ICONDENSATION-Based Hand Tracking and Gesture Recognition with Color and Depth Cues

Presents a hand tracking and gesture recognition algorithm for human-computer interaction with two web cameras.ICONDENSATION technique is employed to fuse color and depth to generate importance particles.To improve the efficiency and accuracy of likelihood evaluation,all samples are validated according to the energy function of active contour models,taking into account the contour information.Gesture recognition is realized through the maximum posterior estimation of several pre-defined templates.Experimental results showed that the proposed method works well even in cluttered scenarios.