Please wait a minute...
FITEE
Front. Inform. Technol. Electron. Eng.  2012, Vol. 13 Issue (6): 472-482    DOI: 10.1631/jzus.C1100311
    
Accurate real-time stereo correspondence using intra- and inter-scanline optimization
Li Yao, Dong-xiao Li, Jing Zhang, Liang-hao Wang, Ming Zhang
Institute of Information and Communication Engineering, Zhejiang University, Hangzhou 310027, China; Zhejiang Provincial Key Laboratory of Information Network Technology, Hangzhou 310027, China
Accurate real-time stereo correspondence using intra- and inter-scanline optimization
Li Yao, Dong-xiao Li, Jing Zhang, Liang-hao Wang, Ming Zhang
Institute of Information and Communication Engineering, Zhejiang University, Hangzhou 310027, China; Zhejiang Provincial Key Laboratory of Information Network Technology, Hangzhou 310027, China
 全文: PDF 
摘要： This paper deals with a novel stereo algorithm that can generate accurate dense disparity maps in real time. The algorithm employs an effective cross-based variable support aggregation strategy within a scanline optimization framework. Rather than matching intensities directly, the use of adaptive support aggregation allows for precisely handling the weak textured regions as well as depth discontinuities. To improve the disparity results with global reasoning, we reformulate the energy function on a tree structure over the whole 2D image area, as opposed to dynamic programming of individual scanlines. By applying both intra- and inter-scanline optimizations, the algorithm reduces the typical ‘streaking’ artifact while maintaining high computational efficiency. The experimental results are evaluated on the Middlebury stereo dataset, showing that our approach is among the best for all real-time approaches. We implement the algorithm on a commodity graphics card with CUDA architecture, running at about 35 fames/s for a typical stereo pair with a resolution of 384×288 and 16 disparity levels.
关键词： Stereo correspondence;  Scanline optimization;  Real-time    
Abstract: This paper deals with a novel stereo algorithm that can generate accurate dense disparity maps in real time. The algorithm employs an effective cross-based variable support aggregation strategy within a scanline optimization framework. Rather than matching intensities directly, the use of adaptive support aggregation allows for precisely handling the weak textured regions as well as depth discontinuities. To improve the disparity results with global reasoning, we reformulate the energy function on a tree structure over the whole 2D image area, as opposed to dynamic programming of individual scanlines. By applying both intra- and inter-scanline optimizations, the algorithm reduces the typical ‘streaking’ artifact while maintaining high computational efficiency. The experimental results are evaluated on the Middlebury stereo dataset, showing that our approach is among the best for all real-time approaches. We implement the algorithm on a commodity graphics card with CUDA architecture, running at about 35 fames/s for a typical stereo pair with a resolution of 384×288 and 16 disparity levels.
Key words: Stereo correspondence    Scanline optimization    Real-time
收稿日期: 2011-10-23 出版日期: 2012-06-05
CLC:  TN911.73  
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章　　
Li Yao
Dong-xiao Li
Jing Zhang
Liang-hao Wang
Ming Zhang

引用本文:

Li Yao, Dong-xiao Li, Jing Zhang, Liang-hao Wang, Ming Zhang. Accurate real-time stereo correspondence using intra- and inter-scanline optimization. Front. Inform. Technol. Electron. Eng., 2012, 13(6): 472-482.

链接本文:

http://www.zjujournals.com/xueshu/fitee/CN/10.1631/jzus.C1100311    或    http://www.zjujournals.com/xueshu/fitee/CN/Y2012/V13/I6/472

[1] Xiao-hu Ma, Yan-qi Tan, Gang-min Zheng. A fast classification scheme and its application to face recognition[J]. Front. Inform. Technol. Electron. Eng., 2013, 14(7): 561-572.
[2] Zhi-bo Wang, Zhi Wang, Hong-long Chen, Jian-feng Li, Hong-bin Li, Jie Shen. HierTrack: an energy-efficient cluster-based target tracking system for wireless sensor networks[J]. Front. Inform. Technol. Electron. Eng., 2013, 14(6): 395-406.
[3] Li Lu, Dong-qin Feng, Jian Chu. Improving the real-time performance of Ethernet for plant automation (EPA) based industrial networks[J]. Front. Inform. Technol. Electron. Eng., 2013, 14(6): 433-448.
[4] Ali Uysal, Raif Bayir. Real-time condition monitoring and fault diagnosis in switched reluctance motors with Kohonen neural network[J]. Front. Inform. Technol. Electron. Eng., 2013, 14(12): 941-952.
[5] Lei Zhang, Mattias Lampe, Zhi Wang. A hybrid genetic algorithm to optimize device allocation in industrial Ethernet networks with real-time constraints[J]. Front. Inform. Technol. Electron. Eng., 2011, 12(12): 965-975.