Skepxels: Spatio-temporal Image Representation of Human Skeleton Joints for Action Recognition.
Jian LiuNaveed AkhtarAjmal MianPublished in: CoRR (2017)
Keyphrases
- image representation
- action recognition
- spatio temporal interest points
- human actions
- spatio temporal
- visual words
- bag of words
- body parts
- bag of features
- spatial temporal
- bag of visual words
- multiscale
- view invariant
- action recognition in videos
- human activities
- image classification
- static images
- human motion
- computer vision
- action classification
- object recognition
- activity recognition
- image features
- image retrieval
- recognition of human actions
- motion history images
- image content
- human detection
- feature space
- sparse representation
- scene classification
- sparse coding
- object categories
- recognizing human actions
- human object interactions
- mid level
- human pose
- moving objects
- image sequences
- scene recognition
- object class
- low level features
- space time