Overview

PyNvVideoCodec is NVIDIA's Python based video codec library for hardware accelerated video encode and decode on NVIDIA GPUs.

PyNvVideoCodec is the successor of VPF (Video Processing Framework).  The library is distributed under MIT license and is officially supported by NVIDIA.

PyNvVideoCodec library internally uses Video Codec SDK's core C/C++ video encode and decode APIs and provides easy to use Python APIs on it. The library offers encode and decode performance close to Video Codec SDK.

Features

Current PyNvVideoCodec version supports following features:

• Codecs: H.264, HEVC, AV1.
• Surface formats: NV12 (8 bit), YUV 4:2:0 (10 bit), YUV 4:4:4 (8 and 10 bit).
• Video  container formats:  MP4, AVI and MKV.
• DLPack support to facilitate data exchange with popular DL frameworks like PyTorch and TensorRT.
• CUDA Array Interface support to facilitate data exchange with NVIDIA's CV-CUDA library.
• CUDA stream support for optimizing throughput.
• Contains Python sample applications demonstrating API usage.

Distribution

PyNvVideoCodec library is distributed in two formats: binary distribution via PyPI and source code distribution via NVIDIA NGC (this page). In both cases, the library and its dependencies can be installed using a single pip install command.

This package on NVIDIA NGC contains PyNvVideoCodec source code, its dependencies, Python sample applications and documents. To install PyNvVideoCodec library please download zip file containing the latest version of PyNvVideoCodec from this NVIDIA NGC catalog page (located in the file browser tab) onto your local machine. Then, open the shell prompt, go to the folder containing the zip file and execute the following command, replacing "PyNvVideCodec.zip'' with the actual name of the downloaded zip file.

pip install PyNvVideoCodec.zip

Documents

• Documents and Python Sample applications demonstrating the API usage can be easily accessed from latest PyNvVideoCodec package on this NVIDIA NGC catalog page.
• For your convenience, these documents are also accessible online on NVIDIA Video Technologies page.