An efficient part-based approach to action recognition from RGB-D video with BoW-pyramid representation.
Jih-Sheng TsaiYen-Pin HsuChengyin LiuLi-Chen FuPublished in: IROS (2013)
Keyphrases
- action recognition
- bag of words
- recognizing human actions
- spatial pyramid matching
- recognition of human actions
- image representation
- action recognition in videos
- view invariant
- human actions
- bag of features
- human detection
- computer vision
- action classification
- activity recognition
- bag of visual words
- multiscale
- body parts
- visual words
- object recognition
- three dimensional
- action primitives
- video dataset
- color space
- image classification
- static images
- mid level
- human pose
- depth information
- human activities
- action detection
- recognizing actions
- feature space
- image sequences