update video info): ditching the differentiated approahces for video info update - video info are always updated unless for preserved keys.

* chore(rebase): fixing rebase merge conflicts

* test(visualization): fixing visualization tests

* feat(docstrings): adding explicit docstring for encoding parameters. Docstrigns will now show up as description in the CLI --help.

* feat(mm as default): adding a global DEFAULT_DEPTH_UNIT variable setting mm as default depth unit

* fix(RGB <-> camera): renaming camera_encoder to rgb_encoder for clarity

* chore(TODO): removing deprecated TODO

* doc(write_u16_plane): improving docstrings for write_u16_plane

* feat(units): adding constants for depth frames units (m and mm)

* fix(spam): replacing spamming warning but a debug log

* feat(leagcy metadata): adding automatic metadata update for legacy 'video.is_depth_map' feature

* fix(copy&reindex): fixing metadat reshaping for single channel frames

* fix(ImageNet): excluding dpeth frames from ImageNet stats

* fix(PyAV container seek): fixing initial  PyAV container seek to be robust againsy codec choice

* feat(lerobot-dataset-viz): adding support for depth in lerobot-dataset-viz

* fix(compress): removing rerun compression for DepthImages

* fix(signle channel squeeze): fixing single channel squeezing

* chore(format): format code

* fix(streaming): adding support for dequantization in streaming_dataset.py

* refactor(read depth): factorizing depth reading methods for realsense camera and adding support for depth-only usage

* chore(renaming): fixing missed RGBEncoderConfig renamings

* docs(renaming): reflecting renamings in a clearer way in the docs

* chore(annotation): excluding depth from the annotation pipeline

* feat(robots): adding depth support in compatible follower robots

* feat(LeSadKiwi): excluding LeKiwi from depth support (for now)

* chore(fail): removing misplaced file

* chore(fail): removing misplaced file

* fix(remove ffv1): removing ffv1 as it does not support MP4

* docs(cheat sheet): adding depth and video encoding to the cheat sheet

* fix(lossless): tuning depth encoding parameters for lossless depth storage

* test(fix): fixing failing tests

* depth(ZMQ): excluding ZMQ from depth support

* Revert "depth(ZMQ): excluding ZMQ from depth support"

This reverts commit b95cf4e4c2.

* fix(image transforms): excluding depth frames from images transforms

* fix(typo): typo

* fix(stats): fixing stats computation for depth frames

* fix(TIFF vs. pytorch): adding an extra uint16 to float32 conversion for depth maps stored as raw TIFF images

* fix(typos): fixing typos

* test(dtype): fixing stats computation typing tests

---------

Signed-off-by: Steven Palma <imstevenpmwork@ieee.org>
Co-authored-by: Wensi (Vince) Ai <59036629+wensi-ai@users.noreply.github.com>
Co-authored-by: Steven Palma <imstevenpmwork@ieee.org>
Co-authored-by: Wensi Ai <wsai@stanford.edu>
2026-06-27 14:21:21 +02:00

222 lines
8.5 KiB
Python

#!/usr/bin/env python
# Copyright 2026 The HuggingFace Inc. team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""PyAV-based compatibility checks for :class:`VideoEncoderConfig`.
Centralises all :mod:`av` introspection of the bundled FFmpeg build.
Checks degrade to a no-op when the target codec isn't available locally.
"""
import functools
import logging
from typing import Any
import av
import numpy as np
logger = logging.getLogger(__name__)
FFMPEG_NUMERIC_OPTION_TYPES = ("INT", "INT64", "UINT64", "FLOAT", "DOUBLE")
FFMPEG_INTEGER_OPTION_TYPES = ("INT", "INT64", "UINT64")
def write_u16_plane(plane: av.video.plane.VideoPlane, src: np.ndarray, fill_value: int | None = None) -> None:
"""Copy a 2D ``uint16`` image into the plane's memory buffer, row by row.
For speed, each row is padded to a wider size than ``width``, so the true row width in
memory is ``plane.line_size`` (bytes), not ``width``. Copying as one straight stream
would skew the image, so we write only the first ``width`` columns of each row and
leave the padding untouched.
Args:
plane: Destination 16-bit plane.
src: Source image, shape ``(height, width)``, dtype ``uint16``.
fill_value: If given, every pixel (padding included) is set to this first, so the
padding holds clean data instead of garbage.
"""
height, width = src.shape
stride_u16 = plane.line_size // np.dtype(np.uint16).itemsize
dst = np.frombuffer(plane, dtype=np.uint16).reshape(height, stride_u16)
if fill_value is not None:
dst.fill(fill_value)
dst[:, :width] = src
@functools.cache
def get_pix_fmt_channels(pix_fmt: str) -> int:
"""Return the number of components (channels) for *pix_fmt*."""
return len(av.VideoFormat(pix_fmt).components)
@functools.cache
def get_codec(vcodec: str) -> av.codec.Codec | None:
"""PyAV write-mode ``Codec`` for *vcodec*, or ``None`` if unavailable."""
try:
return av.codec.Codec(vcodec, "w")
except Exception:
return None
@functools.cache
def _get_codec_options_by_name(vcodec: str) -> dict[str, av.option.Option]:
"""Private-option name → PyAV ``Option`` for *vcodec* (empty if unavailable)."""
codec = get_codec(vcodec)
if codec is None:
return {}
return {opt.name: opt for opt in codec.descriptor.options}
@functools.cache
def _get_codec_video_formats(vcodec: str) -> tuple[str, ...]:
"""Pixel formats accepted by *vcodec* in PyAV's preferred order (empty if unknown)."""
codec = get_codec(vcodec)
if codec is None:
return ()
return tuple(fmt.name for fmt in (codec.video_formats or []))
def detect_available_encoders_pyav(encoders: list[str] | str) -> list[str]:
"""Return the subset of *encoders* available as video encoders in the local FFmpeg build.
Each name is probed directly via :func:`get_codec`; input order is preserved.
"""
if isinstance(encoders, str):
encoders = [encoders]
available: list[str] = []
for name in encoders:
codec = get_codec(name)
if codec is not None and codec.type == "video":
available.append(name)
else:
logger.debug("encoder '%s' not available as video encoder", name)
return available
def _check_option_value(vcodec: str, label: str, value: Any, opt: av.option.Option) -> None:
"""Range-check numeric *value* and choice-check string *value* against *opt*."""
type_name = opt.type.name
if type_name in FFMPEG_NUMERIC_OPTION_TYPES:
if isinstance(value, bool):
raise ValueError(
f"{label}={value!r} is not numeric; codec {vcodec!r} expects a number for this option."
)
elif isinstance(value, str):
try:
num_val = float(value)
except ValueError as e:
raise ValueError(
f"{label}={value!r} is not numeric; codec {vcodec!r} expects a number for this option."
) from e
elif isinstance(value, (float, int)):
num_val = float(value)
else:
raise ValueError(
f"{label}={value!r} is not numeric; codec {vcodec!r} expects a number for this option."
)
# Check integer type compatibility
if type_name in FFMPEG_INTEGER_OPTION_TYPES and not num_val.is_integer():
raise ValueError(
f"{label}={num_val!r} must be an integer for codec {vcodec!r} "
f"(FFmpeg option {opt.name!r} is {type_name}); float values are not allowed."
)
# Check numeric range compatibility
lo, hi = float(opt.min), float(opt.max)
if lo < hi and not (lo <= num_val <= hi):
raise ValueError(
f"{label}={num_val} is out of range for codec {vcodec!r}; must be in [{lo}, {hi}]"
)
elif type_name == "STRING":
if isinstance(value, bool):
raise ValueError(f"{label}={value!r} is not a valid string value for codec {vcodec!r}.")
if isinstance(value, str):
str_val = value
elif isinstance(value, (int, float)):
str_val = str(value)
else:
raise ValueError(f"{label}={value!r} has unsupported type for STRING option on codec {vcodec!r}")
# Check string choice compatibility
choices = [c.name for c in (opt.choices or [])]
if choices and str_val not in choices:
raise ValueError(
f"{label}={str_val!r} is not a supported choice for codec "
f"{vcodec!r}; valid choices: {choices}"
)
else:
return
def _check_pixel_format(vcodec: str, pix_fmt: str) -> None:
formats = _get_codec_video_formats(vcodec)
if formats and pix_fmt not in formats:
raise ValueError(
f"pix_fmt={pix_fmt!r} is not supported by codec {vcodec!r}; "
f"supported pixel formats: {list(formats)}"
)
def _check_pix_fmt_channels(pix_fmt: str, channels: int) -> None:
"""Ensure *pix_fmt* can carry at least *channels* components."""
pix_fmt_channels = get_pix_fmt_channels(pix_fmt)
if pix_fmt_channels < channels:
raise ValueError(
f"pix_fmt={pix_fmt!r} carries only {pix_fmt_channels} component(s) "
f"but the source data has {channels} channel(s)."
)
def _check_codec_options(vcodec: str, codec_options: dict[str, Any]) -> None:
"""Validate merged encoder options (typed) against the codec's published AVOptions."""
supported_options = _get_codec_options_by_name(vcodec)
for key, value in codec_options.items():
# GOP size is not a codec-specific option, it has to be validated separately.
if key == "g":
if isinstance(value, bool) or not isinstance(value, int) or value < 1:
raise ValueError(f"g={value!r} must be a positive integer for codec {vcodec!r}")
continue
if key not in supported_options:
continue
_check_option_value(vcodec, key, value, supported_options[key])
def check_video_encoder_parameters_pyav(
vcodec: str,
pix_fmt: str,
codec_options: dict[str, Any],
channels: int | None = None,
) -> None:
"""Verify *config* is compatible with the bundled FFmpeg build.
Checks pixel format, abstract tuning-field compatibility, and each merged
encoder option from :meth:`~lerobot.configs.video.VideoEncoderConfig.get_codec_options`
against PyAV (including numeric ``extra_options`` present in that dict).
When given, additionally verify that *pix_fmt* carries as many components as the source data channels.
No-op when ``config.vcodec`` isn't in the local FFmpeg build.
Raises:
ValueError: on the first incompatibility encountered.
"""
options = _get_codec_options_by_name(vcodec)
if not options:
raise ValueError(f"Codec {vcodec!r} is not available in the bundled FFmpeg build")
_check_pixel_format(vcodec, pix_fmt)
if channels is not None:
_check_pix_fmt_channels(pix_fmt, channels)
_check_codec_options(vcodec, codec_options)