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-MVS69.58 146.90 137.52 253.15 153.58 357.05 232.83 142.22 248.83 4
DeepC-MVS_fast68.22 346.18 235.80 453.10 253.42 457.06 131.61 240.00 448.84 2
DeepC-MVS68.55 245.84 336.51 352.07 551.95 656.64 331.46 341.56 347.61 7
3Dnovator64.47 545.24 434.37 552.48 457.34 153.53 530.63 538.11 646.56 8
3Dnovator+66.72 444.84 533.14 752.63 356.23 253.78 429.20 637.07 947.89 5
ACMP63.53 642.50 631.11 950.08 648.62 1052.78 724.90 1537.33 848.84 2
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 1041.97 837.54 144.93 1346.78 1346.52 1631.44 443.65 141.48 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 1
ACMM61.98 740.80 1029.05 1248.64 846.87 1153.22 623.47 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 1052.32 544.77 1926.63 1132.05 1247.84 6
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
COLMAPsoft37.35 1328.09 1743.52 1642.34 1946.68 1526.55 1229.63 1941.53 14
LTVRE_ROB55.42 1436.88 1428.57 1542.42 1740.72 2348.22 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 1943.89 1837.61 3026.39 1331.33 1443.20 11
cscdsrtA35.87 1619.38 2746.87 1146.17 1551.96 914.32 3124.45 2542.47 13
OpenMVS_ROBcopyleft52.78 1635.39 1727.98 1840.33 2046.81 1236.10 3424.79 1631.17 1538.08 22
cscd35.32 1818.44 2946.56 1246.00 1752.64 813.32 3223.57 2841.04 17
GSE34.79 1923.63 2442.23 1846.31 1439.18 2723.20 2024.07 2641.20 16
PLCcopyleft56.13 1234.63 2028.70 1438.59 2636.78 2944.65 2026.03 1431.36 1334.33 26
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
AttMVS34.35 2128.45 1638.28 2740.03 2440.06 2423.67 1733.23 1134.75 25
LPCS33.63 2224.80 2139.52 2242.29 2039.01 2823.44 1926.15 2237.27 23
PVSNet43.31 1733.35 2323.83 2239.69 2141.22 2138.75 2921.08 2326.59 2139.10 19
COLMAP_ROBcopyleft52.97 1533.06 2425.38 2038.18 2835.81 3145.04 1822.92 2127.84 2033.69 27
Johannes L. Schönberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
ACMH55.70 1332.91 2523.69 2339.06 2441.22 2136.76 3321.84 2225.54 2339.18 18
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
Pnet-new-32.65 2616.09 3343.68 1546.15 1639.98 2518.51 2613.67 4344.91 10
CIDER32.55 2725.75 1937.08 3039.65 2540.58 2320.91 2430.59 1731.00 31
Qingshan Xu and Wenbing Tao: Learning Inverse Depth Regression for Multi-View Stereo with Correlation Cost Volume. AAAI 2020
ANet30.02 2816.73 3238.88 2539.32 2644.06 2115.61 2817.85 3333.26 28
P-MVSNet29.35 2920.98 2534.93 3134.55 3436.99 3220.06 2521.91 3033.25 29
ANet-0.7529.06 3013.44 3639.47 2339.32 2644.06 2110.78 3616.11 3735.02 24
Pnet_fast27.80 3112.25 3738.17 2937.73 2847.91 137.47 4217.03 3428.88 32
MVSNet27.45 3218.54 2833.39 3229.32 3647.13 1413.18 3323.89 2723.72 35
A-TVSNet + Gipumacopyleft26.68 3316.93 3133.18 3331.29 3539.82 2617.28 2716.57 3628.43 33
R-MVSNet23.84 3418.30 3027.54 3627.82 3733.79 3714.57 3022.03 2921.00 38
cscdsrtB23.80 3519.79 2626.47 3824.88 3837.51 3114.94 2924.65 2417.01 40
Pnet-blend21.89 369.21 4330.34 3434.58 3234.44 353.76 5214.67 4022.00 36
Pnet-blend++21.89 369.21 4330.34 3434.58 3234.44 353.76 5214.67 4022.00 36
MVSNet_plusplus19.10 386.64 5027.41 3736.63 3014.19 463.12 5410.15 4531.41 30
MVSCRF18.49 3912.01 3822.81 3920.51 4132.84 389.61 3714.41 4215.08 43
MVSNet_++16.58 4010.34 4120.75 4022.25 3911.60 491.64 5619.05 3128.39 34
Snet16.01 4111.97 3918.70 4122.09 4017.12 448.46 4015.48 3916.90 41
A1Net14.51 4214.40 3514.59 4213.30 4312.51 4711.95 3416.85 3517.96 39
CPR_FA14.43 4314.90 3414.12 4411.38 4414.82 4511.91 3517.89 3216.15 42
PMVScopyleft28.69 1910.60 444.70 5514.53 4310.82 4830.05 406.66 472.74 552.71 52
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
F/T MVSNet+Gipuma9.78 458.13 4510.88 509.23 5011.92 488.81 387.45 4911.49 44
firsttry9.57 467.40 4811.02 4911.05 4510.54 526.97 457.82 4811.49 44
example9.54 475.97 5411.92 4513.40 4220.72 416.89 465.05 541.63 57
MVSNet + Gipuma9.47 487.91 4610.52 518.91 5211.46 508.80 397.01 5011.18 46
hgnet9.31 496.52 5111.17 4710.91 4620.13 427.15 435.89 512.47 54
DPSNet9.31 496.52 5111.17 4710.91 4620.13 427.15 435.89 512.47 54
MVEpermissive17.77 209.03 519.66 428.62 526.03 5411.40 518.30 4111.02 448.42 47
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
Pnet-eth8.45 5210.49 407.10 549.98 492.92 585.14 5015.84 388.40 48
unMVSv16.85 536.06 537.37 537.56 539.33 536.35 485.76 535.22 49
CMPMVSbinary42.80 186.77 540.01 5911.27 461.93 5931.86 390.03 590.00 590.00 59
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
metmvs_fine4.94 557.02 493.56 563.18 573.86 554.62 519.43 473.63 50
RMVSNet4.93 567.73 473.06 573.93 553.22 575.95 499.52 462.04 56
unMVSmet3.60 572.44 564.38 553.82 566.35 542.69 552.18 562.97 51
confMetMVS2.34 581.33 573.02 583.03 583.47 561.14 571.52 572.56 53
FADENet0.06 590.04 580.07 590.12 600.08 590.05 580.03 580.02 58
dnet0.00 600.00 600.00 600.00 610.00 600.00 600.00 590.00 59
UnsupFinetunedMVSNet9.23 50