Action Recognition with Joints-Pooled 3D Deep Convolutional Descriptors.
Congqi CaoYifan ZhangChunjie ZhangHanqing LuPublished in: IJCAI (2016)
Keyphrases
- action recognition
- action recognition in videos
- body parts
- bag of features
- deep learning
- mid level features
- human actions
- bag of words
- human detection
- computer vision
- activity recognition
- recognizing actions
- action classification
- recognizing human actions
- keypoints
- human body
- static images
- shape descriptors
- spatial temporal
- recognition of human actions
- video dataset
- action primitives
- low level descriptors
- multiscale
- feature descriptors
- human activities
- sift descriptors
- spatio temporal
- image retrieval
- view invariant action recognition
- robust tracking
- human pose
- action detection
- depth sensors
- human motion
- pose estimation