Human Action Recognition in Videos via Principal Component Analysis of Motion Curves

A new approach to human action recognition in videos is presented and evaluated using the Weizmann action dataset. In this approach, motion trajectories are formed by tracking one or more key points on the human body. In particular, points on the hands and feet are tracked. A curve is fitted to each motion trajectory to smooth noise and to form a continuous and differentiable curve. A motion curve is then segmented at peak curvature points, representing each segment by a “basic motion.” To recognize an observed basic motion, a vector of curve features describing the motion is created, the vector is projected to the eigenspace created during PCA training, and the action most similar to a learned action is identified using the k-nearest neighbor