Robust Structure and Motion Estimation by Auto-Scale Random Sample Consensus
Chen Tai, Yunhui Liu · 2006
This paper proposes a robust strategy for structure from planar motion estimation. To improve the robustness, an auto-scale random sample consensus (RANSAC) algorithm is adopted in the motion and structure estimation. So the algorithm can deal with a great number of outliers induced by an automatic matching computation. With the adoption of the auto-scale technique, the algorithm can be used in an entirely automatic application without any prior information or user set parameters. The contribution of this work is the development of an approach to make structure and motion estimation more robust and efficient so as to be applicable to real applications. The experiments indoor and outdoor have been done to verify the feasibility of the algorithm. In the experiments, the feature correspondences in the image sequences are extracted and refined automatically by the relation of the stereo cameras and the property of the planar motion. The results show the algorithm is robust and efficient for applications in planar motions.