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 bysort bysort bysorted bysort bysort bysort bysort bysort by
DeepC-MVS_fast79.65 362.24 350.96 469.75 169.91 371.08 246.04 355.88 468.27 1
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
DeepPCF-MVS80.84 163.41 153.95 169.71 269.72 471.26 148.73 159.18 168.16 2
DeepC-MVS79.81 262.37 252.77 268.77 368.65 570.30 346.95 258.59 367.37 3
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
3Dnovator+77.84 459.53 447.18 767.76 470.63 167.68 443.34 751.01 964.98 6
3Dnovator76.31 559.36 548.19 566.81 570.04 267.59 544.27 552.10 762.80 10
ACMP74.13 656.41 844.60 1264.28 661.46 1465.81 737.07 1852.13 665.56 5
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
COLMAP(SR)56.21 944.16 1364.25 763.87 962.07 1441.86 1046.45 1666.80 4
OpenMVScopyleft72.83 1056.18 1045.66 963.19 864.67 865.13 839.60 1451.71 859.76 15
PCF-MVS73.52 757.06 748.10 663.03 966.85 762.36 1343.32 852.89 559.86 14
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
TAPA-MVS(SR)55.07 1243.33 1562.90 1067.41 657.81 2139.94 1346.71 1563.49 8
TAPA-MVS73.13 958.67 652.34 362.89 1163.68 1063.46 1145.85 458.83 261.53 11
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
ACMM73.20 855.01 1343.19 1662.89 1159.60 1766.07 635.89 2050.48 1063.01 9
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
COLMAP(base)55.51 1144.97 1162.54 1360.85 1562.46 1242.24 947.71 1264.29 7
GSE52.52 1640.81 1860.34 1461.71 1358.23 2038.13 1743.48 1961.07 13
LTVRE_ROB69.57 1253.52 1445.46 1058.89 1558.76 1860.60 1843.91 647.01 1457.32 20
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
COLMAP_ROBcopyleft66.92 1552.32 1742.45 1758.89 1556.18 2561.09 1538.61 1546.28 1759.41 16
Johannes L. Schönberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
PLCcopyleft70.83 1153.50 1545.79 858.64 1756.31 2460.76 1641.70 1149.88 1158.86 17
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
CasMVSNet(SR_A)45.13 2725.99 3357.89 1856.32 2364.81 919.76 3632.23 3252.54 26
CasMVSNet(base)44.49 2824.93 3457.53 1956.38 2264.76 1018.64 3831.23 3451.43 27
ACMH+68.96 1352.02 1843.83 1457.49 2059.85 1651.27 3140.04 1247.62 1361.35 12
LPCS50.60 1940.43 1957.38 2158.37 1956.00 2438.14 1642.71 2157.77 18
test_112645.72 2628.57 3057.15 2262.33 1159.06 1924.22 3032.91 3150.08 28
Pnet-new-42.61 3122.17 3956.24 2358.21 2053.03 2824.08 3120.25 4657.48 19
OpenMVS_ROBcopyleft64.09 1748.56 2038.68 2155.15 2461.90 1249.65 3336.02 1941.33 2353.89 24
PVSNet_057.27 1846.73 2334.96 2654.57 2555.50 2651.93 2931.22 2638.71 2556.27 22
ACMH67.68 1447.97 2138.24 2354.45 2656.88 2150.08 3234.49 2241.99 2256.40 21
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
CIDER47.21 2238.36 2253.11 2755.37 2755.10 2733.25 2543.46 2048.87 31
Qingshan Xu and Wenbing Tao: Learning Inverse Depth Regression for Multi-View Stereo with Correlation Cost Volume. AAAI 2020
PVSNet64.34 1645.78 2536.40 2552.03 2852.49 2949.20 3534.18 2438.61 2654.41 23
Pnet_fast37.64 3617.70 4250.93 2949.72 3160.73 1710.59 5324.81 4142.34 36
A-TVSNet + Gipumacopyleft42.12 3229.09 2950.80 3048.28 3656.53 2329.14 2729.04 3547.59 32
ANet-0.7539.47 3422.61 3850.71 3149.43 3255.75 2518.21 3927.01 3946.96 33
BP-MVSNet43.22 3032.08 2750.65 3252.86 2846.16 4127.25 2836.92 2752.94 25
Christian Sormann, Patrick Knöbelreiter, Andreas Kuhn, Mattia Rossi, Thomas Pock, Friedrich Fraundorfer: BP-MVSNet: Belief-Propagation-Layers for Multi-View-Stereo. 3DV 2020
AttMVS45.85 2439.26 2050.25 3349.36 3451.75 3034.83 2143.70 1849.63 29
ANet39.53 3324.60 3549.48 3449.42 3355.75 2522.08 3227.13 3843.25 35
P-MVSNet44.46 2937.09 2449.37 3549.27 3549.30 3434.35 2339.83 2449.54 30
MVSNet38.33 3526.99 3145.89 3640.49 4157.57 2220.52 3433.47 2939.63 39
test_112431.65 3913.28 5243.90 3744.21 3947.60 3614.80 4111.75 5739.88 38
Pnet-blend31.51 4015.02 4742.51 3847.41 3746.50 396.99 5723.05 4233.62 41
Pnet-blend++31.51 4015.02 4742.51 3847.41 3746.50 396.99 5723.05 4233.62 41
R-MVSNet36.87 3729.78 2841.61 4042.00 4046.99 3724.73 2934.83 2835.83 40
MVSNet_plusplus28.64 4212.93 5339.11 4149.80 3021.55 506.17 5919.69 4745.98 34
CasMVSNet(SR_B)32.16 3826.80 3235.73 4234.57 4246.99 3720.55 3333.05 3025.63 47
MVSCRF28.32 4318.18 4135.09 4332.16 4544.37 4314.66 4221.69 4528.73 44
MVSNet_++25.72 4417.38 4431.27 4433.47 4318.28 552.87 6331.90 3342.06 37
Snet23.85 4517.41 4328.15 4532.72 4424.90 4812.22 4922.60 4426.83 45
PMVScopyleft37.38 2021.09 4811.49 5827.48 4624.09 4644.44 4215.98 407.01 5913.92 55
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
A1Net23.74 4623.95 3623.61 4721.43 5119.62 5219.93 3527.96 3629.78 43
CPR_FA23.28 4723.62 3723.05 4819.33 5223.00 4919.34 3727.89 3726.81 46
example15.69 5610.14 5919.39 4923.73 4730.32 4511.10 529.17 584.13 63
F/T MVSNet+Gipuma17.31 4914.39 4919.26 5016.83 5419.67 5114.03 4414.75 5121.27 48
hgnet16.45 5112.31 5519.21 5121.51 4930.31 4612.80 4611.82 555.81 59
DPSNet16.45 5112.31 5519.21 5121.51 4930.31 4612.80 4611.82 555.81 59
MVSNet + Gipuma16.91 5014.12 5118.77 5316.39 5619.13 5413.98 4514.27 5220.78 49
firsttry15.88 5512.57 5418.08 5417.80 5317.14 5611.54 5113.61 5319.31 50
MVEpermissive26.22 2116.26 5316.97 4515.79 5511.75 5819.40 5314.45 4319.48 4816.21 51
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
Pnet-eth16.08 5418.30 4014.59 5621.78 487.07 629.66 5426.95 4014.91 53
unMVSv113.05 5712.03 5713.73 5714.18 5716.18 5712.13 5011.93 5410.84 56
CMPMVSbinary51.72 197.38 620.03 6512.27 582.37 6534.46 440.06 650.00 650.00 65
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
test_1120copyleft9.53 606.82 6011.33 598.74 609.06 607.57 566.06 6016.20 52
Cas-MVS_preliminary7.24 633.51 629.72 607.03 627.46 614.87 612.16 6314.65 54
unMVSmet7.42 615.82 618.48 617.60 6111.29 585.84 605.81 616.56 58
RMVSNet11.09 5815.49 468.17 629.16 5910.05 5912.57 4818.40 505.29 62
metmvs_fine10.03 5914.31 507.18 636.75 636.98 639.52 5519.10 497.81 57
confMetMVS4.86 643.39 635.85 645.59 646.61 642.94 623.83 625.35 61
FADENet0.17 650.15 640.18 650.31 660.16 650.19 640.10 640.08 64
dnet0.00 660.00 660.00 660.00 670.00 660.00 660.00 650.00 65
UnsupFinetunedMVSNet16.83 54