Hi!
This is a feature request for what the title says. In video frame servers such as VapourSynth and Avisynth, video frame data are usually stored in separated planes instead of packed formats, partially because most filters are selective on planes. As a result, the planes are by design not guaranteed to be equidistant in memory. For now I just create a temp buffer and copy plane data per line to it for the DoTransform call. But I would like to hear from you if this use case might be supported in low effort.
Hi! This is a feature request for what the title says. In video frame servers such as VapourSynth and Avisynth, video frame data are usually stored in separated planes instead of packed formats, partially because most filters are selective on planes. As a result, the planes are by design not guaranteed to be equidistant in memory. For now I just create a temp buffer and copy plane data per line to it for the DoTransform call. But I would like to hear from you if this use case might be supported in low effort.