Dynamically Weighted DTW for Dynamic Full-Body Gesture Recognition
Hyo-Rim Choi, Hye Yeon Cho, Tae Yong Kim · 2015
As sensor technology has become more advanced,research projects exploring new ways in human-computerinteraction have been enabled. Especially, after the Kinectreleased, study of gesture recognition for human-computerinteraction using Kinect has become popular. DynamicTime Warping (DTW) is one of the template matchingalgorithm used in the field of gesture recognition. Byselecting the most similar gesture compared with the DTWto input sequence and reference sequence of the database, itis possible to achieve a gesture recognition. When trying torecognize a dynamic full-body gesture in Kinectenvironment, we have to compare all joints throughout thebody. When calculating the similarity of the two sequences,all joints of the body is not important in the same way. Wepropose a dynamically weighted DTW method based on thedisplacement of the joint. The proposed method showsgood performance in the experimental results.