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-MVS93.97 187.63 182.10 191.31 191.80 191.34 177.64 186.56 190.79 1
DeepC-MVS_fast93.61 286.42 279.88 390.79 291.32 290.73 275.68 384.08 490.32 3
DeepC-MVS93.30 386.02 379.73 490.22 389.69 690.52 374.52 584.94 290.45 2
TAPA-MVS90.10 784.80 480.13 287.91 688.22 888.54 975.35 484.91 386.96 14
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
PCF-MVS89.48 1084.51 576.96 1089.54 491.30 388.76 773.35 880.56 1288.57 4
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
TAPA-MVS(SR)83.88 678.08 787.74 888.99 786.49 1673.17 1082.99 687.76 10
COLMAPsoft83.59 777.50 987.65 986.74 1287.68 1173.30 981.70 988.52 5
PLCcopyleft91.00 683.54 878.38 586.98 1385.38 1687.33 1373.45 683.31 588.23 9
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
GSE83.39 977.82 887.10 1285.70 1587.27 1473.41 782.23 788.34 8
COLMAP(SR)83.28 1076.24 1287.97 587.61 987.90 1072.76 1179.71 1488.39 7
ACMH+87.92 1382.46 1176.63 1186.35 1686.95 1184.52 1871.68 1281.58 1087.56 12
ACMP89.59 982.31 1275.00 1487.18 1184.49 1888.57 868.43 1881.58 1088.49 6
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
LTVRE_ROB88.41 1282.23 1378.11 684.97 1986.24 1382.83 2376.93 279.30 1585.84 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
ACMM89.79 882.20 1475.34 1386.76 1484.94 1787.68 1168.85 1781.84 887.67 11
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
3Dnovator91.36 582.19 1574.09 1687.59 1089.83 590.10 571.54 1376.63 1882.84 19
3Dnovator+91.43 481.91 1673.11 1987.78 790.06 490.17 470.40 1575.82 2083.12 17
COLMAP_ROBcopyleft87.81 1481.57 1774.74 1586.13 1783.81 1987.08 1569.46 1680.01 1387.49 13
Johannes L. Schönberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
ACMH87.59 1580.55 1873.69 1785.11 1886.17 1482.81 2468.19 1979.20 1686.36 15
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
OpenMVScopyleft89.19 1180.38 1971.40 2086.37 1587.42 1089.29 667.70 2075.10 2182.40 21
LPCS78.54 2073.62 1881.82 2179.59 2483.19 2171.25 1475.99 1982.69 20
A-TVSNet + Gipumacopyleft77.14 2167.96 2283.26 2080.53 2286.26 1765.25 2170.67 2282.99 18
PVSNet_082.17 1674.27 2264.50 2380.78 2281.77 2180.64 2761.11 2367.89 2479.92 22
OpenMVS_ROBcopyleft81.14 1773.28 2363.11 2580.06 2382.70 2081.62 2661.11 2365.12 2875.87 23
CIDER72.92 2462.80 2679.66 2480.16 2383.22 2060.69 2564.90 2975.61 25
Qingshan Xu and Wenbing Tao: Learning Inverse Depth Regression for Multi-View Stereo with Correlation Cost Volume. AAAI 2020
R-MVSNet70.54 2563.98 2474.91 2873.85 3077.87 3159.85 2668.11 2373.01 26
P-MVSNet70.42 2671.20 2169.90 3266.73 3572.02 3364.84 2277.55 1770.95 28
ANet-0.7567.51 2758.66 2973.41 3070.12 3179.37 2850.72 2966.61 2670.74 29
AttMVS65.89 2860.14 2769.73 3364.68 3875.04 3258.80 2761.47 3269.47 31
ANet63.85 2952.65 3171.32 3170.12 3179.37 2848.53 3156.77 3564.48 36
Pnet-new-63.42 3042.13 4177.61 2577.71 2679.30 3041.06 3443.20 4775.82 24
cscdsrtA63.08 3145.33 3874.92 2774.36 2783.73 1937.04 4153.62 3866.66 34
cscd61.88 3243.72 3973.99 2974.03 2981.95 2534.99 4452.45 3965.99 35
Pnet_fast59.94 3335.26 5376.39 2674.15 2882.99 2225.02 5145.51 4472.03 27
CPR_FA58.26 3456.35 3059.54 4056.43 4455.34 4350.61 3062.09 3166.83 33
MVSNet56.22 3546.49 3562.70 3757.79 4167.16 3838.93 3754.05 3663.16 38
A1Net54.85 3659.98 2851.43 4347.46 5342.97 5252.72 2867.24 2563.86 37
MVSNet_plusplus54.10 3736.12 5166.09 3478.36 2549.55 4720.40 5651.84 4070.36 30
Pnet-blend53.68 3837.04 4864.76 3568.91 3370.02 3424.13 5249.96 4255.36 41
Pnet-blend++53.68 3837.04 4864.76 3568.91 3370.02 3424.13 5249.96 4255.36 41
MVSCRF51.99 4037.64 4561.55 3858.88 3968.35 3634.99 4440.28 5057.43 40
MVSNet_++50.85 4137.40 4659.81 3965.34 3746.28 489.19 5765.60 2767.81 32
Snet49.32 4236.02 5258.18 4166.02 3650.24 4527.93 5044.12 4558.27 39
cscdsrtB48.99 4345.75 3651.15 4449.07 5162.78 3937.57 3853.93 3741.61 48
Pnet-eth45.62 4449.15 3343.27 5353.31 4629.96 5835.03 4363.26 3046.53 45
unMVSv145.02 4543.12 4046.29 5047.83 5250.14 4642.67 3243.57 4640.92 49
RMVSNet44.67 4650.67 3240.67 5450.79 4741.21 5442.32 3359.02 3330.02 52
MVEpermissive50.73 2044.67 4645.57 3744.07 5235.90 5652.30 4440.89 3550.26 4144.00 47
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
hgnet44.61 4838.48 4348.69 4557.33 4260.08 4037.17 3939.80 5228.65 54
DPSNet44.61 4838.48 4348.69 4557.33 4260.08 4037.17 3939.80 5228.65 54
MVSNet + Gipuma44.11 5038.76 4247.67 4850.05 5045.26 5034.93 4642.60 4847.71 43
F/T MVSNet+Gipuma43.59 5137.31 4747.78 4750.12 4845.64 4934.25 4740.37 4947.60 44
PMVScopyleft53.92 1942.74 5226.50 5553.57 4254.03 4568.05 3736.35 4216.65 5738.64 50
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
firsttry41.11 5336.23 5044.36 5143.82 5443.77 5132.28 4840.18 5145.51 46
example40.49 5430.54 5447.11 4958.75 4058.88 4229.97 4931.12 5423.71 57
metmvs_fine39.41 5548.46 3433.38 5635.53 5731.67 5739.28 3657.64 3432.95 51
unMVSmet32.09 5625.89 5636.22 5536.96 5542.02 5323.79 5427.99 5529.68 53
confMetMVS26.62 5723.72 5728.55 5728.79 5833.00 5621.33 5526.10 5623.86 56
CMPMVSbinary62.92 188.44 580.22 5913.92 584.26 5937.50 550.45 590.00 590.00 59
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
FADENet1.48 591.79 581.27 592.11 601.06 592.23 581.34 580.65 58
dnet0.00 600.00 600.00 600.00 610.00 600.00 600.00 590.00 59
UnsupFinetunedMVSNet50.12 48