Spatio-Temporal Representation Matching-Based Open-Set Action Recognition by Joint Learning of Motion and Appearance.
Yongsang YoonJongmin YuMoongu JeonPublished in: IEEE Access (2019)
Keyphrases
- action recognition
- action recognition in videos
- recognition of human actions
- recognizing human actions
- view invariant
- human actions
- spatio temporal
- static images
- spatial and temporal
- spatial temporal
- bag of words
- image sequences
- human detection
- motion estimation
- activity recognition
- bag of features
- mid level
- space time
- moving objects
- spatio temporal interest points
- appearance model
- body parts
- keypoints
- max margin
- action classification
- object detection
- computer vision