This table lists the benchmark results for the low-res many-view scenario.
The following metrics are evaluated:
Accuracy [%]: The fraction of the reconstruction which is closer to the ground truth than the evaluation threshold distance (*). Larger is better.Completeness [%]: The fraction of the ground truth which is closer to the reconstruction than the evaluation threshold distance (*). Larger is better.F1 score [%]: The harmonic mean of accuracy and completeness, used to rank methods based on both metrics. Larger is better.Time [s]: The runtime of the method. Smaller is better.
(*) 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 .
Set : Training
Metric : F1 score [%]
Tolerance : 2cm
Method
Info all low-res indoor outdoor delivery area electro forest playground terrains
DeepPCF-MVS 79.04 1 62.76 1 66.22 1 60.45 2 53.69 1 65.50 3 68.81 4 47.05 3 78.76 1 DeepC-MVS_fast 78.24 3 62.57 2 64.27 3 61.43 1 53.65 2 65.83 1 70.22 2 48.24 1 74.89 3 DeepC-MVS 78.47 2 61.99 3 64.34 2 60.43 3 53.21 3 65.62 2 68.07 5 47.59 2 75.48 2 3Dnovator+ 75.73 4 58.40 4 59.88 5 57.41 4 46.92 7 62.89 4 70.54 1 38.80 13 72.84 6 3Dnovator 73.76 5 58.20 5 60.10 4 56.93 5 45.58 9 62.04 5 70.17 3 38.57 14 74.61 5 PCF-MVS 73.28 6 57.32 6 59.66 6 55.76 6 48.61 4 56.36 12 67.24 7 43.68 5 70.70 11 Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction . GCPR 2019 ACMP 73.23 7 56.18 7 59.37 7 54.05 9 47.31 5 61.10 7 58.60 19 42.44 7 71.44 9 Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo . AAAI 2020 TAPA-MVS(SR) 56.09 8 56.78 13 55.62 7 43.28 13 58.11 9 65.17 10 43.60 6 70.29 12 OpenMVS 70.44 10 55.58 9 58.77 8 53.44 11 44.82 10 58.83 8 65.59 9 35.91 17 72.72 7 TAPA-MVS 71.42 9 55.13 10 58.21 10 53.07 12 41.62 16 56.79 11 62.68 14 39.75 12 74.79 4 Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo . ICCV 2019 ACMM 72.26 8 55.12 11 54.88 18 55.28 8 38.65 19 61.75 6 60.21 18 43.87 4 71.11 10 Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo . CVPR 2019 COLMAP(SR) 54.99 12 57.09 11 53.58 10 46.78 8 55.24 14 63.74 12 41.76 9 67.41 19 ACMH+ 66.54 12 53.99 13 56.90 12 52.05 13 44.66 11 57.37 10 56.79 22 41.98 8 69.14 14 COLMAP(base) 53.42 14 56.15 14 51.59 14 44.02 12 53.93 15 60.80 17 40.05 11 68.28 17 LTVRE_ROB 59.44 15 53.25 15 56.08 15 51.36 15 39.82 18 53.93 15 62.24 15 37.92 16 72.34 8 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 PLC 68.99 11 51.58 16 55.07 17 49.25 19 41.73 15 52.26 20 56.95 21 38.53 15 68.41 16 Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency . Pacific Graphics 2019 ACMH 65.37 13 51.50 17 53.46 20 50.20 17 38.46 20 52.73 18 57.24 20 40.63 10 68.46 15 Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo . CVPR 2019 GSE 51.31 18 58.55 9 46.48 24 47.08 6 49.58 23 54.32 26 35.53 19 70.02 13 BP-MVSNet 50.87 19 53.66 19 49.01 20 40.10 17 49.76 22 63.03 13 34.23 20 67.21 20 LPCS 50.41 20 55.27 16 47.16 23 43.12 14 51.06 21 54.57 25 35.86 18 67.43 18 COLMAP_ROB 62.73 14 49.91 21 51.76 21 48.68 21 37.30 21 52.31 19 61.83 16 31.91 23 66.23 21 Johannes L. Schönberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo . ECCV 2016 CasMVSNet(SR_B) 49.66 22 48.09 22 50.71 16 33.02 23 55.30 13 67.43 6 29.40 24 63.15 23 CasMVSNet(base) 49.00 23 48.08 24 49.61 18 34.83 22 53.10 17 66.82 8 28.93 26 61.32 25 A-TVSNet + Gipuma 45.36 24 47.40 25 44.00 25 29.86 25 43.85 25 55.31 23 32.84 22 64.94 22 CIDER 45.22 25 41.50 26 47.70 22 28.55 26 46.16 24 64.03 11 32.90 21 54.46 27 Qingshan Xu and Wenbing Tao: Learning Inverse Depth Regression for Multi-View Stereo with Correlation Cost Volume . AAAI 2020 CasMVSNet(SR_A) 41.59 26 48.09 22 37.27 26 33.02 23 27.16 26 55.24 24 29.40 24 63.15 23 PMVS 39.38 17 22.01 27 32.82 27 14.80 27 5.84 30 11.38 28 26.05 27 6.98 29 59.80 26 Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis . PAMI (2010) example 18.45 28 28.74 28 11.60 28 6.01 29 5.21 31 22.21 28 7.37 28 51.47 28 MVE 19.12 18 15.60 29 22.59 32 10.94 29 5.22 31 14.12 27 8.87 31 9.84 27 39.95 32 Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment . EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014) hgnet 15.22 30 24.38 29 9.11 30 6.26 27 5.91 29 17.93 29 3.49 30 42.51 30 DPSNet 15.22 30 24.38 29 9.11 30 6.26 27 5.91 29 17.93 29 3.49 30 42.51 30 CMPMVS 47.78 16 9.53 32 23.73 31 0.07 32 0.00 32 0.00 32 0.20 32 0.00 32 47.47 29 M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces . CVPR 2011 dnet 0.00 33 0.00 33 0.00 33 0.00 32 0.00 32 0.00 33 0.00 32 0.00 33
