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
indooroutdoorlakesidesand boxstorage roomstorage room 2tunnel
sort bysorted bysort bysort bysort bysort bysort bysort bysort by
DeepPCF-MVS96.37 292.61 189.11 194.94 195.58 195.07 185.62 192.60 194.17 1
DeepC-MVS_fast96.70 192.27 288.75 294.63 295.15 394.80 285.14 392.35 293.92 2
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
DeepC-MVS95.98 391.43 387.48 494.06 494.52 594.15 583.69 691.26 393.50 3
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
TAPA-MVS93.98 790.68 488.05 392.44 792.42 1093.52 885.24 290.85 691.36 14
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
TAPA-MVS(SR)90.57 587.48 492.63 692.64 893.17 1183.91 591.04 492.10 12
PCF-MVS93.45 1090.50 685.08 1394.11 395.54 293.44 1082.27 1187.88 1593.34 4
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
COLMAP(SR)90.17 785.79 1193.08 592.63 993.87 682.51 1089.07 1392.74 6
PLCcopyleft95.07 489.96 887.00 691.94 1290.65 1892.68 1583.01 890.98 592.48 7
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
COLMAP(base)89.91 986.37 892.28 991.56 1492.95 1382.53 990.21 792.33 8
ACMH+92.99 1389.84 1085.96 1092.43 892.28 1192.84 1481.89 1290.03 892.17 10
GSE89.80 1186.71 791.87 1390.87 1792.62 1683.49 789.92 992.11 11
ACMM93.85 889.42 1285.42 1292.10 1191.09 1692.98 1281.16 1489.68 1092.23 9
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
ACMP93.49 989.04 1384.30 1692.21 1090.34 1993.47 979.54 1889.07 1392.81 5
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
LTVRE_ROB92.95 1489.00 1486.32 990.78 1991.67 1389.54 2484.87 487.77 1691.13 16
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
ACMH92.88 1588.89 1584.55 1491.78 1491.37 1592.41 1779.88 1689.21 1191.56 13
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
COLMAP_ROBcopyleft93.27 1188.32 1684.41 1590.93 1889.38 2192.11 1879.72 1789.10 1291.28 15
Johannes L. Schönberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
3Dnovator94.51 587.43 1780.90 2191.78 1494.76 494.22 378.96 1982.85 2186.35 20
3Dnovator+94.38 687.07 1880.26 2291.61 1694.34 694.17 478.01 2082.52 2286.34 21
A-TVSNet + Gipumacopyleft86.96 1982.91 1889.65 2188.10 2391.71 1980.23 1585.60 2089.15 18
BP-MVSNet86.71 2081.68 1990.06 2091.85 1287.55 2975.75 2387.61 1790.77 17
Christian Sormann, Patrick Knöbelreiter, Andreas Kuhn, Mattia Rossi, Thomas Pock, Friedrich Fraundorfer: BP-MVSNet: Belief-Propagation-Layers for Multi-View-Stereo. 3DV 2020
OpenMVScopyleft93.04 1286.34 2179.08 2391.18 1793.50 793.77 776.56 2181.60 2586.26 22
LPCS85.52 2283.61 1786.79 2384.99 2988.49 2681.34 1385.87 1986.89 19
PVSNet_088.72 1782.59 2375.14 2787.55 2288.79 2288.38 2773.35 2676.93 2985.48 23
PVSNet91.96 1681.42 2474.82 2885.83 2687.90 2486.77 3074.82 2474.81 3382.83 24
R-MVSNet81.26 2577.44 2483.81 2882.82 3086.25 3174.52 2580.36 2782.37 25
CIDER80.48 2672.14 3086.05 2487.66 2589.77 2271.75 2772.53 3780.72 27
Qingshan Xu and Wenbing Tao: Learning Inverse Depth Regression for Multi-View Stereo with Correlation Cost Volume. AAAI 2020
test_120580.31 2775.46 2683.55 2986.48 2790.90 2070.38 2980.54 2673.26 38
test_112680.30 2871.93 3285.88 2589.75 2090.41 2170.18 3073.68 3577.50 31
OpenMVS_ROBcopyleft86.42 1880.25 2971.88 3385.82 2787.52 2689.56 2370.53 2873.24 3680.39 28
ANet-0.7577.73 3074.78 2979.70 3277.01 3684.71 3467.34 3482.21 2477.39 32
P-MVSNet77.69 3181.61 2075.07 3871.91 4277.33 4076.51 2286.71 1875.98 35
CPR_FA74.23 3272.06 3175.68 3674.30 4070.69 4668.60 3275.52 3182.05 26
ANet74.07 3368.26 3577.94 3577.01 3684.71 3464.75 3571.76 3872.09 39
AttMVS72.35 3468.13 3675.16 3770.98 4680.63 3668.09 3368.16 3973.87 37
Pnet-new-70.55 3553.66 4781.81 3181.27 3285.23 3351.15 4456.17 5378.92 30
CasMVSNet(SR_A)69.66 3655.07 4379.38 3380.08 3387.62 2846.56 4963.58 4670.44 40
Pnet_fast68.89 3747.01 5683.47 3081.84 3189.30 2538.61 5755.42 5479.28 29
CasMVSNet(base)68.63 3853.70 4678.58 3479.92 3485.96 3244.83 5162.57 4869.87 41
A1Net68.49 3975.64 2563.72 4863.53 5351.39 6168.88 3182.41 2376.23 34
MVSNet_plusplus64.09 4048.06 5474.77 3985.29 2862.00 5131.11 6165.01 4077.03 33
Pnet-blend64.02 4152.13 5171.95 4075.81 3878.00 3840.66 5463.61 4462.05 46
Pnet-blend++64.02 4152.13 5171.95 4075.81 3878.00 3840.66 5463.61 4462.05 46
unMVSv163.91 4362.59 3964.80 4766.39 4968.00 4961.62 3663.55 4760.01 49
MVSNet63.58 4456.25 4168.47 4666.00 5071.12 4348.36 4664.13 4168.29 43
MVSCRF63.13 4552.16 5070.45 4270.18 4775.83 4251.17 4353.16 5765.33 44
RMVSNet62.69 4669.01 3458.48 5669.15 4858.56 5559.72 3778.30 2847.75 59
metmvs_fine60.95 4766.50 3757.25 5864.00 5151.92 5959.04 3873.97 3455.84 54
Snet60.77 4847.18 5569.82 4377.90 3561.73 5239.61 5654.76 5569.82 42
Pnet-eth60.43 4965.87 3856.80 5963.59 5245.39 6455.58 4076.16 3061.43 48
MVSNet_++59.91 5045.04 5769.82 4374.28 4160.08 5314.97 6475.11 3275.10 36
hgnet59.89 5154.16 4463.70 4971.39 4370.87 4452.09 4156.24 5148.85 56
DPSNet59.89 5154.16 4463.70 4971.39 4370.87 4452.09 4156.24 5148.85 56
MVEpermissive62.14 2059.41 5360.38 4058.77 5550.39 6269.00 4856.98 3963.78 4356.91 53
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
test_112458.90 5444.17 5968.72 4562.60 5878.44 3738.12 5850.22 5865.10 45
MVSNet + Gipuma57.07 5553.08 4859.73 5462.91 5657.42 5748.41 4557.74 4958.88 50
CasMVSNet(SR_B)56.87 5655.95 4257.48 5756.06 5967.85 5047.86 4864.05 4248.53 58
F/T MVSNet+Gipuma55.78 5749.83 5359.76 5363.10 5457.51 5646.34 5053.31 5658.66 51
example55.06 5844.75 5861.93 5171.21 4569.85 4742.82 5346.68 5944.73 61
firsttry54.62 5952.74 4955.88 6054.90 6055.69 5848.02 4757.46 5057.06 52
PMVScopyleft61.03 2149.24 6032.69 6460.27 5262.70 5776.26 4142.99 5222.38 6441.86 62
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
unMVSmet48.01 6140.73 6152.86 6153.90 6158.72 5436.72 6044.74 6145.96 60
confMetMVS44.56 6241.82 6046.39 6246.63 6351.78 6037.38 5946.25 6040.77 63
test_1120copyleft40.25 6335.81 6343.21 6330.29 6550.38 6229.89 6341.72 6348.95 55
Cas-MVS_preliminary38.91 6436.73 6240.37 6437.57 6448.76 6330.59 6242.87 6234.76 64
CMPMVSbinary66.06 199.32 650.44 6615.23 656.17 6639.51 650.89 660.00 660.00 66
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
FADENet4.02 664.30 653.83 666.03 673.71 665.02 653.58 651.74 65
dnet0.00 670.00 670.00 670.00 680.00 670.00 670.00 660.00 66
UnsupFinetunedMVSNet63.10 54