Ego-motion analysis using average image data intensity
Kojiro Kato, Kris Kitani, Takuya Nojima · 2011
In this paper, we present a new method to perform ego-motion analysis using intensity averaging of image data. The method can estimate general motions from two sequential images on pixel plane by calculating cross correlations. With distance information between camera and objects, this method also enables estimates of camera motion. This method is sufficiently robust even for out of focus image and the calculational overhead is quite low because it uses a simple averaging method. In the future, this method could be used to measure fast motions such as human head tracking, or robot movement. We present a detailed description of the proposed method, and experimental results demonstrating its basic capability. With these results, we verify that our proposed system can detect camera motion even with blurred images. Furthermore, we confirm that it can operate at up to 714 FPS in calculating one dimensional translation motion.