Skip to content

Latest commit

 

History

History
338 lines (265 loc) · 12.5 KB

CHANGELOG.md

File metadata and controls

338 lines (265 loc) · 12.5 KB

📈 Changelog

Version 0.6.0 (2025-1-25)

New Audio API in VideoReader

Retrieve Audio Data as a Tensor or File

A new .audio property has been added to VideoReader, allowing direct access to the Audio object:

reader = VideoReader("test.mp4")
if reader.has_audio:
    audio = reader.audio  # Access the audio object
    tensor = audio.tensor()  # Retrieve audio as a PyTorch tensor, 1D. NOTE. THIS HAS BEEN MINIMALLY TESTED
    success = audio.file("output.wav")  # Extract audio to a WAV file

Audio Class Features

  • tensor() – Extracts the audio stream as a PyTorch tensor.
  • file(output_path) – Saves the audio to a specified file path.
  • Read-only Properties:
    • sample_rate: Audio sample rate in Hz.
    • channels: Number of audio channels.
    • bit_depth: Bit depth of the audio.
    • codec: Audio codec format.
    • bitrate: Audio bitrate.

Example usage:

print(audio.sample_rate)  # Get the sample rate
print(audio.channels)  # Number of channels

Version 0.5.8.5 (2025-1-24)

  • Fixed an issue with ranges being off by 1 frame.

Version 0.5.8 (2025-1-24)

Improved Property Access for VideoReader

  • Added direct property access for video metadata.

  • Users can now retrieve video properties directly instead of accessing properties["key"].

  • Example:

    reader = VideoReader("test.mp4")
print(reader.width)  # Instead of video.properties["width"]
print(reader.fps)  # Instead of video.properties["fps"]

  • Available properties:
    width, height, fps, min_fps, max_fps, duration, total_frames, pixel_format, has_audio, audio_bitrate, audio_channels, audio_sample_rate, audio_codec, bit_depth, aspect_ratio, codec

__getitem__ Seeking Behavior Update

  • __getitem__ now only accepts seconds (float) for seeking.
  • Frame-based seeking (int) is currently not supported via __getitem__.
  • Example usage: 
    reader = VideoReader("test.mp4")
frame = reader[2.5]  # Seeks to 2.5 seconds into the video

Version 0.5.7 (2025-1-23)

  • New color format support
    Added CPU-based conversions for:

    • 12-bit YUV (420, 422, and 444)
    • 10-bit YUV444
    • ProRes 4444 with alpha (10-bit or 12-bit)
    • Anything NOT 8-bit uses uint-16 tensors.

  • Notes on rawvideo
    Raw .yuv files still require specifying resolution and pixel format manually.
    (-f rawvideo -pix_fmt yuv420p -s 1920x1080 …)

    • IN PROGRESS**

  • Other improvements

    • Minor cleanups in converter classes.
    • Updated tests to cover the newly supported pixel formats.

Version 0.5.6.1 (2025-1-23)

= Adusted __call__ method handling. - Int values seek as frames, Float values seek as times.

  • Added several tests against OpenCV and FFMPEG for confirmation on frame/time ranges.
  • Added tests for color formats.

Version 0.5.6 (2025-1-22)

= Removed CUDA Dependency in favor of CPU decoding. (It's faster anyways.)

  • Updated repo + docs
  • Tested and adjusted time range-- should match OpenCV behavior in all cases. = Looking into color space differences...

Version 0.5.4 (2024-12-27)

  • Added Support for YUV422P10LE to RGB48 conversion. (CPU)
  • Added Support for GBRP Conversion. (CPU)
  • Added New Argument, tensor_shape
    • Default is HWC, but can be set to CHW, etc.

Version 0.5.3 (2024-11-10)

  • Fixed issue with floating point ranges.
  • Fixed/added quicktime and RGB24 support.

Version 0.5.2 (2024-11-05)

  • Finalized fixes for CPU Color Conversions.
  • Added Support for RGBA, BGRA pixel formats.
  • Adjusted __call__ and set_range methods for VideoReader.
    • Now takes int for frame steps, float for timestamp steps.
      • Pass int or float and the reader will handle things internally.
  • No new benchmarks. Not need for this release.

Version 0.5.1.2 (2024-11-05)

  • Fixed issue with Color Conversion.

Version 0.5.1.1 (2024-11-05)

  • Testing out use of timestamps for setting range.

Version 0.5.1 (2024-11-04)

  • Fixed an issue where if no filters were added, decoder would not run properly.

Version 0.5.0 (2024-11-03)

  • Some Major Refactoring and changes made.
    • Parsed and created Filter classes for every (video) Ffmpeg filter.
      • Filters defined within Celux.pyi
        • Not all are tested. For Full documentation of arguments and usage, see: ffmpeg-filter-docs
        • Please create a new issue if any problems occur!
    • Fixed an issue with Filter initialization and unwanted output messages.
    from celux import VideoReader, Scale #, CreateFilter, FilterType


scale_filter = Scale(width = "1920", height = "1080")

# scale_filter = CreateFilter(FilterType.Scale)
# scale_filter.setWidth("1920")
# scale_filter.setHeight"1080")
# scale_filter.setFlags("bicubic")

with VideoReader("input.mp4", device = "cpu", filters = [scale_filter]) as reader:
  for frame in reader:
    # will be a scaled frame

Version 0.4.5.5 (2024-10-30)

  • Added some safety checks for filters.
    • Fixed issue that occurs when using scale.

Version 0.4.5 (2024-10-29)

  • Implemented filters for cpu usage.
    • usage should be familiar to those who've used ffmpeg:
filters = [("scale", "1280:720"), ("hue", "0.5")]
reader = cx.VideoReader("/path/to/input", device = "cpu", filters = filters)

Version 0.4.4 (2024-10-29)

  • Removed Stream Parameter in VideoReader: The VideoReader no longer accepts an external CUDA stream.
  • Introduced event-based synchronization between frame reading operations to ensure proper and consistent output.
  • Use of nvdec directly.

Version 0.4.3.5 (2024-10-29)

  • Testing some changes, partial release, may end up reverting.
  • Use nvdec directly instead of _cuvid.
  • Some small refactoring and testing, nothing major.

Version 0.4.3 (2024-10-29)

  • New Features:
    • Added num_threads arg to control decoder threads internally used.
    • Fixed VideoReader() calls, now properly sets frame range.
    • Potentially fixed issue with cuda synchronizations.

Version 0.4.2 (2024-10-28)

  • Focus on VideoReader:

    • Removed VideoWriter to streamline the library and enhance focus on reading capabilities.
    • Fixed call method of VideoReader, now properly seeks frames.

  • New Features:

    • Added __getitem__ method to VideoReader for easier access to properties, allowing users to retrieve metadata using dictionary-like syntax (e.g., reader['width']).
    • Expanded VideoReader.get_properties() to include new metadata properties:
      • codec: The name of the codec being used.
      • bit_depth: The bit-depth of the video.
      • has_audio: Indicates whether the video contains an audio track.
      • audio_bitrate: Bitrate of the audio stream.
      • audio_channels: Number of audio channels.
      • audio_sample_rate: Sample rate of the audio stream.
      • audio_codec: Codec used for audio.
      • min_fps: Minimum frames per second of the video.
      • max_fps: Maximum frames per second of the video.
      • aspect_ratio: Aspect ratio of the video.

  • New Converter Formats:

    • Completed the implementation of the following converters to support new video formats:
      • YUV420P to RGB
      • YUV420P10LE to RGB48
      • BGR to RGB
      • RGB to RGB
      • P010LE to RGB48

  • Supported Codecs:

    • The following codecs can be worked with using the VideoReader, based on supported pixel formats:
      • H.264 (AVC): YUV420P, YUV420P10LE
      • H.265 (HEVC): YUV420P, YUV420P10LE
      • VP8/VP9: YUV420P, YUV420P10LE
      • AV1: YUV420P, YUV420P10LE
      • MPEG-2: YUV420P
      • ProRes: YUV420P, YUV422, YUV444
      • DNxHD/DNxHR: YUV422, YUV444
      • DV (Digital Video): YUV420P
      • Uncompressed RGB: RGB, BGR
      • P010LE: P010LE

  • Testing Improvements:

    • Updated tests to ensure compatibility with various bit-depths and codec types.
    • Added tests to verify the correct functionality of the new features and converters.

Version 0.4.0 (2024-10-23)

  • Moved to FFmpeg static libraries!
    • Startup times are improved. All libs that can be static, are static.
  • Adjusted logging to flow a little bit better, not overcrowd console unless desired.
    • Logging details more info on codecs. The Decoder selects the BEST codec for the video.
  • Need to investigate if NVDEC is bottlenecked, or I've reached max performance capabilities.
    • It is curious that cpu benches at 1859 fps and gpu benches at 1809 fps.

Version 0.3.9 (2024-10-21)

  • Pre-Release Update:
    • Prep for 0.4.0 release.
      • 0.4.x release will be characterized by new codec and pixel format support!
      • Removed d_type and buffer_size arguments from VideoReader and VideoWriter.
        • Output and Input tensors are now, by standard, UINT8, HWC format, [0,255].
      • Standardized to YUV420P for now.
      • Swapped custom CUDA kernels for nppi.
      • various cleanup and small refactorings.

Version 0.3.8 (2024-10-21)

  • Pre-Release Update:
    • Removed Buffering from VideoWriter, resulting in INSANE performance gains.
    • Fixed threading issue with VideoWriter, now properly utilizes available threads.
    • Removed sync method from VideoWriter.
      • Synchronization can be manually handled by the user or by letting the VideoWriter do so on destruction.
    • Updated Benchmarks to reflect new version.

Version 0.3.7 (2024-10-21)

  • Pre-Release Update:
    • Fixed remaining issues with VideoWriter class.
      • Both cpu and cuda arguments NOW work properly.
    • Few Small bug fixes regarding synchronization and memory management.

Version 0.3.6 (2024-10-19)

  • Pre-Release Update:
    • Fixed VideoWriter class.
      • Both cpu and cuda arguments now work properly.
    • Encoder Functionality:
      • Enabled encoder support for both CPU and CUDA backends.
      • Users can now encode videos directly from PyTorch tensors.
    • Update Github Actions, add tests.

Version 0.3.5 (2024-10-19)

  • Pre-Release Update:
    • (somewhat) Fixed VideoWriter class. Working on cuda for now, but cpu still has incorrect output.
    • Added VideoWriter, and LogLevel definitions to .pyi stub file.
    • Adjusted github actions to publish to pypi.

Version 0.3.4.1 (2024-10-19)

  • Pre-Release Update:
    • Added logging utility for debugging purposes. 
      import celux
celux.set_log_level(celux.LogLevel.debug)

Version 0.3.3 (2024-10-19)

  • Pre-Release Update:
    • Added buffer_size and stream arguments.
      • Choose Pre-Decoded Frame buffer size, and pass your own cuda stream.
    • Some random cleanup and small refactorings.

Version 0.3.1 (2024-10-17)

  • Pre-Release Update:
    • Adjusted Frame Range End in VideoReader to be exclusive to match cv2 behavior.
    • Removed unnecessary error throws.
    • Encoder Functionality: Now fully operational for both CPU and CUDA.

Version 0.3.0 (2024-10-17)

  • Pre-Release Update:
    • Renamed from ffmpy to CeLux.
    • Created official pypi release.
    • Refactored to split cpu and cuda backends.

Version 0.2.6 (2024-10-15)

  • Pre-Release Update:
    • Removed Numpy support in favor of PyTorch tensors with GPU/CPU support.
    • Added NV12ToBGR, BGRToNV12, and NV12ToNV12 conversion modules.
    • Fixed several minor issues.
    • Updated documentation and examples.

Version 0.2.2 (2024-10-14)

  • Pre-Release Update:

    • Fixed several minor issues.

    • Made VideoReader and VideoWriter callable.

    • Created BGR conversion modules.

    • Added frame range (in/out) arguments.

      with VideoReader('input.mp4')([10, 20]) as reader:
    for frame in reader:
        print(f"Processing frame {frame}")

Version 0.2.1 (2024-10-13)

  • Pre-Release Update:
    • Adjusted Python bindings to use snake_case.
    • Added .pyi stub files to .whl.
    • Adjusted dtype arguments to (uint8, float32, float16).
    • Added GitHub Actions for new releases.
    • Added HW Accel Encoder support, direct encoding from numpy/tensors.
    • Added has_audio property to VideoReader.get_properties().

Version 0.1.1 (2024-10-06)

  • Pre-Release Update:
    • Implemented support for multiple data types (uint8, float, half).
    • Provided example usage and basic documentation.