Accurate 3D action recognition using learning on the Grassmann manifold

作者：

Highlights：

• A human action recognition approach which represents skeletal sequence as point on the Grassmann manifold.

• A new learning algorithm is introduced for learning human actions.

• Experiments are performed on three public datasets.

• Promising success rates are achieved, showing accuracy and better latency performances.

摘要

Highlights•A human action recognition approach which represents skeletal sequence as point on the Grassmann manifold.•A new learning algorithm is introduced for learning human actions.•Experiments are performed on three public datasets.•Promising success rates are achieved, showing accuracy and better latency performances.

论文关键词：Human action recognition,Grassmann manifold,Observational latency,Depth images,Skeleton,Classification

论文评审过程：Received 11 February 2014, Revised 23 July 2014, Accepted 13 August 2014, Available online 24 August 2014.

论文官网地址：https://doi.org/10.1016/j.patcog.2014.08.011