Time-Conditioned Generative Modeling of Object-Centric Representations for Video Decomposition and Prediction.
Chengmin GaoBin LiPublished in: CoRR (2023)
Keyphrases
- video sequences
- video data
- object description
- object detection and tracking
- multimedia
- video content
- prediction accuracy
- probabilistic topic models
- video frames
- d objects
- object representation
- prediction model
- real time
- successive frames
- object motion
- prediction error
- object model
- complex objects
- video streams
- space time
- generative model
- video surveillance
- moving objects
- object tracking
- scalable video coding
- multiple objects
- video objects
- probabilistic model
- unsupervised learning
- object recognition
- motion estimation