在立體匹配演算法中，區域性方法被認為是較為快速且兼具準度，其中又以十字生成的支持區域更為代表。故本論文將以十字生成為基礎，儘管原方法已經為平行化演算法，然而方法的形式若能使用到共享記憶體，將會再次提升其速度。所以原來不能以共享記憶體實現的步驟，在不改變其精神下做修改，以期望準度上不犧牲太多，且得到速度上的大幅提升。
另外，在序列影像中，總是有靜止的部分(如:背景)，隱含著我們並非需要每個畫面都算整張圖，靜止部分的資訊可由之前的計算得知。但是攝影機卻非一直靜止，幸運地攝影機的速度變化總是微小。在提出的方法中，我會在一段時間週期偵測一次攝影機速度，藉由攝影機速度的補償判斷該方格是否靜止，若為靜止我將複製上次計算的支持區域以減少計算量。
在面對高解析圖像的輸入，使用共享記憶體，會受限於硬體的規格，然而用到的資訊都是區域性的，我們仍可以切割為一塊塊，做完再拼起來，儘管其中有其額外負擔，相比共享記憶體所帶來的提速，仍是值得。

In stereo matching algorithm, the local methods are considered as the approach balancing the speed and accuracy, besides the cross-based one is outstanding of them. In this essay, it’ll be based on the cross-based, although it has been a parallel algorithm, and if the algorithm can be applied the shared memory, it is going to speed up again. These steps which cannot be applied shared memory in cross-based method will be improved under the same principle without sacrificing the accuracy much.
Moreover, in the sequential image, there’s always something stationary (e.g. background), and it implies that we don’t have to calculate for each frames, and the information of the stationary part can be known from the previous. But the camera won’t be stationary, however fortunately it changes slightly. In the proposed method, we’ll detect the camera motion in period, and judge whether stationary or not with camera motion compensation. For the stationary part, we won’t calculate its support region, but copy these from previous instead.
For high-resolution image input, it’ll be restricted by hardware if the shared memory is applied. However, the information of each pixel is local so that we can divide the image into tiles and handle them one by one, and then combine them. Although it leads the overhead, it’ll be still worthwhile to do, considering the speed-up by shared memory.

[1] C. Fehn, "Depth-image-based rendering (DIBR), compression, and transmission for a new approach on 3D-TV," Proc. SPIE 5291, Stereoscopic Displays and Virtual Reality Systems XI, 93, 21 5 2004.
[2] Y. Boykov, O. Veksler and R. Zabih, "Fast approximate energy minimization via graph cuts," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 23, no. 11, pp. 1222-1239, 2001.
[3] Yu-Chih Wang, Cheng-Ping Tung, Pau-Choo Chung, "Efficient disparity estimation using hierarchical bilateral disparity structure based graph cut algorithm with a foreground boundary refinement mechanism," Circuits and Systems for Video Technology, IEEE Transactions on, pp. 784-801, 5 2013.
[4] J. Sun, N.-N. Zheng and H.-Y. Shum, "Stereo Matching Using Belief Propagation," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 25, no. 7, pp. 787-800, July 2003.
[5] Q. Yang, L. Wang, R. Yang and H. Stewen, "Stereo Matching with Color-Weighted Correlation, Hierarchical Belief Propagation, and Occlusion Handling," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. vol.31, no. 3, pp. 492-504, 2009.
[6] O. Veksler, "Stereo correspondence by dynamic programming on a tree," IEEE Conference on Computer Vision and Pattern Recognition, vol. 2, pp. 384-390, 2005.
[7] C. Richardt, D. Orr, I. Davies, A. Criminisi and N. A. Dodgson, "Real-Time Spatiotemporal Stereo Matching Using the Dual-Cross-Bilateral Grid," European conference on CompurterVision, pp. 510-523, 2010.
[8] O. Faugeras, B. Hotz, H. Mathieu, T. Viéville, Z. Zhang, P. Fua, E. Théron, L. Moll, G. Berry, J. Vuillemin, P. Bertin and C. Proy, "Real time correlation-based stereo: algorithm, implementations and applications," INRIA Technical Report, p. 2013, 1993.
[9] C. Stentoumis, L. Grammatikopoulos, I. Kalisperakis, E. Petsa and G. Karras, "A Local Adaptive Approach for Dense Stereo Matching in Architectural Scene Reconstruction," in 3D Virtual Reconstruction and Visualization of Complex Architectures, Trento, Italy, 2013.
[10] K. J. Yoon, I. S. Kweon, "Adaptive support-weight approach for correspondence search," IEEE TPAMI, 28(4), pp. 650-656, 2006.
[11] Q. Yang, "A non-local cost aggregation method for stereo matching," Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference on, pp. 1402 - 1409, 6 2012.
[12] S. D. Sharghi and F. A. Kamangar, "Geometric Feature-based Matching in Stereo Images," Information, Decision and Control, pp. 65-70, 1999.
[13] C.-I. Kim and S.-Y. Park, "Fast Stereo Matching of Feature Links," International Conference on 3D Imaging, Modeling, Processing, Visualization and Transmission, pp. 268-274, 2011.
[14] K. Zhang, J. Lu and G. Lafruit, "Cross-Based Local Stereo Matching Using Orthogonal Integral Images," in IEEE Transactions on circuit and systems for video technology, vol. 19, no. 7, 2009.
[15] Xing Mei, Xun Sun, Mingcai Zhou, Shaohui Jiao, Haitao Wang, Xiaopeng Zhang, "On building an accurate stereo matching system on graphics hardware," Computer Vision Workshops (ICCV Workshops), 2011 IEEE International Conference on, pp. 467 - 474, 11 2011.
[16] H. Hirschm¨uller and D. Scharstein., "Evaluation of stereo matching costs on images with radiometric differences," IEEE TPAMI, pp. 1582-1599, 2009.
[17] G. Yao, Y. Liu, B. Lei and D. Ren, "A Rapid Stereo Matching Algorithm Based on Disparity Interpolation," Conference on Dependable Computing, pp. 5-10, 2010.
[18] M.V. Srinivasan;S. Venkatesh;R. Hosie, "Qualitative estimation of camera motion parameters from video sequences," Pattern Recognition, pp. 593-606, 4 1997.
[19] Tomasi C. , Manduchi R., "Bilateral filtering for gray and color images," Computer Vision, 1998. Sixth International Conference on, pp. 839-846, 1 1998.