Partitioning variability in animal behavioral videos using semi-supervised variational autoencoders.
Matthew R. WhitewayDan BidermanYoni FriedmanMario DipoppaEstefany Kelly BuchananAnqi WuJohn ZhouNiccolò BonacchiNathaniel J. MiskaJean-Paul NoelErica RodriguezMichael SchartnerKarolina SochaAnne E. UraiC. Daniel SalzmanInternational Brain LaboratoryJohn P. CunninghamLiam PaninskiPublished in: PLoS Comput. Biol. (2021)
Keyphrases
- semi supervised
- fully labeled
- denoising
- semi supervised learning
- video sequences
- supervised learning
- pairwise
- semi supervised classification
- labeled data
- active learning
- multi view
- co training
- video frames
- human activities
- unlabeled data
- unsupervised learning
- video clips
- optical flow
- video database
- video data
- image segmentation
- restricted boltzmann machine
- semi supervised clustering
- video content
- human behavior
- generative model
- graph partitioning
- variational methods
- free energy
- weakly supervised
- pairwise constraints
- dynamic scenes
- video surveillance
- clustering algorithm