Recovering Epipolar Geometry by Reactive Tabu Search

We propose a new approach to recover epipolar geometry from a pair of uncalibrated images. By minimizing a proposed cost function, our approach matches the detected feature points from an image pair, discards the outliers and recovers the epipolar geometry simutaneously. Experiments on real images show that this approach is effective and fast.

• Type: Conference
• Year: 1998
• Where: ICCV
• Authors: Qifa Ke, Gang Xu, Songde Ma
• Keywords: Computer Vision, ICCV 1998, Image Pair, Uncalibrated Images