Time-Conditioned Generative Modeling of Object-Centric Representations for Video Decomposition and Prediction.
Chengmin GaoBin LiPublished in: UAI (2023)
Keyphrases
- prediction accuracy
- video data
- object motion
- video sequences
- d objects
- object detection and tracking
- generative model
- successive frames
- multimedia
- object model
- video streams
- video frames
- real time
- models built
- predictive modeling
- multiple objects
- object description
- object segmentation
- dynamic scenes
- video clips
- video surveillance
- video content
- moving objects
- video retrieval
- spatial relationships
- complex objects
- prediction error
- video analysis
- object models
- data objects
- discriminative learning
- key frames
- video scene
- temporal information
- space time
- object detection