Diving Deep into the Motion Representation of Video-Text Models.
Chinmaya DevarajCornelia FermüllerYiannis AloimonosPublished in: ACL (Findings) (2024)
Keyphrases
- human motion
- motion estimation
- spatial and temporal
- recognition of human actions
- motion analysis
- information retrieval
- video sequences
- motion features
- video data
- space time
- camera motion
- temporal structure
- static images
- motion capture
- moving camera
- temporal filtering
- motion model
- image representation
- dynamical models
- recognizing human actions
- input video
- video representation
- motion patterns
- temporal coherence
- configuration space
- visual cues
- video content
- key frames
- image sequences