基于金字塔变换跨尺度代价聚合的立体匹配

系统仿真学报 ›› 2016, Vol. 28 ›› Issue (9): 2227-2234.

基于金字塔变换跨尺度代价聚合的立体匹配

姚莉^1,2,3, 刘助奎³, 王秉凤³

1.计算机网络和信息集成教育部重点实验室(东南大学),南京 211189;
2.计算机软件新技术国家重点实验室(南京大学),南京 210023;
3.东南大学计算机科学与工程学院,南京 211189

收稿日期:2016-05-26 修回日期:2016-07-14 出版日期:2016-09-08 发布日期:2020-08-14
作者简介:姚莉(1977-),女,湖北,博士,副教授,研究方向为计算机图形学与视觉。
基金资助:
国家高技术研究发展计划(863计划) (2015 AA015904), CCF-腾讯犀牛鸟基金(RAGR20150120), NSFC-广东联合基金（第二期）超级计算科学应用研究专项资助和国家超级计算广州中心支持

Stereo Matching based on Pyramid transform Cross-Scale Cost Aggregation

Yao Li^1,2,3, Liu Zhukui³, Wang Bingfeng³

1. Key Laboratory of Computer Network and Information Integration (Southeast University), Nanjing 211189, China;
2. State Key Laboratory for Novel Software Technology(Nanjing University), Nanjing 210032, China;
3. School of Computer Science and Engineering, Southeast University, Nanjing 211189, China

Received:2016-05-26 Revised:2016-07-14 Online:2016-09-08 Published:2020-08-14

摘要/Abstract

摘要： 人眼视觉系统在不同尺度上处理接收到的视觉信号,而传统的立体匹配算法都在原立体匹配图像对最大尺度下进行立体匹配,在低纹理区域、无纹理区域立体匹配错误率高。模拟人眼视觉系统在多个尺度上进行立体匹配,再融合不同尺度下的匹配结果能减少错误率。通过改进一种基于高斯金字塔变换跨尺度代价聚合的立体匹配方法,在原框架上,在代价聚合阶段加入拉普拉斯金字塔变换,在视差精细化阶段加入基于加权联合双边滤波的视差精细化方法,使新框架能获得比原论文更好的视差图。

关键词: 立体匹配, 跨尺度, 金字塔变换, 加权联合双边滤波, 视差精细化

Abstract: Human visual system processes the received visual signals on different scales, however, the traditional stereo matching algorithms get the disparity map from the original image in the biggest scale, which will lead to high stereo matching error rate in the area of low-texture, texture-less area. Simulation of the human visual system in stereo matching on multiple scales can reduce error rate. A stereo matching method based on the Gaussian pyramid cross-scale transform was improved, by adding the Laplace Pyramid Transform to the original cross-scale framework and adding the weighted joint bilateral filter in the disparity refinement stage. The new cross-scale framework can get a better disparity map than the original method.

Key words: stereo matching, cross-scale, pyramid transform, weighted joint bilateral filter, disparity refinement

中图分类号:

TP391.4

姚莉, 刘助奎, 王秉凤. 基于金字塔变换跨尺度代价聚合的立体匹配[J]. 系统仿真学报, 2016, 28(9): 2227-2234.

Yao Li, Liu Zhukui, Wang Bingfeng. Stereo Matching based on Pyramid transform Cross-Scale Cost Aggregation[J]. Journal of System Simulation, 2016, 28(9): 2227-2234.

参考文献 15

[1]	塞利斯基 (Szeliski, R.). 计算机视觉 [M]. 北京: 清华大学出版社, 2012: 410.
[2]	曹之乐, 严中红, 王洪. 双目立体视觉匹配技术综述[J]. 重庆理工大学学报(自然科学版), 2015, 29(2): 70-75.
[3]	Scharstein D, Szeliski R.A Taxonomy and Evaluation of Dense Two-Frame Stereo Correspondence Algorithms[C]// Stereo and Multi-Baseline Vision. USA: IEEE, 2001: 131-140.
[4]	Yoon K J, Kweon I S.Adaptive Support-Weight Approach for Correspondence Search[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence (S0162-8828), 2006, 28(4): 650-656.
[5]	Hosni A, Rhemann C, Bleyer M, et al.Fast Cost-Volume Filtering for Visual Correspondence and Beyond[J]. IEEE Transactions on Software Engineering (S0098-5589), 2013, 35(2): 504-511.
[6]	Yang Q.A non-local cost aggregation method for stereo matching[C]// IEEE Conference on Computer Vision and Pattern Recognition. USA: IEEE Computer Society, 2012: 1402-1409.
[7]	Mei X, Sun X, Dong W, et al.Segment-Tree based Cost Aggregation for Stereo Matching[C]// 2013 IEEE Conference on Computer Vision and Pattern Recognition, USA: IEEE, 2013: 313-320. DOI: 10.1109/CVPR. 2013. 47
[8]	才溪. 多尺度图像融合理论与方法 [M]. 北京: 电子工业出版社, 2014.
[9]	Zhang K, Fang Y, Min D, et al.Cross-Scale Cost Aggregation for Stereo Matching[C]// 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA. USA: IEEE, 2014: 1590-1597. DOI: 10.1109/CVPR.2014.206
[10]	X Qin, J Shen, X Mao, et al.Robust match fusion using optimization[J]. IEEE Transactions on Cybernetics (S2168-2267), 2015, 45(8): 1549-1560.
[11]	X Qin, J Shen, X Mao, et al.Structured-patch optimization for dense correspondence[J]. IEEE Transactions on Multimedia (S1520-9210), 2015, 17(3): 295-306.
[12]	Milanfar P.A Tour of Modern Image Filtering: New Insights and Methods, Both Practical and Theoretical[J]. IEEE Signal Processing Magazine (S1053-5888), 2013, 30(1): 106-128.
[13]	Matsuo T, Fukushima N, Ishibashi Y.Weighted joint bilateral filter with slope depth compensation filter for depth map refinement[C]// International Conference on Computer Vision Theory and Applications, 2013. USA: IEEE, 2013: 300-309.
[14]	Scharstein B D, Szeliski R. Middlebury stereo vision [DB/OL]. http://vision.middlebury.edu/stereo. 2010. 2015-12-05/ 2016-08-24.
[15]	Yoon K J, Kweon I S.Locally adaptive support-weight approach for visual correspondence search[C]// Computer Vision & Pattern Recognition. USA: IEEE Computer Society, 2005: 924-931.