Motion Guided Attention Fusion to Recognize Interactions from Videos.
Tae Soo KimJonathan D. JonesGregory D. HagerPublished in: CoRR (2021)
Keyphrases
- periodic motion
- motion features
- video sequences
- space time
- moving camera
- key frames
- input video
- optical flow
- static images
- camera motion
- recognition of human actions
- video representation
- video frames
- motion analysis
- multi modal fusion
- image sequences
- motion estimation
- motion field
- motion model
- video data
- temporal coherence
- dynamic scenes
- video content
- motion cues
- dynamic textures
- motion patterns
- spatial and temporal
- motion detection
- motion segmentation
- successive frames
- visual attention
- moving objects
- traffic scenes
- adjacent frames
- human actions
- action recognition
- motion capture
- information fusion
- group activities
- motion trajectories
- motion capture data
- fusion method
- visual features
- human motion
- reference frame
- video stabilization
- recognizing human actions
- motion parameters
- video indexing
- low frame rate
- view invariant
- video retrieval
- data fusion
- video scene