Stereo Correspondence with Global and Local Traits

Global based and local based methods are two main aspects of stereo correspondence. Up to now, they both have achieved some significant results but still have their own problems. The global based approach mostly makes the result too smooth to include more details, while the local based method makes the result noisier. This paper proposes a novel method which combines the global and local traits as much as possible to fill the gap between them. Firstly, layered result is obtained by graph cut based algorithms. Secondly, several sub-pixel local results are obtained under global result´s guidance with window correlation method. Thirdly, these local results are repeatedly merged in the MRF model with a second order smooth prior energy function. Experiments show that this method can achieve better results than that of either global or local based method individually.