In this thesis, we present a novel tree-based dynamic programming (TDP) algorithm for efficient stereo reconstruction. We employ the geodesic distance transformation for tree construction, which results in sound image over-segmentation and can be easily parallelized on graphic processing unit (GPU). Instead of building a single tree to convey message in dynamic programming (DP), we construct multiple trees according to the image geodesic distance to allow for parallel message passing in DP. In addition to efficiency improvement, the proposed algorithm provides visually sound stereo reconstruction results. Compared with previous related approaches, our experimental results demonstrate superior performance of the proposed algorithm in terms of efficiency and accuracy.

在這一篇論文中，我們提出了一個以測地線之樹為基礎的動態規劃來達到快的雙眼立體重建的演算法。我們利用測地線的轉換來達到建樹的效果，此方法不但可以得到過渡的影像分割的結果也可以很容易的使用圖形處理器(GPU)去產生平行的效果。與其只建立一棵樹來傳遞訊息，我們將會建立多棵的測地線之樹來達至平行傳遞訊息的效果。除了提高效率外，我們的算法也提供了良好的雙眼立體重建的結果。與此前的相關辦法，我們的實驗結果著有著相當有效率的執行速度且準確性。

[1] A. Criminisi, T. Sharp and A. Blake, GeoS: geodesic image segmentation, in Proc. of European Conference on Computer Vision, 2008.
[2] A. Klaus, M. Sormann and K. Karner, Segment-based stereo matching using belief propagation and a self-adapting dissimilarity measure, in Proc. International Conference on Pattern Recognition, 2006.
[3] C. E. Leiserson and I. B. Mirman, How to survive the multicore software revolution, CILK ARTS, 2008.
[4] C. Lei, J. Selzer and Y. H. Yang, Region–tree based stereo using dynamic programming optimization, in Proc. IEEE Conference on Computer Vision and Pattern Recognition, 2006.
[5] D. Scharstein and R. Szeliski, A taxonomy and evaluation of dense two-frame stereo correspondence algorithms, International Journal of Computer Vision, 47:7-42, 2001.
[6] G. E. Blelloch, Prefix sum and their application. Technical Report CMU-CS-90-190, School of Computer Science, Carnegie Mellon University, 1990.
[7] http://developer.download.nvidia.com/compute/cuda/2_2/toolkit/docs/NVIDIA_CUDA_Programming_Guide_2.2.pdf
[8] http://www.middleburry.edu/stereo
[9] http://www.openmp.org/
[10] I. J. Cox, S. Hingorani, S. Rao and B. Maggs, A maximum likelihood stereo algorithm, Computer Vision and Image Understanding, 63:542-567, 1996.
[11] J. C. Kim, K. M. Lee, B. T. Choi and S. U. Lee, A dense stereo matching using two-pass dynamic programming with generalized ground control points, in Proc. IEEE Conference on Computer Vision and Pattern Recognition, 2005.
[12] J. Sun, N. N. Zheng and H. Y. Shum, Stereo mathing using belief propagation, IEEE Trans. on Pattern Analysis and Machine Intelligence, 25:787-800, 2003.
[13] L. Wang, M. Liao, M. Gong, R. Yang and D. Nist, High-quality real-time stereo using adaptive cost aggregation and dynamic programming, 3DPVT, 2006.
[14] M. Gong and Y. H. Yang, Fast stereo matching using reliability-based dynamic programming and consistency constraints, In Proc. of Int. Conf. on Computer Vision, 1: 610-617, 2003.
[15] N. Satish, M. Harris and M. Garland, Designing efficient sorting algorithms for manycore GPUs, Proc. 23rd IEEE Int’l Parallel & Distributed Processing Symposium, 2009.
[16] O. Grygorash, Y. Zhou, and Z. Jorgensen, Minimum spanning tree based clustering algorithms, ICTAI, 2006.
[17] O. Veksler, Stereo correspondence by dynamic Programming on a tree, in Proc. IEEE Conference on Computer Vision and Pattern Recognition, 2005.
[18] Q. Yang, L. Wang, R. Yang, S. Wang, M. Liao and D. Nist. Real-time global stereo matching using hierarchical belief propagation, BMVC, 2006.
[19] Q. X. Yang, L. Wang, R. Yang and H. Stewenius, Stereo Matching with color-weighted correlation, hierarchical belief, IEEE Trans. on Pattern Analysis and Machine Intelligence, 31:492-504, 2009.
[20] S. Birchfield and C. Tomasi, A pixel dissimilarity measure that is insensitive to image sampling, IEEE Trans. on Pattern Analysis and Machine Intelligence, 20:401-406, 1998.
[21] T. Cormen, C. Leiserson, and R. Rivest, Introduction to Algorithms, MIT Press, 1990.
[22] T. Kanade and M. Okutomi, A stereo matching algorithm with an adaptive window: theory and experiment, IEEE Trans. on Pattern Analysis and Machine Intelligence 16:920-932, 1994.
[23] V. Kolmogorov and R. Zabih, Multi-camera scene reconstruction via graph cuts, in Proc. of European Conference on Computer Vision, 2002.
[24] Y. Boykov, O. Veksler and R. Zabih, Fast approximate energy minimization via graph cuts. IEEE Trans. on Pattern Analysis and Machine Intelligence, 23:1222-1239, 2001.