PyNvVideoCodec 2.0 introduces major enhancements for GPU-accelerated video processing in Python. Key improvements include flexible frame sampling and seeking, decoder caching, threaded decoding, buffer-based processing, and multi-GPU support. The library now supports extended codec formats (H.264, HEVC, AV1, VP8, VP9) and 4:2:2 encoding/decoding on Blackwell architecture. New features enable live encoder reconfiguration, SEI message handling, and segment-based transcoding optimized for deep learning workflows.

3m read timeFrom developer.nvidia.com
Post cover image
Table of contents
What’s new in PyNvVideoCodec 2.0?Get started with PyNvVideoCodec 2.0

Sort: