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-MVS96.37 292.61 189.11 194.94 195.58 195.07 185.62 192.60 194.17 1
DeepC-MVS_fast96.44 191.83 287.74 394.57 295.34 394.64 284.48 491.00 393.72 3
DeepC-MVS96.23 391.57 387.58 494.23 394.49 594.42 383.58 591.59 293.78 2
TAPA-MVS93.98 790.68 488.05 292.44 592.42 893.52 785.24 290.85 591.36 10
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
PCF-MVS93.45 1090.50 585.08 994.11 495.54 293.44 982.27 787.88 1193.34 4
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
PLCcopyleft95.07 489.96 687.00 591.94 990.65 1392.68 1283.01 690.98 492.48 6
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
ACMH+92.99 1389.84 785.96 792.43 692.28 992.84 1181.89 890.03 692.17 8
ACMM93.85 889.42 885.42 892.10 891.09 1292.98 1081.16 989.68 792.23 7
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
ACMP93.49 989.04 984.30 1292.21 790.34 1493.47 879.54 1389.07 1092.81 5
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
LTVRE_ROB92.95 1489.00 1086.32 690.78 1591.67 1089.54 1884.87 387.77 1291.13 12
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
ACMH92.88 1588.89 1184.55 1091.78 1091.37 1192.41 1379.88 1189.21 891.56 9
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
COLMAP_ROBcopyleft93.27 1188.32 1284.41 1190.93 1489.38 1592.11 1479.72 1289.10 991.28 11
Johannes L. Schönberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
3Dnovator94.51 587.43 1380.90 1591.78 1094.76 494.22 478.96 1482.85 1586.35 14
3Dnovator+94.38 687.07 1480.26 1691.61 1294.34 694.17 578.01 1582.52 1686.34 15
A-TVSNet + Gipuma86.96 1582.91 1389.65 1688.10 1691.71 1580.23 1085.60 1489.15 13
OpenMVScopyleft93.04 1286.34 1679.08 1791.18 1393.50 793.77 676.56 1681.60 1986.26 16
R-MVSNet81.26 1777.44 1883.81 1982.82 1986.25 1974.52 1880.36 2082.37 17
CIDER80.48 1872.14 2186.05 1787.66 1789.77 1671.75 1972.53 2380.72 18
Qingshan Xu and Wenbing Tao: Learning Inverse Depth Regression for Multi-View Stereo with Correlation Cost Volume. AAAI 2020
OpenMVS_ROBcopyleft86.42 1680.25 1971.88 2285.82 1887.52 1889.56 1770.53 2073.24 2280.39 19
ANet-0.7577.73 2074.78 2079.70 2077.01 2284.71 2067.34 2382.21 1877.39 20
P-MVSNet77.69 2181.61 1475.07 2371.91 2477.33 2376.51 1786.71 1375.98 22
ANet74.07 2268.26 2477.94 2177.01 2284.71 2064.75 2471.76 2472.09 24
AttMVS72.35 2368.13 2575.16 2270.98 2880.63 2268.09 2268.16 2573.87 23
A1Net68.49 2475.64 1963.72 2963.53 3351.39 3868.88 2182.41 1776.23 21
unMVSv163.91 2562.59 2664.80 2866.39 3168.00 3161.62 2563.55 2860.01 29
MVSNet63.58 2656.25 2868.47 2766.00 3271.12 2648.36 3264.13 2668.29 27
MVSCRF63.13 2752.16 3370.45 2470.18 2975.83 2551.17 3053.16 3665.33 28
RMVSNet62.69 2869.01 2358.48 3769.15 3058.56 3459.72 2678.30 2147.75 36
Pnet_fast60.77 2947.18 3569.82 2577.90 2061.73 3239.61 3754.76 3369.82 25
Snet60.77 2947.18 3569.82 2577.90 2061.73 3239.61 3754.76 3369.82 25
DPSNet59.89 3154.16 2963.70 3071.39 2570.87 2752.09 2856.24 3148.85 34
hgnet59.89 3154.16 2963.70 3071.39 2570.87 2752.09 2856.24 3148.85 34
MVEpermissive62.14 1859.41 3360.38 2758.77 3650.39 3969.00 3056.98 2763.78 2756.91 33
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
MVSNet + Gipuma57.07 3453.08 3159.73 3562.91 3657.42 3648.41 3157.74 2958.88 30
F/T MVSNet+Gipuma55.78 3549.83 3459.76 3463.10 3457.51 3546.34 3453.31 3558.66 31
example55.06 3644.75 3761.93 3271.21 2769.85 2942.82 3646.68 3744.73 37
firsttry54.62 3752.74 3255.88 3854.90 3855.69 3748.02 3357.46 3057.06 32
PMVScopyleft61.03 1949.24 3832.69 3860.27 3362.70 3776.26 2442.99 3522.38 3841.86 38
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
CMPMVSbinary66.06 179.32 390.44 3915.23 396.17 4039.51 390.89 390.00 390.00 39
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
dnet0.00 400.00 400.00 400.00 410.00 400.00 400.00 390.00 39
UnsupFinetunedMVSNet63.10 34