在傳統的立體匹配演算法中，全域方法被認為是比較精準的，尤其在圖中的遮蔽區域表現較為出色，當然也要耗費較多時間。另一方面，區域方法速度較快，但是精確度略嫌不足，而且相當容易受到圖片雜訊干擾。
本論文提出一種較為新穎的方式來求得兩張圖的深度圖。沿用區域方法的基本架構，但是在中間做成本匯聚時我們採用一種類似全域整合步驟的方式。這個成本匯聚的做法是利用雙邊濾波器的概念來產生權重圖，再使用權重圖找出點與點間的關係來加總，最終目的是讓同一物體內的所有點的對應成本能互通有無。
得到初始的深度圖後，我們利用左右兩張深度圖來檢查出是否有遮蔽處或是其他無法對應到的點。接著找出鄰近的最小深度值來填補遮蔽區，最後用雙邊濾波器洗掉其他雜訊。
雖然以上整套演算法流程是設計給較低階的硬體使用的，例如智慧型手機，但是它依然可以用GPU或雲端機房來加速。當我們在高階硬體上使用時，表現並不比其他演算法差。

In traditional stereo matching method, global method is more accurate but spending more time, and have more correct rate in occlusion area. On the contrary, local method is usually fast but have bad performance, and easily is influence by noise.
This paper proposed a novel method to compute disparity between two images. It is based on local method, but its cost be aggregated in like-global way. This aggregation is processed by a weight map which created by bilateral filter concept. Every pixel transfer its own cost information to all pixels on the same object, but this information would be restricted by weight map.
After finishing preliminary depth map, we use L-R check to find occlusion and mismatch pixel. Then fix occlusion by the smallest disparity nearby. At last, we use bilateral filter clean up whole depth map.
All of above computing process can be parallelized on GPU machine or cloud sever. Although this algorithm is designed for low-level machine, it still exerts high performance in the world of high-level hardware.

[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] D. Scharstein, R. Szeliski, "Middlebury stereo evaluation," [Online]. Available: http://vision.middlebury.edu/stereo/eval/. [Accessed 2 2010].
[3] Jian Sun, Nan-Ning Zheng, Heung-Yeung Shum, "Stereo matching using belief propagation," Pattern Analysis and Machine Intelligence, IEEE Transactions on (Volume:25 , Issue: 7 ), pp. 787 - 800, 7 2003.
[4] Kolmogorov V. , Zabih R., "Computing visual correspondence with occlusions using graph cuts," Computer Vision, 2001. ICCV 2001. Proceedings. Eighth IEEE International Conference on, pp. 508 - 515 vol.2, 7 2001.
[5] A. Klaus, M. Sormann, K. Karner, "Segment-based stereo matching using belief propagation and a self-Adapting dissimilarity measure," Pattern Recognition, 2006. ICPR 2006. 18th International Conference on (Volume:3 ), pp. 15-18, 2006.
[6] Q. Yang, C. Engels, A. Akbarzadeh, "Near real-time stereo for weakly-textured scenes," Proc. BMVC, pp. 80-87, 2008.
[7] Q. Yang, L. Wang, R. Yang, H. Stewenius, D. Nister, "Stereo matching with color-weighted correlation, hierarchical belief propagation, and occlusion handling," Pattern Analysis and Machine Intelligence, IEEE Transactions on (Volume:31 , Issue: 3 ), pp. 492-504, 2009.
[8] 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.
[9] Tomasi C. , Manduchi R., "Bilateral filtering for gray and color images," Computer Vision, 1998. Sixth International Conference on, pp. 839-846, 1 1998.
[10] Jiangyu Liu, Jian Sun, "Parallel graph-cuts by adaptive bottom-up merging," Computer Vision and Pattern Recognition (CVPR), 2010 IEEE Conference on, pp. 2181-2188, 13 6 2010.
[11] H. Hirschmuller, "Accurate and efficient stereo processing by semi-global matching and mutual information," Computer Vision and Pattern Recognition, 2005. CVPR 2005. IEEE Computer Society Conference on (Volume:2 ), pp. 807 - 814 vol. 2, 6 2005.
[12] Fife W.S., Archibald J.K., "Improved census transforms for resource-optimized stereo vision," Circuits and Systems for Video Technology, IEEE Transactions on (Volume:23 , Issue: 1 ), pp. 60-73, 1 2013.
[13] 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.
[14] A. Hosni, M. Bleyer, M. Gelautz, "Near real-time stereo with adaptive support weight approaches," Proc. 3DPVT, 2010.
[15] A. Hosni, M. Bleyer, M. Gelautz, C. Rheman, "Local stereo matching using geodesic support weights," Proc. ICIP, pp. 2093-2096, 2009.
[16] Federico Tombari, Stefano Mattoccia, Luigi Di Stefano, "Segmentation-based adaptive support for accurate stereo correspondence," Proc. PSIVT, pp. 427-438, 2007.
[17] K. J. Yoon, I. S. Kweon, "Adaptive support-weight approach for correspondence search," IEEE TPAMI, 28(4), pp. 650-656, 2006.
[18] Y. Wei, C. Tsuhan, F. Franz, C. H. James, "High performance stereo vision designed for massively data parallel platforms," IEEE TCSVT, 99, pp. 1-11, 2010.
[19] 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.
[20] C. John, "A computational approach to edge detection," Pattern Analysis and Machine Intelligence, IEEE Transactions on, pp. 679 - 698, 11 1986.