多视密集匹配并行传播GPU-PatchMatch算法

doi:10.11947/j.AGCS.2020.20180459

测绘学报 ›› 2020, Vol. 49 ›› Issue (2): 181-190.doi: 10.11947/j.AGCS.2020.20180459

多视密集匹配并行传播GPU-PatchMatch算法

邓非^1,2, 颜青松¹, 肖腾¹

1. 武汉大学测绘学院, 湖北武汉 430079;
2. 自然资源部城市国土资源监测与仿真重点实验室, 广东深圳 518000

收稿日期:2018-10-10 修回日期:2019-06-15 发布日期:2020-03-03
通讯作者: 颜青松 E-mail:yanqs_whu@whu.edu.cn
作者简介:邓非(1976-),男,博士,教授,研究方向为摄影测量与计算机视觉。E-mail:fdeng@sgg.whu.edu.cn
基金资助:
自然资源部城市国土资源监测与仿真重点实验室开放基金（KF-2018-03-025）

A GPU-PatchMatch multi-view dense matching algorithm based on parallel propagation

DENG Fei^1,2, YAN Qingsong¹, XIAO Teng¹

1. School of Geodesy and Geomatics, Wuhan University, Wuhan 430079, China;
2. Key Laboratory of Urban Land Resources Monitoring and Simulation, MNR, Shenzhen 518000, China

Received:2018-10-10 Revised:2019-06-15 Published:2020-03-03
Supported by:
The Open Fund of Key Laboratory of Urban Land Resources Monitoring and Simulation, MNR (No. KF-2018-03-025)

摘要/Abstract

摘要： 针对多视密集匹配的效率较低的问题，提出了GPU-PatchMatch多视密集匹配算法。该算法使用GPU提高PatchMatch的计算效率；同时充分利用稀疏场景信息，对深度信息进行规则初始化；为提高传播效率，使用了金字塔红黑板并行传播深度信息。最后在DTU、Strecha和Vaihigen数据集上进行了试验，并与常用的多视密集匹配算法进行对比。试验结果表明，本文算法在重建效率上有较大提高，与CPU算法（PMVS、MVE、OpenMVS）相比有7倍以上提升，与GPU算法相比也有2.5倍以上提升，表明本文算法的有效性。

关键词: 三维重建, 多视密集匹, 块匹配, 并行计算

Abstract: Aiming at the problem of low efficiency of multi-view dense matching, a GPU-PatchMatch multi-view dense matching algorithm is proposed. The algorithm uses GPU to improve the computational efficiency of PatchMatch. At the same time, it also makes full use of sparse scene information to initialize the depth information. In addition, in order to improve the propagation efficiency, it uses the pyramid red-blackboard to propagate the depth information in parallel. Finally, the experiments are carried out on the DTU, Strecha and Vaihigen datasets, and compared with the commonly used multi-view dense matching algorithms. The results show that our algorithm has a significant improvement in reconstruction efficiency, which is 7 times higher than the CPU algorithm (such as PMVS, MVE and OpenMVS), and 2.5 times higher than the GPU algorithm (such as Gipuma), which proves the effectiveness of the proposed method.

Key words: 3D reconstruction, multi-view dense matching, PatchMatch, parallel computation

中图分类号:

P237

邓非, 颜青松, 肖腾. 多视密集匹配并行传播GPU-PatchMatch算法[J]. 测绘学报, 2020, 49(2): 181-190.

DENG Fei, YAN Qingsong, XIAO Teng. A GPU-PatchMatch multi-view dense matching algorithm based on parallel propagation[J]. Acta Geodaetica et Cartographica Sinica, 2020, 49(2): 181-190.

参考文献

[1] SEITZ S M, CURLESS B, DIEBEL J, et al. A comparison and evaluation of multi-view stereo reconstruction algorithms[C]//Proceedings of 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. New York:IEEE, 2006:519-528.
[2] REMONDINO F, SPERA M G, NOCERINO E, et al. State of the art in high density image matching[J]. Photogrammetric Record, 2014, 29(146):144-166.
[3] FURUKAWA Y, PONCE J. Accurate, dense, and robust multiview stereopsis[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(8):1362-1376.
[4] 赵红蕊, 陆胜寒. 基于特征尺度分布与对极几何约束的高清影像快速密集匹配方法[J]. 测绘学报, 2018, 47(6):790-798. DOI:10.11947/j.AGCS.2018.20170630. ZHAO Hongrui, LU Shenghan. Dense high-definition image matching strategy based on scale distribution of feature and geometric constrain[J]. Acta Geodaetica et Cartographica Sinica, 2018, 47(6):790-798. DOI:10.11947/j.AGCS.2018.20170630.
[5] 王竞雪, 朱庆, 王伟玺. 多匹配基元集成的多视影像密集匹配方法[J]. 测绘学报, 2013, 42(5):691-698. WANG Jingxue, ZHU Qing, WANG Weixi. A dense matching algorithm of multi-view image based on the integrated multiple matching primitives[J]. Acta Geodaetica et Cartographica Sinica, 2013, 42(5):691-698.
[6] GOESELE M, SNAVELY N, CURLESS B, et al. Multi-view stereo for community photo collections[C]//Proceedings of 2007 IEEE 11th International Conference on Computer Vision. Rio de Janeiro:IEEE, 2007:1-8.
[7] ROTHERMEL M, WENZEL K, FRITSCH D, et al. SURE:photogrammetric surface reconstruction from imagery[C]//Proceedings of LC3D Workshop. Berlin:ISPRS, 2012:1-9.
[8] SHEN Shuhan. Accurate multiple view 3D reconstruction using patch-based stereo for large-scale scenes[J]. IEEE Transactions on Image Processing, 2013, 22(5):1901-1914.
[9] GALLIANI S, LASINGER K, SCHINDLER K. Massively parallel multiview stereopsis by surface normal diffusion[C]//Proceedings of 2015 IEEE International Conference on Computer Vision. Santiago:IEEE, 2015:873-881.
[10] 闫利, 费亮, 陈长海, 等. 利用网络图进行高分辨率航空多视影像密集匹配[J]. 测绘学报, 2016, 45(10):1171-1181. DOI:10.11947/j.AGCS.2016.20160068. YAN Li, FEI Liang, CHEN Changhai, et al. A multi-view dense matching algorithm of high resolution aerial images based on graph network[J]. Acta Geodaetica et Cartographica Sinica, 2016, 45(10):1171-1181. DOI:10.11947/j.AGCS.2016.20160068.
[11] 吴军, 姚泽鑫, 程门门. 融合SIFT与SGM的倾斜航空影像密集匹配[J]. 遥感学报, 2015, 19(3):431-442. WU Jun, YAO Zexin, CHENG Menmen. Airborne oblique stereo image dense matching by integrating SIFT and SGM algorithm[J]. Journal of Remote Sensing, 2015, 19(3):431-442.
[12] TOLDO R, FANTINI F, GIONA L, et al. Accurate multiview stereo reconstruction with fast visibility integration and tight disparity bounding[C]//Proceedings of 3D Virtual Reconstruction and Visualization of Complex Architectures. Trento:ISPRS, 2013:243-249.
[13] KAR A, HÄNE C, MALIK J. Learning a multi-view stereo machine[C]//Proceedings of the 30st Conference on Neural Information Processing Systems. Long Beach:Neural Information Processing Systems Foundation, 2017:365-376.
[14] YAO Yao, LUO Zixin, LI Shiwei, et al. MVSNet:depth inference for unstructured multi-view stereo[C]//Proceedings of the 15th European Conference on Computer Vision. Munich:Springer, 2018:785-801.
[15] JI Mengqi, GALL J, ZHENG Haitian, et al. SurfaceNet:an end-to-end 3D neural network for multiview stereopsis[C]//Proceedings of 2017 IEEE International Conference on Computer Vision. Venice:IEEE, 2017:2326-2334.
[16] BARNES C, SHECHTMAN E, FINKELSTEIN A, et al. PatchMatch:a randomized correspondence algorithm for structural image editing[J]. ACM Transactions on Graphics (TOG), 2009, 28(3):24.
[17] BLEYER M, RHEMANN C, ROTHER C. PatchMatch stereo-stereo matching with slanted support windows[C]//Proceedings of 2011 22nd British Machine Vision Conference. Dundee:BMVA, 2011:1-11.
[18] HIRSCHMULLER H. Stereo processing by semiglobal matching and mutual information[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008, 30(2):328-341.
[19] 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.
[20] TOLA E, LEPETIT V, FUA P DAISY:an efficient dense descriptor applied to wide-baseline stereo[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(5):815-830.
[21] BOISSONNAT J D, DEVILLERS O, PION S, et al. Triangulations in CGAL[C]//Proceedings of the 16th Annual Symposium on Computational Geometry. Hong Kong, China:ACM, 2000:11-18.
[22] AANAES H, JENSEN R R, VOGIATZIS G, et al. Large-scale data for multiple-view stereopsis[J]. International Journal of Computer Vision, 2016, 120(2):153-168.
[23] STRECHA C, VON HANSEN W, VAN GOOL L, et al. On benchmarking camera calibration and multi-view stereo for high resolution imagery[C]//Proceedings of 2008 IEEE Conference on Computer Vision and Pattern Recognition. Anchorage:IEEE, 2008:1-8.
[24] CRAMER M. The DGPF-test on digital airborne camera evaluation-overview and test design[J]. Photogrammetrie-Fernerkundung-Geoinformation, 2010, 2010(2):73-82.
[25] ALTMAN S, XIAO Wen, GRAYSON B. Evaluation of LOW-COST terrestrial Photogrammetry for 3D reconstruction of complex buildings[C]//Proceedings of ISPRS Geospatial Week 2017. Wuhan:Copernicus GmbH, 2017:199-206.

多视密集匹配并行传播GPU-PatchMatch算法

A GPU-PatchMatch multi-view dense matching algorithm based on parallel propagation

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	邓非, 陈欣, 颜青松, 曲英杰. 带线约束的摄影测量网格变分精化算法[J]. 测绘学报, 2020, 49(4): 469-479.
[2]	张春森, 张会, 郭丙轩, 彭哲. 城市场景结构感知的网格模型简化算法[J]. 测绘学报, 2020, 49(3): 334-342.
[3]	朱艳, 颜青松, 曲英杰, 陈欣, 邓非. 影像一致性优化的视图选取策略[J]. 测绘学报, 2020, 49(11): 1463-1472.
[4]	张一, 姜挺, 江刚武, 余岸竹, 于英. 特征法视觉SLAM逆深度滤波的三维重建[J]. 测绘学报, 2019, 48(6): 708-717.
[5]	范军, 李涛, 左小清, 陈乾福, 张祥, 禄競. 利用参数独立分解的星载SAR干涉测量检校方法[J]. 测绘学报, 2019, 48(6): 737-746.
[6]	尚洋, 孙晓亮, 张跃强, 李由, 于起峰. 三维目标位姿跟踪与模型修正[J]. 测绘学报, 2018, 47(6): 799-808.
[7]	郑顺义, 何源, 徐刚, 王辰, 朱锋博. 三维点云数据实时管理的Hash map方法[J]. 测绘学报, 2018, 47(6): 825-832.
[8]	袁一, 程亮, 宗雯雯, 李舒怡, 李满春. 互联网众源照片的三维重建定位技术[J]. 测绘学报, 2018, 47(5): 631-643.
[9]	梁伟, 徐新禹, 李建成, 朱广彬. 联合EGM2008模型重力异常和GOCE观测数据构建超高阶地球重力场模型SGG-UGM-1[J]. 测绘学报, 2018, 47(4): 425-434.
[10]	赵传, 张保明, 陈小卫, 郭海涛, 卢俊. 一种利用点云邻域信息的建筑物屋顶面高精度自动提取方法[J]. 测绘学报, 2017, 46(9): 1123-1134.
[11]	郑茂腾, 周顺平, 熊小东, 朱俊峰. GPU并行区域网平差[J]. 测绘学报, 2017, 46(9): 1193-1201.
[12]	崔阳, 吕志平, 李林阳, 陈正生, 孙大双, 邝英才. GNSS大网双差模型并行快速解算方法[J]. 测绘学报, 2017, 46(7): 848-856.
[13]	林祥国, 张继贤. 架空输电线路机载激光雷达点云电力线三维重建[J]. 测绘学报, 2016, 45(3): 347-353.
[14]	谢传节, 龙舟, 马益杭, 由志杰. 多边形间空间关系查询的异构多核架构并行算法[J]. 测绘学报, 2016, 45(1): 119-126.
[15]	王萌萌, 黄国满, 花奋奋, 魏钜杰, 卢丽君. 机载双天线InSAR联合定标算法[J]. 测绘学报, 2014, 43(12): 1259-1265.