SVM-based human action recognition and its remarkable motion features discovery algorithm

This paper proposes a discovery algorithm of knowledge of remarkable motion features in SVM-based action recognition. The main characteristics of the proposed method are a) basic scheme of the algorithm is based on Support Vector Learning and its generalization error, b) remarkabale motion features are discovered in response to kernel parameters optimization through generalization error minimization. Experimental results show that this proposed algorithm makes the recognition robust and finds remarkable motion features that are intuitive for human.