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-MVS69.38 247.34 138.25 153.40 254.29 456.62 332.95 143.55 249.30 3
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
DeepPCF-MVS69.58 146.90 237.52 353.15 353.58 557.05 232.83 242.22 348.83 5
DeepC-MVS_fast68.24 346.61 334.94 454.40 155.37 357.55 131.02 438.86 450.28 1
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
3Dnovator64.47 545.24 434.37 552.48 557.34 153.53 530.63 538.11 646.56 8
3Dnovator+66.72 444.84 533.14 752.63 456.23 253.78 429.20 637.07 947.89 6
ACMP63.53 642.50 631.11 950.08 648.62 1052.78 724.90 1537.33 848.84 4
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
OpenMVScopyleft61.03 942.28 732.86 848.56 951.94 750.82 1027.02 938.70 542.92 12
TAPA-MVS59.36 1041.97 837.54 244.93 1346.78 1446.52 1631.44 343.65 141.48 15
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
COLMAP(SR)41.20 929.58 1048.95 748.70 948.18 1228.26 830.89 1649.98 2
ACMM61.98 740.80 1029.05 1248.64 846.87 1253.22 623.47 1834.63 1045.82 9
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
TAPA-MVS(SR)40.72 1129.34 1148.31 1052.32 644.77 2026.63 1132.05 1247.84 7
PCF-MVS61.88 840.12 1233.16 644.75 1449.39 846.16 1728.63 737.70 738.70 20
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
COLMAP(base)37.35 1328.09 1743.52 1642.34 2046.68 1526.55 1229.63 1941.53 14
LTVRE_ROB55.42 1436.88 1428.57 1542.42 1840.72 2448.22 1126.97 1030.17 1838.32 21
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
ACMH+57.40 1136.48 1528.86 1341.57 2043.89 1937.61 3226.39 1331.33 1443.20 11
CasMVSNet(SR_A)35.87 1619.38 2946.87 1146.17 1651.96 914.32 3524.45 2642.47 13
OpenMVS_ROBcopyleft52.78 1635.39 1727.98 1840.33 2146.81 1336.10 3824.79 1631.17 1538.08 22
CasMVSNet(base)35.32 1818.44 3246.56 1246.00 1852.64 813.32 3623.57 2941.04 17
GSE34.79 1923.63 2542.23 1946.31 1539.18 2923.20 2024.07 2741.20 16
PLCcopyleft56.13 1234.63 2028.70 1438.59 2736.78 3244.65 2126.03 1431.36 1334.33 29
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
AttMVS34.35 2128.45 1638.28 2840.03 2540.06 2623.67 1733.23 1134.75 28
LPCS33.63 2224.80 2139.52 2342.29 2139.01 3023.44 1926.15 2337.27 24
PVSNet_043.31 1833.35 2323.83 2339.69 2241.22 2238.75 3121.08 2426.59 2139.10 19
test_112633.11 2418.21 3443.04 1747.43 1145.33 1815.33 3221.10 3436.35 26
COLMAP_ROBcopyleft52.97 1533.06 2525.38 2038.18 2935.81 3445.04 1922.92 2227.84 2033.69 30
Johannes L. Schönberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
ACMH55.70 1332.91 2623.69 2439.06 2541.22 2236.76 3621.84 2325.54 2439.18 18
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
Pnet-new-32.65 2716.09 3743.68 1546.15 1739.98 2718.51 2713.67 4744.91 10
CIDER32.55 2825.75 1937.08 3239.65 2640.58 2520.91 2530.59 1731.00 34
Qingshan Xu and Wenbing Tao: Learning Inverse Depth Regression for Multi-View Stereo with Correlation Cost Volume. AAAI 2020
PVSNet50.76 1732.44 2924.74 2237.57 3138.43 2936.66 3723.03 2126.46 2237.62 23
ANet30.02 3016.73 3638.88 2639.32 2744.06 2215.61 3117.85 3733.26 31
BP-MVSNet29.92 3120.53 2736.17 3337.89 3033.68 4217.55 2823.52 3036.95 25
Christian Sormann, Patrick Knöbelreiter, Andreas Kuhn, Mattia Rossi, Thomas Pock, Friedrich Fraundorfer: BP-MVSNet: Belief-Propagation-Layers for Multi-View-Stereo. 3DV 2020
P-MVSNet29.35 3220.98 2634.93 3434.55 3836.99 3420.06 2621.91 3333.25 32
ANet-0.7529.06 3313.44 4039.47 2439.32 2744.06 2210.78 4116.11 4135.02 27
test_120528.37 3419.29 3034.43 3534.50 3941.99 2415.81 3022.78 3126.79 39
Pnet_fast27.80 3512.25 4138.17 3037.73 3147.91 137.47 4717.03 3828.88 36
MVSNet27.45 3618.54 3133.39 3729.32 4147.13 1413.18 3723.89 2823.72 40
A-TVSNet + Gipumacopyleft26.68 3716.93 3533.18 3831.29 4039.82 2817.28 2916.57 4028.43 37
test_112423.88 388.92 4933.86 3635.80 3536.79 3510.99 406.84 5528.99 35
R-MVSNet23.84 3918.30 3327.54 4127.82 4233.79 4114.57 3422.03 3221.00 43
CasMVSNet(SR_B)23.80 4019.79 2826.47 4324.88 4337.51 3314.94 3324.65 2517.01 45
Pnet-blend++21.89 419.21 4730.34 3934.58 3634.44 393.76 5814.67 4422.00 41
Pnet-blend21.89 419.21 4730.34 3934.58 3634.44 393.76 5814.67 4422.00 41
MVSNet_plusplus19.10 436.64 5527.41 4236.63 3314.19 513.12 6010.15 4931.41 33
MVSCRF18.49 4412.01 4222.81 4420.51 4632.84 439.61 4214.41 4615.08 48
MVSNet_++16.58 4510.34 4520.75 4522.25 4411.60 541.64 6319.05 3528.39 38
Snet16.01 4611.97 4318.70 4622.09 4517.12 498.46 4515.48 4316.90 46
A1Net14.51 4714.40 3914.59 4713.30 4812.51 5211.95 3816.85 3917.96 44
CPR_FA14.43 4814.90 3814.12 4911.38 4914.82 5011.91 3917.89 3616.15 47
PMVScopyleft28.69 2010.60 494.70 6014.53 4810.82 5330.05 456.66 522.74 612.71 59
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
F/T MVSNet+Gipuma9.78 508.13 5010.88 559.23 5511.92 538.81 437.45 5311.49 49
firsttry9.57 517.40 5311.02 5411.05 5010.54 576.97 507.82 5211.49 49
example9.54 525.97 5911.92 5013.40 4720.72 466.89 515.05 591.63 64
MVSNet + Gipuma9.47 537.91 5110.52 568.91 5711.46 558.80 447.01 5411.18 52
DPSNet9.31 546.52 5611.17 5210.91 5120.13 477.15 485.89 562.47 61
hgnet9.31 546.52 5611.17 5210.91 5120.13 477.15 485.89 562.47 61
MVEpermissive17.77 219.03 569.66 468.62 576.03 6011.40 568.30 4611.02 488.42 53
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
Pnet-eth8.45 5710.49 447.10 609.98 542.92 655.14 5515.84 428.40 54
unMVSv16.85 586.06 587.37 597.56 589.33 586.35 535.76 585.22 56
CMPMVSbinary42.80 196.77 590.01 6611.27 511.93 6631.86 440.03 660.00 660.00 66
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
test_1120copyleft6.40 604.10 617.94 586.69 595.86 604.95 563.26 6011.26 51
metmvs_fine4.94 617.02 543.56 633.18 643.86 624.62 579.43 513.63 57
RMVSNet4.93 627.73 523.06 643.93 623.22 645.95 549.52 502.04 63
Cas-MVS_preliminary4.12 631.79 635.67 614.49 614.19 612.85 610.74 648.32 55
unMVSmet3.60 642.44 624.38 623.82 636.35 592.69 622.18 622.97 58
confMetMVS2.34 651.33 643.02 653.03 653.47 631.14 641.52 632.56 60
FADENet0.06 660.04 650.07 660.12 670.08 660.05 650.03 650.02 65
dnet0.00 670.00 670.00 670.00 680.00 670.00 670.00 660.00 66
UnsupFinetunedMVSNet9.23 55