Disparities Maps Generation Employing Multi-resolution Analysis and Perceptual Grouping

Inspired by human vision system, the stereo vision is a very important area in computer vision. It is related with the 3D information recovery from a pair of images. Although it has being studied for quite a long time, the stereo vision presents some difficult problems, related to the change of visualization perspective. Among the different problems originated from point of view changes, occlusions and ambiguities have special attention and compose the foundation of stereo problem, named correspondence problem. In the related works, the local techniques are fast, but produce poor results whilst the global techniques produce good results but are very computationally expensive. This work aims to give a contribution to the correspondence problem in stereo matching using a local approach with two strategies as solution to ambiguities and occlusions: the multiresolution analysis with images pyramids and the perceptual grouping weight, called Gestalt Theory in the psychology. The obtained results were closer to the ones generated by global techniques, with the advantage of requiring less computational complexity. The use of Gestalt Theory makes this a modern disparity estimation method, due to this theory has received special attention in computer vision researches.