作者: Kazunori Umeda , Kota Irie , Masahito Takahashi , Hidetoshi Ogishima , Kenji Terabayashi
DOI:
关键词:
摘要: In this paper, we propose a method to recognize periodic gestures from images. The proposed uses power spectrum and phase that are obtained by applying FFT time series of intesity is applied each pixel the low-resolution consisted 2 steps. First, detects motion regions spectrum. Second, in detected region classify gestures. also robust lighting condition indi- vidual difference skin color, because it does not use color information. To demonstrate effectiveness method, have conducted several experiments show support vector machines well suited