v2.13.0

Intel® Video Processing Library (Intel® VPL) provides access to hardware accelerated video decode, encode, and processing capabilities on Intel® GPUs to support AI visual inference, media delivery, cloud gaming, and virtual desktop infrastructure use cases.

Added

• Intel® VPL API 2.13 support, including new APIs for AV1 screen content tools,
  encoded picture quality information, alpha channel encoding, AI-based frame
  interpolation, AI-based super resolution, and Battlemage platform
• hello-encode-jpeg example

Known Issues

• vpl-infer example will fail to load model if built with CMake version higher than 3.25.3 on Windows

v2.12.0

Intel® Video Processing Library (Intel® VPL) provides access to hardware accelerated video decode, encode, and processing capabilities on Intel® GPUs to support AI visual inference, media delivery, cloud gaming, and virtual desktop infrastructure use cases.

Added

• Intel® VPL API 2.12 support, including new APIs for AI-based frame
  interpolation and Vulkan zero-copy surface sharing.

Known Issues

• vpl-infer example will fail to load model if built with CMake version higher than 3.25.3 on Windows

v2.11.0

Added

• Intel® VPL API 2.11 support

Removed

• Command line tools. They have been moved to a separate repository
  (https://github.com/intel/libvpl-tools)

Known Issues

• vpl-infer example will fail to load model if built with CMake version higher than 3.25.3 on Windows

v2.10.2

Fixed

• "®" symbol in pkg-config file breaking downstream build tools

Known Issues

• sample_decode may report "DeliverOutput return error = -1" when run using -r option
• vpl-infer example will fail to load model if built with CMake version higher than 3.25.3 on Windows

v2.10.1

Changed

• Project name to Intel® Video Processing Library (Intel® VPL)
• Project versioning scheme to align with library and API version

Added

• Intel® VPL specification and associated build scripts

Known Issues

• sample_decode may report "DeliverOutput return error = -1" when run using -r option
• vpl-infer example will fail to load model if built with CMake version higher than 3.25.3 on Windows

v2023.4.0

Added

• oneVPL API 2.10 support
• New command line parameters to sample_multi_transcode for setting
  B Frame QP Offsets based on Pyramid Level
• New command line parameters to the sample_vpp tool to enable video signal
  information
• New APIs for importing and exporting shared surfaces
• New APIs for string-based parameter configuration

Removed

• Environment module support
• Unused files and folders

Fixed

• NV12 wayland render for MTL
• Buffer starvation when wayland window is moved
• Missing profile strings in vpl-inspect tool
• Rdrm to use card node instead of render node
• Rdrm tile4 selection based DRM/KMS query
• Rdrm NV12 and P010 render for MTL

Changed

• Directory layout and name of dispatcher directory
• Location of image files used by documentation
• Runtime library search path to remove current directory

v2023.3.1

Added

• Three new command line parameters to the sample_multi_transcode tool to enable
  video quality tuning for different use cases.
• New command line parameter to the sample_multi_transcode tool to enable
  adaptiveCQM control

Removed

• Outdated docker files

Fixed

• Unsafe calling of SetCurrentDirectory during MFXEnumImplementations
• Parameter parsing in sample_vpp tool

v2023.3.0

Removed

• USE_ONEAPI_INSTALL_LAYOUT build shortcut
• '-sw' option from examples

Added

• New MFXSetConfigFilterProperty property: mfxExtendedDeviceId.RevisionID

Fixed

• Fix HDR luminance value
• Handling of special cases of CMAKE_INSTALL_PREFIX
• X11 rendering corruption issue
• Intel® Distribution of OpenVINO™ Toolkit installation steps in vpl-infer README
• Intermittent Sample_multi_transcode segfault on Wayland

Changed

• quick-start scripts to be more simple and easy to use

V2023.2.1

Added

• Documentation on security policy and vulnerability reporting process

Removed

• Fix for VPP processing for YUV input

v2023.2.0

Added

• oneVPL API 2.9 support
• Perceptual encoding prefilter option to sample_multi_transcode

Fixed

• VPP processing for YUV input
• Sample_multi_transcode segfault on wayland
• Missing prerequisites in vpl-infer README

Changed

• vpl-infer Dockerfile and Linux setup steps to better align with https://dgpu-docs.intel.com/

Removed

• dpcpp-blur example