Motion Guided Attention Fusion to Recognize Interactions from Videos.
Tae Soo KimJonathan D. JonesGregory D. HagerPublished in: ICCV (2021)
Keyphrases
- periodic motion
- moving camera
- motion features
- multi modal fusion
- key frames
- successive frames
- image sequences
- space time
- recognizing human actions
- human motion
- static images
- dynamic scenes
- input video
- video sequences
- video representation
- motion capture data
- recognition of human actions
- motion analysis
- motion cues
- motion estimation
- motion model
- camera motion
- motion trajectories
- video objects
- surveillance videos
- motion parameters
- action recognition
- data fusion
- motion patterns
- temporal domain
- reference frame
- video data
- video content
- arbitrary viewpoint
- temporal coherence
- motion sequences
- video shots
- optical flow
- object motion
- video summarization
- moving objects
- visual data
- video analysis
- motion detection
- video frames
- adjacent frames
- spatial and temporal
- visual attention
- video stabilization
- feature points
- human body
- low frame rate
- long video sequences
- traffic scenes
- motion field
- motion segmentation
- information fusion
- image fusion
- global motion
- motion capture