全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

GPU近实时线性双目立体代价聚合

DOI: 10.11834/jig.20141010

Keywords: 双目视觉,代价聚合,GPU通用计算,并行计算

Full-Text   Cite this paper   Add to My Lib

Abstract:

目的近年来双目视觉领域的研究重点逐步转而关注其“实时化”策略的研究,而立体代价聚合是双目视觉中最为复杂且最为耗时的步骤,为此,提出一种基于GPU通用计算(GPGPU)技术的近实时双目立体代价聚合算法。方法选用一种匹配精度接近于全局匹配算法的局部算法――线性立体匹配算法(linearstereomatching)作为代价聚合策略;结合线性代价聚合的原理,对其主要步骤(代价计算、均值滤波及系数求解等)的计算流程进行有针对性地并行优化。结果对于相同的实验样本,用本文方法在NVIDAGTX780实验平台上能在更短的时间计算出代价矩阵,与原有的CPU实现方法相比,代价聚合的效率平均有了数十倍的提升。结论实时双目立体代价聚合方法,为在个人通用PC平台上实时获取高质量双目视觉深度信息提供了一个高效可靠的途径。

References

[1]  Mattoccia S. Stereo vision: algorithms and applications [EB/OL].[2013-11-20]. http://www.vision.deis.unibo.it/smatt/.
[2]  Scharstein D, Szeliski R. Middlebury stereo vision [DB/OL].[2013-11-20]. http://vision.middlebury.edu/stereo.
[3]  Scharstein D, Szeliski R. A taxonomy and evaluation of dense two-frame stereo correspondence algorithms[J]. International Journal of Computer Vision, 2002, 47(1-3): 7-42. [DOI: 10.1023/A:1014573219977]
[4]  Szeliski R. Computer Vision: Algorithms and Applications [M]. Berlin Heidelberg: Springer, 2011:467-499.[DOI: 10.1007/978-1-84882-935-0]
[5]  Hirschmuller H. Accurate and efficient stereo processing by semi-global matching and mutual information[C] //Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. San Diego, CA: IEEE, 2005(2):807-814. [DOI: 10.1109/CVPR.2005.56]
[6]  Hirschmuller H. Stereo processing by semi- global matching and mutual information[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008, 30(2): 328-341. [DOI: 10.1109/TPAMI.2007.1166]
[7]  Mattoccia S, Giardino S, Gambini A. Accurate and efficient cost aggregation strategy for stereo correspondence based on approximated joint bilateral filtering[C]// Proceedings of Computer Vision -ACCV 2009. Berlin Heidelberg: Springer, 2010: 371-380. [DOI: 10.1007/978-3-642-12304-7_35]
[8]  Hosni A, Bleyer M, Gelautz M, et al. Local stereo matching using geodesic support weights[C]// Proceedings of IEEE International Conference on Image Processing. Cairo, Egypt: IEEE, 2009: 2093-2096. [DOI: 10.1109/ICIP.2009.5414478]
[9]  Wang L, Liao M, Gong M, et al. High-quality real-time stereo using adaptive cost aggregation and dynamic programming[C]//Proceedings of 3D Data Processing, Visualization, and Transmission. Chapel Hill: IEEE, 2006: 798-805. [DOI: 10.1109/3DPVT.2006.75]
[10]  Yoon K J, Kweon I S. Adaptive support-weight approach for correspondence search[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006, 28(4): 650-656. [DOI: 10.1109/TPAMI.2006.70]
[11]  Darabiha A, Rose J, Maclean J W. Video-rate stereo depth measurement on programmable hardware[C]// Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Los Alamitos, California: IEEE, 2003(1): 203-210. [DOI: 10.1109/CVPR.2003.1211355]
[12]  Park S, Jeong H. Real-time stereo vision FPGA chip with low error rate[C]//Proceedings of International Conference on Multimedia and Ubiquitous Engineering. Seoul, Korea: IEEE, 2007: 751-756. [DOI: 10.1109/MUE.2007.180]
[13]  Gehrig S K, Eberli F, Meyer T. A real-time low-power stereo vision engine using semi-global matching[J]. Computer Vision Systems, 2009: 134 -143. [DOI: 10.1007/978-3-642-04667-4_14]
[14]  De-Maeztu L, Mattoccia S, Villanueva A, et al. Linear stereo matching[C]//Proceedings of IEEE International Conference on Computer Vision. Barcelona, Spain: IEEE, 2011: 1708-1715. [DOI: 10.1109/ICCV.2011.6126434]
[15]  Porikli F. Constant time O (1) bilateral filtering[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Anchorage, AK: IEEE, 2008: 1-8. [DOI: 10.1109/CVPR.2008.4587843]
[16]  Ju M H, Kang H B. Constant time stereo matching[C]// Proceedings of Machine Vision and Image Processing Conference. Dublin, Ireland: IEEE,2009:13-17.[DOI: 10.1109/IMVIP.2009.10 ]
[17]  He K, Sun J, Tang X. Guided image filtering[C]// Proceedings of Computer Vision-ECCV. Berlin Heidelberg: Springer, 2010: 1-14. [DOI: 10.1007/978-3-642-15549-9_1]
[18]  NVIDIA. CUDA C programming guide [EB/ OL].[2013-11-20]. http://www.NVIDIA. com.
[19]  Crow F C. Summed-area tables for texture mapping[C]//Proceedings of ACM SIGGRAPH Computer Graphics. New York, NY: ACM, 1984, 18(3): 207-212. [DOI: 10.1145/800031.808600]
[20]  Podlozhnyuk V. Image convolution with CUDA [J]. NVIDIA Corporation white paper, 2007, 2097(3): 9-14.
[21]  更多...
[22]  Mattoccia S, Viti M, Ries F. Near real-time fast bilateral stereo on the GPU[C]//Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops. Colorado Springs, CO: IEEE, 2011: 136-143. [DOI: 10.1109/CVPRW.2011.5981835 ]
[23]  Chen B, Chen H. A realization of semi-global matching stereo algorithm on GPU for real-time application[C]//Seventh International Symposium on Multispectral Image Processing and Pattern Recognition. Guilin, China: SPIE, 2011,8004:1-9.[DOI: 10.1117/12.902015]

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133