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-MVS98.18 396.68 195.10 197.74 198.30 197.78 492.41 297.79 197.12 1
DeepC-MVS_fast98.59 196.44 294.71 397.59 298.20 297.75 592.08 397.35 596.83 2
DeepC-MVS98.44 296.08 394.14 1097.37 397.83 597.46 1090.75 1397.52 396.81 3
COLMAP(SR)96.07 494.39 797.19 596.86 897.92 291.54 1097.24 796.79 4
TAPA-MVS(SR)95.82 594.80 296.49 996.19 1197.58 992.04 497.56 295.71 12
PCF-MVS97.08 1395.68 693.36 1497.23 498.16 396.93 1491.43 1195.29 1896.61 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 1297.20 1392.04 496.94 1096.10 9
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
ACMH+97.24 995.61 893.77 1396.83 696.69 997.71 690.70 1496.85 1196.09 10
TAPA-MVS97.07 1495.48 994.55 496.10 1295.68 1597.30 1191.73 897.37 495.32 14
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
COLMAPsoft95.42 1094.03 1296.35 1195.96 1396.92 1591.10 1296.97 896.16 8
PLCcopyleft97.94 495.35 1194.28 996.06 1395.26 1896.59 1691.61 996.96 996.34 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 1596.65 796.32 1097.93 189.82 1696.68 1395.69 13
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
ACMP97.20 1095.11 1393.01 1796.52 895.68 1597.26 1289.74 1796.27 1596.61 5
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
LTVRE_ROB97.16 1195.10 1494.13 1195.75 1495.69 1495.73 2092.87 195.39 1795.83 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 1594.33 895.45 1694.81 1996.53 1891.98 696.68 1395.02 16
A-TVSNet + Gipumacopyleft94.96 1694.54 595.24 1895.31 1796.58 1791.82 797.27 693.83 17
COLMAP_ROBcopyleft97.56 694.43 1793.09 1695.33 1794.55 2096.37 1989.45 1896.72 1295.07 15
Johannes L. Schönberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
3Dnovator97.25 892.47 1887.97 2495.47 1597.93 497.85 386.02 2289.92 2490.64 21
3Dnovator+97.12 1292.17 1987.57 2595.24 1897.49 697.68 785.24 2389.91 2590.54 22
LPCS92.07 2092.93 1891.49 2490.32 2693.29 2690.66 1595.20 1990.87 20
OpenMVScopyleft96.50 1591.77 2186.65 2795.19 2097.38 797.66 884.60 2588.69 2990.53 23
R-MVSNet90.93 2289.80 2091.68 2390.81 2594.15 2587.67 1991.93 2390.07 24
PVSNet94.43 1690.46 2385.77 2893.59 2193.82 2295.33 2284.62 2486.92 3191.61 19
CPR_FA89.35 2488.37 2390.00 2789.38 2887.69 3487.29 2189.46 2792.93 18
CIDER88.55 2583.13 3392.16 2293.95 2195.61 2183.79 2882.47 3486.92 26
Qingshan Xu and Wenbing Tao: Learning Inverse Depth Regression for Multi-View Stereo with Correlation Cost Volume. AAAI 2020
ANet-0.7587.89 2689.58 2186.76 2884.37 4090.29 3084.51 2694.65 2085.61 29
OpenMVS_ROBcopyleft92.34 1787.46 2781.75 3491.27 2592.55 2395.21 2381.21 3282.29 3586.05 27
P-MVSNet85.35 2891.57 1981.21 3778.76 4782.87 4187.56 2095.58 1682.00 33
ANet85.16 2985.20 3085.14 3184.37 4090.29 3082.00 3088.39 3080.78 37
unMVSv183.89 3084.37 3283.58 3384.00 4485.63 3683.17 2985.56 3281.11 36
metmvs_fine83.50 3184.90 3182.56 3584.72 3581.80 4481.09 3388.72 2881.17 35
A1Net82.90 3289.39 2278.57 4685.19 3263.18 5784.36 2794.42 2187.34 25
RMVSNet80.62 3386.71 2676.57 4784.14 4278.09 4680.22 3493.20 2267.48 53
Pnet_fast80.38 3465.52 5190.29 2690.03 2794.91 2460.54 5070.51 5185.93 28
AttMVS80.32 3578.27 3581.68 3678.98 4686.16 3579.76 3576.79 4879.89 38
Pnet-new-79.46 3669.54 4786.08 2984.76 3490.31 2965.13 4473.95 5083.16 31
cscdsrtA79.00 3769.56 4685.29 3085.95 3092.82 2762.08 4777.05 4477.11 39
hgnet78.82 3876.20 3780.57 4184.50 3682.32 4275.56 3776.84 4574.88 42
DPSNet78.82 3876.20 3780.57 4184.50 3682.32 4275.56 3776.84 4574.88 42
Pnet-blend78.13 4073.62 4181.13 3884.46 3888.20 3266.50 4180.74 3770.74 51
Pnet-blend++78.13 4073.62 4181.13 3884.46 3888.20 3266.50 4180.74 3770.74 51
cscd78.10 4268.51 4884.50 3285.47 3191.56 2859.79 5277.22 4376.47 40
Pnet-eth77.75 4385.52 2972.57 5275.57 4965.89 5681.44 3189.61 2676.26 41
MVSCRF77.21 4474.57 4078.98 4579.29 4583.87 3975.01 3974.12 4973.78 46
MVEpermissive76.82 1876.63 4577.52 3676.03 4970.40 5585.49 3775.68 3679.36 3972.20 48
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
MVSNet_plusplus76.50 4666.38 4983.25 3491.25 2473.79 5153.62 5679.14 4084.70 30
example74.48 4765.59 5080.41 4384.10 4383.16 4062.82 4668.36 5373.98 45
Snet74.39 4864.28 5381.13 3888.04 2973.18 5460.15 5168.41 5282.18 32
MVSNet73.65 4970.01 4576.07 4876.39 4877.59 4761.93 4878.09 4274.23 44
MVSNet + Gipuma72.61 5071.20 4373.54 5075.39 5073.46 5365.60 4376.81 4771.78 49
firsttry72.49 5175.56 3970.44 5468.80 5770.25 5569.31 4081.82 3672.28 47
MVSNet_++71.81 5259.04 5680.33 4484.86 3374.23 5033.06 5785.02 3381.88 34
cscdsrtB70.07 5371.18 4469.33 5569.73 5677.06 4864.06 4578.31 4161.20 55
F/T MVSNet+Gipuma69.80 5464.46 5273.36 5175.05 5173.57 5260.74 4968.18 5471.46 50
unMVSmet66.94 5560.67 5571.11 5371.91 5479.70 4554.62 5466.73 5561.74 54
confMetMVS63.51 5660.95 5465.21 5760.95 5876.48 4955.53 5366.38 5658.21 56
PMVScopyleft70.75 1958.19 5743.22 5768.18 5673.06 5384.09 3854.23 5532.21 5747.37 57
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
FADENet18.58 5815.51 5820.63 5821.45 5932.74 5915.09 5815.93 587.68 58
CMPMVSbinary69.68 2011.01 590.97 5917.70 5910.09 6043.01 581.93 590.00 590.00 59
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
dnet0.00 600.00 600.00 600.00 610.00 600.00 600.00 590.00 59
UnsupFinetunedMVSNet75.05 51