RCF-UnsupVideoSeg
RCF-UnsupVideoSeg copied to clipboard
dataset
Hello, if I want to use my own dataset, So "Download pre-extracted flow from RAFT (trained with chairs and things) and decompress to data/data_davis". How should we change it?
You can run RAFT on your own to extract the optical flow. We use the official code and checkpoints.
Thank you very much.