Learning Disentangled Representation in Object-Centric Models for Visual Dynamics Prediction via Transformers.
Sanket Gandhi AtulSamanyu MahajanVishal SharmaRushil GuptaArnab Kumar MondalParag SinglaPublished in: CoRR (2024)
Keyphrases
- learning systems
- visual representation
- learning algorithm
- prior knowledge
- object representations
- accurate models
- reinforcement learning
- learning process
- dynamic bayesian networks
- learning models
- learned models
- spatial relations
- model selection
- learning tasks
- dynamic model
- complex objects
- complex systems
- structured prediction
- visual learning
- contextual cues
- category specific
- visual properties
- visual objects
- relational models
- active learning
- online learning
- radial basis function network
- predictive model
- visual information
- hidden variables
- machine learning
- prediction model