SIMONe: View-Invariant, Temporally-Abstracted Object Representations via Unsupervised Video Decomposition.
Rishabh KabraDaniel ZoranGoker ErdoganLoic MattheyAntonia CreswellMatthew BotvinickAlexander LerchnerChristopher P. BurgessPublished in: CoRR (2021)
Keyphrases
- object representations
- view invariant
- human actions
- complex objects
- spatio temporal
- action recognition
- object categorization
- video sequences
- real world objects
- human motion
- single view
- space time
- object models
- multi view
- object categories
- video data
- visual features
- semi supervised
- human activities
- d objects
- active learning
- object recognition
- object classes
- supervised learning
- gabor wavelets
- object representation
- viewpoint