Skip to Main Content
This paper proposes a method to detect small-waving hand from image sequences. The method is based on frequency analysis of intensity value change for each pixel of low-resolution images converted from input images. Handwaving makes periodical change of intensity value of image pixels located in the moving region. The periodical change is detected by thresholding of a value accumulating frequency features obtained through FFT calculation. The proposed method can stably detect small-waving hand with 3-5 Hz in a range of 2-6 m. We demonstrate an application of spatial memory through measuring 3D position of small-waving hand detected by a distributed camera system.