View-independent representation with frame interpolation method for skeleton-based human action recognition

• Published in: International journal of machine learning and cybernetics Vol. 11; no. 12; pp. 2625 - 2636
• Main Authors: Jiang, Yingguo, Xu, Jun, Zhang, Tong
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.12.2020; Springer Nature B.V
• Subjects: Accuracy; Artificial Intelligence; Complex Systems; Computational Intelligence; Computer vision; Control; Data collection; Engineering; Hip joint; Human activity recognition; Human mechanics; Interpolation; Mechatronics; Methods; Neural networks; Original Article; Pattern Recognition; Representations; Robotics; Semantics; Systems Biology; Action recognition; View-independent representation; Transfer CNN; Self-attention mechanism; Frame interpolation
• ISSN: 1868-8071; 1868-808X
• DOI: 10.1007/s13042-020-01132-4

Human action recognition is an important branch of computer vision science. It is a challenging task based on skeletal data because of joints’ complex spatiotemporal information. In this work, we propose a method for action recognition, which consists of three parts: view-independent representation, frame interpolation, and combined model. First, the action sequence becomes view-independent representations independent of the view. Second, when judgment conditions are met, differentiated frame interpolations are used to expand the temporal dimensional information. Then, a combined model is adopted to extract these representation features and classify actions. Experimental results on two multi-view benchmark datasets Northwestern-UCLA and NTU RGB+D demonstrate the effectiveness of our complete method. Although using only one type of action feature and a simple architecture combined model, our complete method still outperforms most of the referential state-of-the-art methods and has strong robustness.