Sparse feature auto-combination deep network for video action recognition.
Qicong WangDingxi GongMaozhen LiChong ZhaoYunqi LeiPublished in: ICNC-FSKD (2017)
Keyphrases
- action recognition
- human actions
- spatio temporal interest points
- motion features
- action classification
- spatial temporal
- video dataset
- bag of words
- recognition of human actions
- action detection
- activity recognition
- recognizing human actions
- static images
- mid level
- human activities
- human detection
- space time interest points
- computer vision
- mid level features
- body parts
- video sequences
- depth sensors
- motion history images
- action primitives
- view invariant action recognition
- human pose
- bag of features
- space time
- multimedia
- view invariant
- human activity recognition
- atomic actions
- video shots
- independent component analysis
- video frames
- action recognition in videos
- video data
- image sequences