Human action recognition based on 3D skeleton has become an active research field in recent years with the recently developed\ncommodity depth sensors. Most published methods analyze an entire 3D depth data, construct mid-level part representations, or\nuse trajectory descriptor of spatial-temporal interest point for recognizing human activities. Unlike previous work, a novel and\nsimple action representation is proposed in this paper which models the action as a sequence of inconsecutive and discriminative\nskeleton poses, named as key skeleton poses.The pairwise relative positions of skeleton joints are used as feature of the skeleton poses\nwhich are mined with the aid of the latent support vector machine (latent SVM). The advantage of our method is resisting against\nintraclass variation such as noise and large nonlinear temporal deformation of human action.We evaluate the proposed approach\non three benchmark action datasets captured by Kinect devices: MSR Action 3D dataset, UT Kinect Action dataset, and Florence\n3D Action dataset. The detailed experimental results demonstrate that the proposed approach achieves superior performance to\nthe state-of-the-art skeleton-based action recognition methods.
Loading....