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
DeepC-MVS_fast98.69 196.77 195.30 197.74 198.24 297.85 392.74 297.86 197.13 1
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
DeepPCF-MVS98.18 396.68 295.10 297.74 198.30 197.78 592.41 397.79 297.12 2
DeepC-MVS98.35 296.09 394.20 1197.34 397.94 497.39 1190.85 1497.56 396.71 4
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
COLMAP(SR)96.07 494.39 897.19 596.86 897.92 291.54 1097.24 896.79 3
TAPA-MVS(SR)95.82 594.80 396.49 996.19 1297.58 1092.04 497.56 395.71 12
PCF-MVS97.08 1395.68 693.36 1597.23 498.16 396.93 1591.43 1195.29 2096.61 5
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
ACMM97.58 595.68 694.49 696.48 1096.12 1397.20 1492.04 496.94 1196.10 9
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
ACMH+97.24 995.61 893.77 1496.83 696.69 1097.71 790.70 1596.85 1296.09 10
TAPA-MVS97.07 1495.48 994.55 496.10 1295.68 1697.30 1291.73 897.37 695.32 15
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
COLMAP(base)95.42 1094.03 1396.35 1195.96 1496.92 1691.10 1396.97 996.16 8
PLCcopyleft97.94 495.35 1194.28 1096.06 1395.26 2096.59 1791.61 996.96 1096.34 7
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
ACMH97.28 795.29 1293.25 1696.65 796.32 1197.93 189.82 1796.68 1595.69 13
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
ACMP97.20 1095.11 1393.01 1996.52 895.68 1697.26 1389.74 1896.27 1796.61 5
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
LTVRE_ROB97.16 1195.10 1494.13 1295.75 1495.69 1595.73 2192.87 195.39 1995.83 11
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
GSE95.00 1594.33 995.45 1794.81 2196.53 1991.98 696.68 1595.02 17
A-TVSNet + Gipumacopyleft94.96 1694.54 595.24 1995.31 1996.58 1891.82 797.27 793.83 18
BP-MVSNet94.57 1793.20 1795.49 1596.77 994.34 2989.67 1996.74 1395.35 14
Christian Sormann, Patrick Knöbelreiter, Andreas Kuhn, Mattia Rossi, Thomas Pock, Friedrich Fraundorfer: BP-MVSNet: Belief-Propagation-Layers for Multi-View-Stereo. 3DV 2020
COLMAP_ROBcopyleft97.56 694.43 1893.09 1895.33 1894.55 2296.37 2089.45 2096.72 1495.07 16
Johannes L. Schönberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
IB-MVS95.67 1794.34 1994.41 794.29 2295.45 1897.77 691.37 1297.44 589.67 26
3Dnovator97.25 892.47 2087.97 2895.47 1697.93 597.85 386.02 2489.92 2890.64 22
3Dnovator+97.12 1292.17 2187.57 2995.24 1997.49 697.68 885.24 2789.91 2990.54 23
LPCS92.07 2292.93 2091.49 2790.32 3193.29 3190.66 1695.20 2190.87 21
OpenMVScopyleft96.50 1591.77 2386.65 3195.19 2197.38 797.66 984.60 3088.69 3390.53 24
R-MVSNet90.93 2489.80 2291.68 2690.81 3094.15 3087.67 2191.93 2690.07 25
PVSNet_094.43 1890.46 2585.77 3293.59 2393.82 2495.33 2484.62 2986.92 3591.61 20
test_112689.66 2688.06 2790.73 2993.40 2595.29 2584.99 2891.13 2783.50 34
test_120589.53 2789.08 2589.83 3292.63 2795.71 2285.34 2692.82 2581.15 40
CPR_FA89.35 2888.37 2690.00 3189.38 3387.69 3987.29 2389.46 3192.93 19
PVSNet96.02 1689.22 2984.53 3692.35 2493.29 2694.60 2885.85 2583.20 3889.17 27
CIDER88.55 3083.13 3892.16 2593.95 2395.61 2383.79 3382.47 3986.92 29
Qingshan Xu and Wenbing Tao: Learning Inverse Depth Regression for Multi-View Stereo with Correlation Cost Volume. AAAI 2020
ANet-0.7587.89 3189.58 2386.76 3384.37 4590.29 3584.51 3194.65 2285.61 32
OpenMVS_ROBcopyleft92.34 1987.46 3281.75 3991.27 2892.55 2895.21 2681.21 3782.29 4086.05 30
P-MVSNet85.35 3391.57 2181.21 4278.76 5282.87 4787.56 2295.58 1882.00 37
ANet85.16 3485.20 3485.14 3684.37 4590.29 3582.00 3588.39 3480.78 42
unMVSv183.89 3584.37 3783.58 3884.00 4985.63 4183.17 3485.56 3681.11 41
metmvs_fine83.50 3684.90 3582.56 4084.72 4081.80 5081.09 3888.72 3281.17 39
A1Net82.90 3789.39 2478.57 5185.19 3763.18 6484.36 3294.42 2387.34 28
RMVSNet80.62 3886.71 3076.57 5384.14 4778.09 5280.22 3993.20 2467.48 59
Pnet_fast80.38 3965.52 5690.29 3090.03 3294.91 2760.54 5570.51 5885.93 31
AttMVS80.32 4078.27 4081.68 4178.98 5186.16 4079.76 4076.79 5379.89 43
Pnet-new-79.46 4169.54 5286.08 3484.76 3990.31 3465.13 4973.95 5583.16 35
CasMVSNet(SR_A)79.00 4269.56 5185.29 3585.95 3592.82 3262.08 5277.05 4977.11 44
hgnet78.82 4376.20 4280.57 4684.50 4182.32 4875.56 4276.84 5074.88 48
DPSNet78.82 4376.20 4280.57 4684.50 4182.32 4875.56 4276.84 5074.88 48
Pnet-blend++78.13 4573.62 4681.13 4384.46 4388.20 3766.50 4680.74 4270.74 57
Pnet-blend78.13 4573.62 4681.13 4384.46 4388.20 3766.50 4680.74 4270.74 57
CasMVSNet(base)78.10 4768.51 5384.50 3785.47 3691.56 3359.79 5777.22 4876.47 45
Pnet-eth77.75 4885.52 3372.57 5875.57 5465.89 6281.44 3689.61 3076.26 46
MVSCRF77.21 4974.57 4578.98 5079.29 5083.87 4575.01 4474.12 5473.78 52
MVEpermissive76.82 2076.63 5077.52 4176.03 5570.40 6085.49 4375.68 4179.36 4472.20 54
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
MVSNet_plusplus76.50 5166.38 5483.25 3991.25 2973.79 5753.62 6379.14 4584.70 33
example74.48 5265.59 5580.41 4884.10 4883.16 4662.82 5168.36 6073.98 51
Snet74.39 5364.28 5881.13 4388.04 3473.18 6060.15 5668.41 5982.18 36
MVSNet73.65 5470.01 5076.07 5476.39 5377.59 5361.93 5378.09 4774.23 50
MVSNet + Gipuma72.61 5571.20 4873.54 5675.39 5573.46 5965.60 4876.81 5271.78 55
firsttry72.49 5675.56 4470.44 6068.80 6370.25 6169.31 4581.82 4172.28 53
MVSNet_++71.81 5759.04 6380.33 4984.86 3874.23 5633.06 6585.02 3781.88 38
test_112471.63 5863.97 5976.73 5269.39 6285.55 4256.73 5871.21 5675.26 47
CasMVSNet(SR_B)70.07 5971.18 4969.33 6169.73 6177.06 5464.06 5078.31 4661.20 62
F/T MVSNet+Gipuma69.80 6064.46 5773.36 5775.05 5673.57 5860.74 5468.18 6171.46 56
unMVSmet66.94 6160.67 6271.11 5971.91 5979.70 5154.62 6066.73 6261.74 61
confMetMVS63.51 6260.95 6165.21 6360.95 6476.48 5555.53 5966.38 6358.21 63
test_1120copyleft58.74 6357.03 6459.88 6449.42 6664.79 6349.65 6464.41 6465.45 60
PMVScopyleft70.75 2158.19 6443.22 6568.18 6273.06 5884.09 4454.23 6132.21 6547.37 64
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
Cas-MVS_preliminary56.71 6562.57 6052.80 6550.07 6561.96 6554.15 6271.00 5746.38 65
FADENet18.58 6615.51 6620.63 6621.45 6732.74 6715.09 6615.93 667.68 66
CMPMVSbinary69.68 2211.01 670.97 6717.70 6710.09 6843.01 661.93 670.00 670.00 67
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
dnet0.00 680.00 680.00 680.00 690.00 680.00 680.00 670.00 67
UnsupFinetunedMVSNet75.05 56