Mining actionlet ensemble for action recognition with depth cameras
Northwestern University · Microsoft (United States) · +1 more institution
Abstract
Human action recognition is an important yet challenging task. The recently developed commodity depth sensors open up new possibilities of dealing with this problem but also present some unique challenges. The depth maps captured by the depth cameras are very noisy and the 3D positions of the tracked joints may be completely wrong if serious occlusions occur, which increases the intra-class variations in the actions. In this paper, an actionlet ensemble model is learnt to represent each action and to capture the intra-class variance. In addition, novel features that are suitable for depth data are proposed. They are robust to noise, invariant to translational and temporal misalignments, and capable of…
Citation impact
- FWCI
- 105.71
- Percentile
- 100%
- References
- 33
Authors
4Topics & keywords
- Computer science
- Action recognition
- Artificial intelligence
- Action (physics)
- Computer vision
- Pattern recognition (psychology)
- Physics