Quasi-Motion Extraction Utilizing Edge Field Analysis

M. Matsumoto and S. Hashimoto (Japan)


Vector field analysis, Edge field analysis, Feature extrac tion, Quasi-motion extraction


Recognizing the motion in the images is an important part of computer vision. In order to recognize the motion in the images, it is significant to distinguish three motion primi tives as translation, rotation and divergence that can be ex tracted based on the optical flow. In this paper, we intro duce a method for extracting quasi-motion utilizing edge flow in one image, namely edge field analysis. In general, the optical flow is estimated utilizing two or more succes sive images. On the other hand, this method directly esti mates translation, rotation and divergence not utilizing two or more images but utilizing only one image without the in formation of the optical flow of the motion. We firstly for mulate the problem of extracting the motion primitives as a vector field analysis of the velocity vector field. We then propose an algorithm for estimating three motion primi tives from one image regarding edge as pseudo-velocity vector. The experimental results in terms of estimating the eye of the storm, Whirlpool Galaxy and the center of the explosion are also demonstrated to evaluate the proposed method.

Important Links:

Go Back