This table lists the benchmark results for the low-res many-view scenario. The following metrics are evaluated:

(*) For exact definitions, detailing how potentially incomplete ground truth is taken into account, see our paper.

The datasets are grouped into different categories, and result averages are computed for a category and method if results of the method are available for all datasets within the category. Note that the category "all" includes both the high-res multi-view and the low-res many-view scenarios.

Methods with suffix _ROB may participate in the Robust Vision Challenge.

Click a dataset result cell to show a visualization of the reconstruction. For training datasets, ground truth and accuracy / completeness visualizations are also available. The visualizations may not work with mobile browsers.




Method Infoalllow-res
many-view
indooroutdoordelivery areaelectroforestplaygroundterrains
sort bysorted bysort bysort bysort bysort bysort bysort bysort by
DeepC-MVS_fast78.24 362.57 264.27 361.43 153.65 265.83 170.22 248.24 174.89 3
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
DeepC-MVS78.47 261.99 364.34 260.43 353.21 365.62 268.07 647.59 275.48 2
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
DeepPCF-MVS79.04 162.76 166.22 160.45 253.69 165.50 368.81 547.05 378.76 1
3Dnovator+75.73 458.40 459.88 557.41 446.92 762.89 470.54 138.80 1372.84 6
ACMM72.26 855.12 1254.88 1955.28 938.65 2061.75 660.21 1943.87 471.11 11
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
ACMP73.23 756.18 759.37 754.05 1047.31 561.10 858.60 2042.44 771.44 10
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
OpenMVScopyleft70.44 1055.58 1058.77 853.44 1244.82 1058.83 965.59 1035.91 1872.72 7
3Dnovator73.76 558.20 560.10 456.93 545.58 962.04 570.17 338.57 1474.61 5
PCF-MVS73.28 657.32 659.66 655.76 748.61 456.36 1367.24 843.68 570.70 12
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
PLCcopyleft68.99 1151.58 1755.07 1849.25 2041.73 1552.26 2156.95 2238.53 1568.41 17
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
ACMH+66.54 1353.99 1456.90 1252.05 1444.66 1157.37 1156.79 2341.98 869.14 15
ACMH65.37 1451.50 1853.46 2150.20 1838.46 2152.73 1957.24 2140.63 1068.46 16
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
TAPA-MVS71.42 955.13 1158.21 1053.07 1341.62 1656.79 1262.68 1539.75 1274.79 4
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
COLMAP_ROBcopyleft62.73 1549.91 2251.76 2248.68 2237.30 2252.31 2061.83 1731.91 2466.23 22
Johannes L. Schönberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
IB-MVS66.94 1255.84 955.92 1655.78 640.33 1761.21 768.84 437.29 1771.51 9
Christian Sormann, Mattia Rossi, Andreas Kuhn and Friedrich Fraundorfer: IB-MVS: An Iterative Algorithm for Deep Multi-View Stereo based on Binary Decisions. BMVC 2021
LTVRE_ROB59.44 1653.25 1656.08 1551.36 1639.82 1953.93 1662.24 1637.92 1672.34 8
Andreas Kuhn, Heiko Hirschmüller, Daniel Scharstein, Helmut Mayer: A TV Prior for High-Quality Scalable Multi-View Stereo Reconstruction. International Journal of Computer Vision 2016
CMPMVSbinary47.78 179.53 3423.73 330.07 340.00 340.00 340.20 340.00 3447.47 30
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
PMVScopyleft39.38 1822.01 2932.82 2814.80 295.84 3211.38 3026.05 296.98 3159.80 27
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
MVEpermissive19.12 1915.60 3122.59 3410.94 315.22 3314.12 298.87 339.84 2939.95 34
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
unsupervisedMVS_cas30.63 2831.53 2930.04 2821.80 2827.12 2840.58 2822.40 2841.26 33
BP-MVSNet50.87 2053.66 2049.01 2140.10 1849.76 2363.03 1434.23 2167.21 21
Christian Sormann, Patrick Knöbelreiter, Andreas Kuhn, Mattia Rossi, Thomas Pock, Friedrich Fraundorfer: BP-MVSNet: Belief-Propagation-Layers for Multi-View-Stereo. 3DV 2020
CasMVSNet(SR_A)41.59 2748.09 2337.27 2733.02 2427.16 2755.24 2529.40 2563.15 24
CasMVSNet(SR_B)49.66 2348.09 2350.71 1733.02 2455.30 1467.43 729.40 2563.15 24
TAPA-MVS(SR)56.09 856.78 1355.62 843.28 1358.11 1065.17 1143.60 670.29 13
CasMVSNet(base)49.00 2448.08 2549.61 1934.83 2353.10 1866.82 928.93 2761.32 26
GSE51.31 1958.55 946.48 2547.08 649.58 2454.32 2735.53 2070.02 14
LPCS50.41 2155.27 1747.16 2443.12 1451.06 2254.57 2635.86 1967.43 19
COLMAP(SR)54.99 1357.09 1153.58 1146.78 855.24 1563.74 1341.76 967.41 20
COLMAP(base)53.42 1556.15 1451.59 1544.02 1253.93 1660.80 1840.05 1168.28 18
dnet0.00 350.00 350.00 350.00 340.00 340.00 350.00 340.00 35
CIDER45.22 2641.50 2747.70 2328.55 2746.16 2564.03 1232.90 2254.46 28
Qingshan Xu and Wenbing Tao: Learning Inverse Depth Regression for Multi-View Stereo with Correlation Cost Volume. AAAI 2020
A-TVSNet + Gipumacopyleft45.36 2547.40 2644.00 2629.86 2643.85 2655.31 2432.84 2364.94 23
hgnet15.22 3224.38 319.11 326.26 295.91 3117.93 313.49 3242.51 31
example18.45 3028.74 3011.60 306.01 315.21 3322.21 307.37 3051.47 29
DPSNet15.22 3224.38 319.11 326.26 295.91 3117.93 313.49 3242.51 31