Video Action Recognition Using Motion and Multi-View Excitation with Temporal Aggregation.
Yuri Yudhaswana JoefrieMasaki AonoPublished in: Entropy (2022)
Keyphrases
- multi view
- dynamic scenes
- view invariant
- temporal aggregation
- space time
- motion capture
- human actions
- scene flow estimation
- single view
- free viewpoint
- moving camera
- spatio temporal
- multiple views
- depth video
- video sequences
- d objects
- depth map
- human motion
- motion estimation
- camera calibration
- temporal databases
- three dimensional
- key frames
- video data
- semi supervised
- action recognition
- multiple cameras
- temporal data
- depth image based rendering
- image sequences
- view synthesis
- moving objects
- optical flow
- video objects
- stereoscopic video
- multi view images
- multi view video coding
- multiview video
- machine learning
- light field
- video frames
- motion model
- human body
- depth information
- motion field
- viewpoint
- object recognition
- database