Visual hand gestures classification using temporal motion templates and wavelet transforms

This paper presents a technique for classifying human hand gestures based on stationary wavelet transform (SWT). This approach uses a cumulative image-difference technique where the time between the sequences of images is implicitly captured in the representation of action. This results in the construction of temporal history templates (THT). These THTs are decomposed into 4 subimages using SWT, an average image (f_ll), and three detail images (f_lh, f_hb, f_hh) respectively. The average image (f_ll) is fed as the global image descriptors to the ANN for classification. The preliminary experiments show that such a system can classify human hand gestures with a classification accuracy of 97%.