Multi-scale sparse feature point correspondence by graph cuts
This paper presents a global stereo sparse matching technique based on graph cut theory to obtain accurate correspondence of stereo point pair in vision measurement applications. First, in order to obtain accurate location of feature points, a new multi-scale corner detection algorithm is proposed where wavelet coefficients are used to determine feature points by calculating an auto-correlation matrix. A sparse graph is constructed based on the feature points according to the graph cut theory. Then, the feature point correspondence problem is transformed into a labeling problem in the sparse graph which can be solved by energy minimization. Multi-scale analysis is utilized to improve the precision of matching results. It was found that the use of sparse feature points in the construction of the graph can lead to both a simple graph structure and a reduced computational complexity. It was also found that node labeling in the graph can be performed using fewer disparity values instead of all disparity values. Our experimental results show that the new global stereo sparse matching technique can obtain more accurate results than the existing techniques.
Keywordsmulti-scale feature point extraction graph cut theory stereo correspondence sparse point matching
Unable to display preview. Download preview PDF.
- 1.Olga V. Semi-dense stereo correspondence with dense features. In: Proceedings of IEEE Workshop on Stereo and Multi-Baseline Vision, Kauai, HI, 2001. 149–157Google Scholar
- 2.Ping L, Dirk F. Contrast-invariant feature point correspondence. In: Proceedings of IEEE International Conference ICASSP, Honolulu, HI, 2007. I–477, I-488Google Scholar
- 3.Beltran J R, Garcia L J. Edge detection and classification using Mallat’s wavelet. Image Process, 1994. 1: 293–297Google Scholar
- 6.Roy S, Cox I. A maximum-flow formulation of the n-camera stereo correspondence problem. In: Proceedings of the Sixth International Conference on Computer Vision, Bombay, 1998. 492–499Google Scholar
- 7.Huq S, Koschan A. Efficient BP stereo with automatic paramemeter estimation. In: Proceedings of ICIP, San Diego, CA, 2008. 301–304Google Scholar
- 9.Mokhtarian F, Suomela R. Curvature scale space for robust image corner detection. In: Proceedings of the 14th International Conference on Pattern Recognition, Brisbane, Qld. 1998. 1819–1821Google Scholar
- 10.Harris C, Stephens M. A combined corner and edge detector. In: Proceedings of the 4th Alvey Vsion Conference, Manchester, 1988. 189–192Google Scholar
- 13.Pedersini F, Pozzoli E, Sarti A. Multi-resolution corner detection. In: Proceedings of IEEE International Conference ICIP, 2000. 150–153Google Scholar
- 14.Gao X, Sattar F. Corner detection of gray level images using Gabor wavelets. In: Proceedings of IEEE International Conference ICIP, Singapore, 2004. 2669–2672Google Scholar
- 15.Flore F. A fast method to improve the stability of interest point detection under illumination changes. In: Proceedings of IEEE International Conference ICIP, Singapore, 2004. 2673–2676Google Scholar
- 16.Swee E G T, Elangovan S. Applications of symlets for denoising and load forecasting. In: Proceedings of the IEEE Signal Processing Workshop on High-Order Statist, Caesarea, 1999. 165–169Google Scholar
- 18.Kolmogorov V, Zabih R. Computing visual correspondence with occlusions using graph cuts. In: Proceedings of Int’l Conf. Computer vision, Vancouver, BC, 2001. 508–515Google Scholar