GPU-based Stereo Matching Algorithm with the Strategy of Population-based Incremental Learning
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
To solve the general problems surrounding the application of genetic algorithms in stereo matching, two measures are proposed. Firstly, the strategy of simplified population-based incremental learning (PBIL) is adopted to reduce the problems with memory consumption and search inefficiency， and a scheme for controlling the distance of neighbors for disparity smoothness is inserted to obtain a wide-area consistency of disparities. In addition, an alternative version of the proposed algorithm, without the use of a probability vector, is also presented for simpler set-ups. Secondly, programmable graphics-hardware (GPU) consists of multiple multi-processors and has a powerful parallelism which can perform operations in parallel at low cost. Therefore, in order to decrease the running time further, a model of the proposed algorithm, which can be run on programmable graphics-hardware (GPU), is presented for the first time. The algorithms are implemented on the CPU as well as on the GPU and are evaluated by experiments. The experimental results show that the proposed algorithm offers better performance than traditional BMA methods with a deliberate relaxation and its modified version in terms of both running speed and stability. The comparison of computation times for the algorithm both on the GPU and the CPU shows that the former has more speed-up than the latter, the bigger the image size is.
Abstract
To solve the general problems surrounding the application of genetic algorithms in stereo matching, two measures are proposed. Firstly, the strategy of simplified population-based incremental learning (PBIL) is adopted to reduce the problems with memory consumption and search inefficiency， and a scheme for controlling the distance of neighbors for disparity smoothness is inserted to obtain a wide-area consistency of disparities. In addition, an alternative version of the proposed algorithm, without the use of a probability vector, is also presented for simpler set-ups. Secondly, programmable graphics-hardware (GPU) consists of multiple multi-processors and has a powerful parallelism which can perform operations in parallel at low cost. Therefore, in order to decrease the running time further, a model of the proposed algorithm, which can be run on programmable graphics-hardware (GPU), is presented for the first time. The algorithms are implemented on the CPU as well as on the GPU and are evaluated by experiments. The experimental results show that the proposed algorithm offers better performance than traditional BMA methods with a deliberate relaxation and its modified version in terms of both running speed and stability. The comparison of computation times for the algorithm both on the GPU and the CPU shows that the former has more speed-up than the latter, the bigger the image size is.
참고문헌 (19)
- D.Luebke, "GPGPU: general purpose computation on graphics hardware", ACM SIGGRAPH 2004 Course Notes, pp.33~ (2004)
- P. H. Winston, "Artificial Intelligence-3rd edition", pp.505~528 (1993), Addison-Wesley Publishing Co.
- Fang-Chih Tien, "Solving Line-Feature Stereo Matching with Genetic Algorithms in Hough Space", Journal of the Chinese Institute of Industrial Engineers, 21(5), pp.516~526 (2004)
- Régis Vaillant, "Genetic algorithms applied to binocular stereovision", Computer Vision- ECCV '94, 801, pp.193~198 (2006)
- Kyu-Phil Han, "A Simple Stereo Matching Algorithm Using PBIL and its Alternative", Korea Information Processing Society, 12-B(4), pp.429~436 (2005)
- Population-based Incremental Learning:A Method for Integrating Genetic Search Based Function Optimization and Competitive Learning, "Shumeet Baluja", Carnegie Mellon Univ., ~, pp.Technical reports CMU-CS-94-163 (1994)
- Q. Yang, "Stereo Matching with Color-Weighted correlation,hierarchical Belief Propagation and occlusion Handling", Proceedings of the 2006 IEEE Computer society Conference on Computer Vision and Pattern recognition(CVPR’06), pp.2347~2354 (2006)
- John R. Jordan, "Using Chromatic Information in Edge-based Stereo Correspondence", CVGIP:Image Understanding, 54(12), pp.98~118 (1991)
- John(Juyang)Weng, "Image Matching Using the Windowed Fourier Phase", International Journal of Computer Vision, 11(3), pp.211~236 (1993)
- Yong-Suk Kim, "Stereo Matching Algorithm Based on Modified Wavelet Decomposition Process", Pattern Recognition, 30, pp.929~952 (1997)
- Kyu-Phil Han, "Stereo Matching Using Genetic Algorithm with Adaptive Chromosomes", Pattern Recognition, 34(9), pp.1729~1740 (2001)
- Ondřej Fialka, "FFT and Convolution Performance in Image Filtering on GPU", Proceedings of the 10th International Conference on Information Visualisation, pp.609~614 (2006)
- Kenneth Moreland, "The FFT on a GPU", SIGGRAPH, 2003, 112~119
- J. KRÜGER, "Linear algebra operators for GPU implementation of numerical algorithms", ACM SIGGRAPH 2005 Courses, pp.234~ (2005), ACM
- J. BOLZ, "Sparse matrix solvers on the GPU: conjugate gradients and multigrid", ACM SIGGRAPH 2003 Papers, pp.917~924 (2003)
- M. Pollefeys, "Visual Modeling with a Hand-held Camera", International Journal of Computer Vision, 59(3), pp.207~232 (2004)
- D. V. Papadimitriou, "Epipolar Line Estimation and Rectification for Stereo Image Pairs", IEEE Transactions on Image Processing, 5(4), pp.672~676 (1996)
- Kyu-Phil Han, "Hybrid Stereo Matching with a New Relaxation Scheme of Preserving Disparity Discontinuity", Pattern Recognition, 33(5), pp.767~785 (2000)
- D. Marr, "A Computational Theory of Human Stereo Vision", Proc. Royal Soc., B204, pp.301~328 (1979)