Enhancing Stereo Matching With Classification
Open Access
- 22 May 2014
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 2, 485-499
Abstract
This paper presents a novel approach that employs classification to enhance the accuracy of the stereo matching problem. First, the images are treated in order to improve their pixel to pixel correspondence and reduce illumination differences. After that, stereo matching is addressed using different methods with emphasis on local ones like the sum of absolute distances and normalized cross correlation. Other state-of-the-art approaches are also considered. Then, and for every pixel, different features are computed from the input stereo image and the initially found depth map. Afterward, boosting and neural networks, as classification methods, are used to handle occlusion and enhance stereo matching by finding the erroneous disparity values. These values are then corrected through a completion stage. The accuracy of the proposed implementation improves on the problem in an efficient manner. A timing analysis of the method is provided to validate the real time performance. This paper further clarifies some of the possible developments based on various discussions.Keywords
Funding Information
- Lebanese National Council for Scientific Research
- Nvidia through the GPUs
This publication has 44 references indexed in Scilit:
- Biologically and psychophysically inspired adaptive support weights algorithm for stereo correspondenceRobotics and Autonomous Systems, 2010
- Stereo and motion correspondences using nonlinear optimization methodComputer Vision and Image Understanding, 2006
- Stereovision matching through support vector machinesPattern Recognition Letters, 2003
- A cooperative algorithm for stereo matching and occlusion detectionIEEE Transactions on Pattern Analysis and Machine Intelligence, 2000
- Additive logistic regression: a statistical view of boosting (With discussion and a rejoinder by the authors)The Annals of Statistics, 2000
- Improving stereovision matching through supervised learningPattern Analysis and Applications, 1998
- A pixel dissimilarity measure that is insensitive to image samplingIEEE Transactions on Pattern Analysis and Machine Intelligence, 1998
- A Decision-Theoretic Generalization of On-Line Learning and an Application to BoostingJournal of Computer and System Sciences, 1997
- A stereo matching algorithm with an adaptive window: theory and experimentIEEE Transactions on Pattern Analysis and Machine Intelligence, 1994
- Learning and feature selection in stereo matchingIEEE Transactions on Pattern Analysis and Machine Intelligence, 1994