• Improve --vpp-kfm performance for long processing.
• Extend temporal options in --vpp-fft3d.
• Improve diagnostics when ONNX Runtime fails to load.
• Avoid nan in CUDA --vmaf. ( #781 )
• Fix Linux VMAF CUDA support. ( #781 )
• Enable libvship build on Linux.
• Add --vpp-rife-ov RIFE v4.x frame interpolation filter.
• Add SDR to HDR model support to --vpp-onnx.
  • New model files can be downloaded from the link ( https://github.com/rigaya/HWEnc-onnx-models/releases ).
• Clarify error for non-existent VMAF json model. ( #783 )
• Add OpenVINO RIFE frame interpolation filter --vpp-rife-ov.
• Add coring and hue range to --vpp-tweak.
• Add interpolation mode and fix all in --vpp-curves.
• Add detailed parameters to --vpp-hqdering.
• Add chroma to --vpp-cas.
• Add valid source size to --vpp-descale.
• Add keep and fix state management in --vpp-mpdecimate.
• Add detection parameters to --vpp-ivtc.
• Add planes to --vpp-nnedi.
• Add temporal radius and improve temporal processing in --vpp-knn.
• Fix yuv444→y410 conversion.
• Fix mask output in --vpp-msmooth.
• Fix chroma processing in --vpp-vinverse.
• Propagate Yadif copyFrameAsync error in --vpp-yadif.
• Fix nvvfx output crop height check.
• Fix alpha input buffer width.
• Fix smooth QP table reference coordinate clamp.
• Fix Anime4K spline36 linear term in --vpp-anime4k-shader.
• Fix AFS mie_spot8 4th element reference in --vpp-afs.
• Fix colorspace LUT3D green axis scale in --vpp-colorspace.
• Fix libplacebo radius warning argument in --vpp-libplacebo-shader.
• Fix libplacebo deband grain target in --vpp-libplacebo-deband.
• Fix delogo two-point least squares intercept in --vpp-delogo.
• Fix decimate diff grid height in --vpp-decimate.
• Fix decomb stripe grid Y stride in --vpp-decomb.
• Fix chroma reference line in RGB→YUV420 conversion.
• Apply denoise filter fixes.
• Fix threshold in --vpp-convolution3d.