Diving Deep into the Motion Representation of Video-Text Models.
Chinmaya DevarajCornelia FermüllerYiannis AloimonosPublished in: CoRR (2024)
Keyphrases
- video sequences
- dynamical models
- space time
- recognizing human actions
- human motion
- video data
- video representation
- image representation
- image sequences
- temporal structure
- key frames
- temporal filtering
- temporal consistency
- temporal coherence
- video search
- video analysis
- dynamic textures
- information retrieval
- motion analysis
- motion estimation
- spatial and temporal
- video streams
- multimedia
- dynamic scenes
- moving objects
- optical flow
- input video
- text detection
- temporal domain
- video content
- video scene
- motion features
- static images
- object motion
- motion segmentation
- visual cues