is it possible to produce optical flow between two frames using co tracker

this might sound stupid, i am searching for a deep learning based optical flow method that also produces confidence score and occlusion,
i dont know if co-tracker can track all pixels in 1 frame and produce something like optical flow, if yes what should i do?

The issue names CoTracker, optical flow, confidence scores, and occlusion handling but no files, tests, or entry points. First clarify whether tracking every pixel and producing these outputs is supported or requires a new capability; done would require an agreed scope and documented, verifiable outputs.

Written by the indexing model from the issue text.