github Netflix/vmaf v3.2.1

5 hours ago

Small release with a fix for #1587.

What's Changed

  • doc/vmaf_v1: update Netflix techblog url by @kylophone in #1545
  • build(deps): bump actions/checkout from 6 to 7 by @dependabot[bot] in #1546
  • build(deps): bump actions/cache from 4 to 6 by @dependabot[bot] in #1550
  • libvmaf/thread_pool: fix thread pool queue depth to restore backpressure by @kylophone in #1557
  • libvmaf/feature: enable speed_chroma/speed_temporal without enable_float by @kylophone in #1558
  • .github/ci: fix release asset name collision across build matrix by @kylophone in #1559
  • libvmaf/speed_chroma: remove bilinear prescale index/weight computation from per-pixel loop by @kylophone in #1565
  • build(deps): bump actions/setup-python from 6 to 7 by @dependabot[bot] in #1569
  • cuda: fix row stride in calculate_motion_score_kernel_16bpc (wrong scores for >8-bit input) by @BardieJoensen in #1552
  • Fix race condition with generated header under parallel builds by @StormBytePP in #1578
  • Add ARM NEON implementation for 8-bit integer motion feature by @kjg0724 in #1570
  • AVX512: Drop -mavx512vbmi from generic AVX-512 compile flags by @StormBytePP in #1577
  • Add support for Intel CET (SHSTK) by @StormBytePP in #1575

New Contributors

Full Changelog: v3.2.0...v3.2.1

Don't miss a new vmaf release

NewReleases is sending notifications on new releases.