STFC: Spatio-temporal feature chain for skeleton-based human action recognition

Human action recognition and analysis has been of interest to researchers of computer vision for many years. This paper presents a method to recognize human actions from sequences of 3D joint positions. The major contributions of this paper include: (1) An action decomposition method that uses motion velocities, the direction of motion, and the curvatures of trajectories to encode the temporal decomposition of action into a sequence of meaningful atomic actions (actionlets); and (2) the concept of the Spatio-Temporal Feature Chain (STFC) that is introduced to represent the characteristic parameters of temporal sequential patterns, which exhibit greater robustness to noise and temporal misalignment. The effectiveness of the proposed method is evaluated on three challenging 3D action datasets captured by commodity depth cameras. The experimental evaluations show that the proposed approach achieves promising results compared to other state of the art algorithms.