Segmenting Video Into Classes of Algorithm-Suitability
OverviewGiven a set of algorithms, which one(s) should you apply to, i) compute optical flow, or ii) perform feature matching? Would looking at the sequence in question help you decide? It is unclear if even a person with intimate knowledge of all the different algorithms and access to the sequence itself could predict which one to apply. Our hypothesis is that the most suitable algorithm can be chosen for each video automatically, through supervised training of a classifier. The classifier treats the different algorithms as black-box alternative "classes," and predicts when each is best because of their respective performances on training examples where ground truth flow was available.
Data and CodeIf you find the following useful, please cite our CVPR paper.
UCL Ground Truth Optical Flow Dataset v1.1
This dataset contains 4 different rigid scenes with marked occlusion regions, imaged at different locations, with alternate textures.
Maya python code for generating ground truth optical flow.
[Maya GetFlow Code]
Code for recreating experiments in paper to compute flow confidence.
Errata: A mistake was found in the GT data for the Robot sequence resulting in incorrect errors reported in the paper.
Contact: omacaodh (@) cs.ucl.ac.uk