Stereo matching has gained the popularity in computer vision and image processing. The objective of stereo correspondence matching is to obtain dense depth information of objects for 3D reconstruction and modeling. Several stereo correspondence algorithms have been developed in last couple of years. However, they are not suitable for real time applications due to their limitations of high computational cost. This paper proposes an efficient algorithm for stereo correspondence matching, which is fast and capable of tackling additive noise. The algorithm estimates average disparity within a search range and attains the benefit of mean filtering. Experimental evaluations demonstrates the effectiveness of our proposed algorithm comparable to the existing stereo methods.
|Number of pages||4|
|Journal||International Journal of Computer Theory and Engineering|
|Publication status||Published - Feb 2017|