Walking Motion Analysis Using 3D Non-Harmonic Analysis
The spatiotemporal spectra of a video that contains a moving object form a plane in the three-dimensional (3D) frequency domain. This plane, which is referred to as the theoretical motion plane, reflects the velocity, which is calculated from the slope, of the moving objects. In the present paper, we propose a high-resolution frequency analysis method, referred to as 3D non-harmonic analysis (NHA), which is only weakly influenced by the analysis window. In addition, we estimate the motion vectors of objects in a video using the plane-clustering method, in conjunction with the least-squares method (LSM), for 3D NHA spatiotemporal spectra. In experiments, we verify the accuracy of 3D NHA and its usefulness for a sequence containing complex motions, such as crossover motion, by comparison with 3D fast Fourier transform (FFT). The experimental results show that increasing the frequency resolution contributes to high-accuracy estimation of a motion plane.
Image motion analysis Least squares methods Clustering methods
Kenta Fujii Takaaki Ueda Shigeki Hirobayashi Toshio Yoshizawa Tadanobu Misawa
国际会议
2011 4th International Congress on Image and Signal Processing(第四届图像与信号处理国际学术会议 CISP 2011)
上海
英文
456-460
2011-10-15（万方平台首次上网日期，不代表论文的发表时间）