Joint Dynamic Pose Image and Space Time Reversal for Human Action Recognition from Videos.
Mengyuan LiuFanyang MengChen ChenSongtao WuPublished in: AAAI (2019)
Keyphrases
- space time
- input video
- human actions
- video sequences
- physically plausible
- super resolution reconstruction
- dynamic scenes
- temporal domain
- spatial and temporal
- spatio temporal
- single image
- video representation
- image retrieval
- dynamic textures
- image features
- key frames
- input image
- high resolution
- image classification
- image collections
- pose estimation
- human pose
- action recognition
- motion patterns
- scan statistic
- frame rate
- feature vectors
- video frames
- super resolution
- video data