Human action recognition based on skeleton splitting

• We present a novel human action recognition technique.
• We use segmented human body parts’ eigenvalues and eigenvectors, derived using diffusion tensor fields.
• With a multiple kernel SVM-approach this yields an efficient and effective human action recognition system.
• Experiments on publicly available data sets show better recognition.

Human action recognition, defined as the understanding of the human basic actions from video streams, has a long history in the area of computer vision and pattern recognition because it can be used for various applications. We propose a novel human action recognition methodology by extracting the human skeletal features and separating them into several human body parts such as face, torso, and limbs to efficiently visualize and analyze the motion of human body parts.Our proposed human action recognition system consists of two steps: (i) automatic skeletal feature extraction and splitting by measuring the similarity between neighbor pixels in the space of diffusion tensor fields, and (ii) human action recognition by using multiple kernel based Support Vector Machine. Experimental results on a set of test database show that our proposed method is very efficient and effective to recognize the actions using few parameters.