Pose-Enhanced Relation Feature for Action Recognition in Still Images.
Jiewen WangShuang LiangPublished in: MMM (1) (2022)
Keyphrases
- action recognition
- image features
- static images
- bag of features
- human pose
- human actions
- three dimensional
- mid level
- object recognition
- video dataset
- recognizing actions
- human detection
- image classification
- input image
- image retrieval
- bag of words
- activity recognition
- action detection
- action classification
- computer vision
- motion features
- body parts
- view invariant
- image matching
- keypoints
- d objects
- low level descriptors
- class specific
- bounding box
- feature vectors
- feature points
- object detection