Optical Flow
The optical flow code used was a part of code developed for CVPR 2017 paper on Unsupervised Learning using unlabeled videos.
The link for the github repo: https://github.com/pathak22/unsupervised-video
Optical flow is basically based upon using the motion of cilia in the images to segment the cilia in an unsupervised way. The key objective here is that if there is a single object moving with respect to the background through the entire video, then pixels on this object will move differently from pixels in the background. Analyzing the optical flow, will give us which pixels belong to the foreground. Since only a part of the pixels moves in each frame, this has to be aggregated over a number of frames.
