High-res multi-view results

This table lists the benchmark results for the high-res multi-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.

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	all	high-res multi-view	indoor	outdoor	courty.	delive.	electro	facade	kicker	meadow	office	pipes	playgr.	relief	relief.	terrace	terrai.

APDe-MVS		98.87 1	98.96 1	98.77 1	99.58 1	99.53 2	99.44 1	97.81 1	98.22 7	97.33 2	98.70 2	99.33 7	98.86 7	98.96 3	98.40 11	99.63 3	99.57 5

ESAPD		98.59 2	98.77 4	98.39 6	99.46 8	99.50 4	99.11 13	97.80 2	97.20 33	96.06 13	98.56 3	99.83 2	98.43 23	98.84 7	98.03 23	99.45 30	99.45 11

LS3D		95.46 50	95.14 63	95.84 46	97.91 49	98.90 57	98.58 28	97.79 3	97.07 38	83.65 110	88.71 95	88.64 87	97.82 31	97.49 47	97.42 42	99.26 76	97.72 145

APD-MVS		98.36 12	98.32 19	98.41 5	99.47 6	99.26 19	99.12 12	97.77 4	96.73 43	96.12 12	97.27 24	98.88 19	98.46 22	98.47 15	98.39 12	99.52 13	99.22 31
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
HPM-MVS++		98.34 13	98.47 12	98.18 13	99.46 8	99.15 27	99.10 14	97.69 5	97.67 20	94.93 22	97.62 16	99.70 4	98.60 18	98.45 16	97.46 40	99.31 66	99.26 26

HSP-MVS		98.59 2	98.65 7	98.52 3	99.44 10	99.57 1	99.34 3	97.65 6	97.36 29	96.62 8	98.49 5	99.65 5	98.67 17	98.60 11	97.44 41	99.40 50	99.46 10

ACMMP_Plus		98.20 15	98.49 10	97.85 22	99.50 4	99.40 7	99.26 9	97.64 7	97.47 27	92.62 43	97.59 17	99.09 16	98.71 15	98.82 9	97.86 30	99.40 50	99.19 35

SMA-MVS		98.58 4	98.88 2	98.24 10	99.58 1	99.44 6	99.05 16	97.63 8	97.76 14	94.92 23	97.94 12	99.84 1	98.85 9	98.95 4	98.70 5	99.44 40	99.50 7

zzz-MVS		98.43 9	98.31 20	98.57 2	99.48 5	99.40 7	99.32 6	97.62 9	97.70 17	96.67 6	96.59 28	99.09 16	98.86 7	98.65 10	97.56 37	99.45 30	99.17 39

MCST-MVS		98.20 15	98.36 15	98.01 19	99.40 12	99.05 30	99.00 19	97.62 9	97.59 24	93.70 30	97.42 23	99.30 8	98.77 13	98.39 20	97.48 39	99.59 4	99.31 20

SD-MVS		98.52 5	98.77 4	98.23 12	98.15 45	99.26 19	98.79 24	97.59 11	98.52 1	96.25 11	97.99 11	99.75 3	99.01 3	98.27 23	97.97 24	99.59 4	99.63 1

CNVR-MVS		98.47 8	98.46 13	98.48 4	99.40 12	99.05 30	99.02 18	97.54 12	97.73 15	96.65 7	97.20 25	99.13 14	98.85 9	98.91 6	98.10 18	99.41 48	99.08 45

NCCC		98.10 18	98.05 27	98.17 15	99.38 16	99.05 30	99.00 19	97.53 13	98.04 10	95.12 20	94.80 44	99.18 12	98.58 19	98.49 14	97.78 32	99.39 52	98.98 61

HFP-MVS		98.48 7	98.62 8	98.32 7	99.39 15	99.33 14	99.27 8	97.42 14	98.27 5	95.25 19	98.34 8	98.83 21	99.08 1	98.26 24	98.08 20	99.48 21	99.26 26

MP-MVS		98.09 19	98.30 21	97.84 23	99.34 17	99.19 25	99.23 10	97.40 15	97.09 37	93.03 37	97.58 18	98.85 20	98.57 20	98.44 18	97.69 33	99.48 21	99.23 29

DeepC-MVS	94.87 4	96.76 42	96.50 46	97.05 32	98.21 44	99.28 17	98.67 25	97.38 16	97.31 30	90.36 62	89.19 92	93.58 62	98.19 25	98.31 21	98.50 7	99.51 17	99.36 15
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
ACMMPR		98.40 10	98.49 10	98.28 9	99.41 11	99.40 7	99.36 2	97.35 17	98.30 4	95.02 21	97.79 14	98.39 31	99.04 2	98.26 24	98.10 18	99.50 19	99.22 31

CP-MVS		98.32 14	98.34 18	98.29 8	99.34 17	99.30 15	99.15 11	97.35 17	97.49 26	95.58 17	97.72 15	98.62 28	98.82 11	98.29 22	97.67 34	99.51 17	99.28 21

AdaColmap		97.53 27	96.93 40	98.24 10	99.21 20	98.77 62	98.47 31	97.34 19	96.68 45	96.52 10	95.11 41	96.12 50	98.72 14	97.19 55	96.24 71	99.17 89	98.39 104

SteuartSystems-ACMMP		98.38 11	98.71 6	97.99 20	99.34 17	99.46 5	99.34 3	97.33 20	97.31 30	94.25 26	98.06 9	99.17 13	98.13 26	98.98 2	98.46 9	99.55 9	99.54 6
Skip Steuart: Steuart Systems R&D Blog.
X-MVS		97.84 21	98.19 24	97.42 27	99.40 12	99.35 10	99.06 15	97.25 21	97.38 28	90.85 51	96.06 32	98.72 24	98.53 21	98.41 19	98.15 17	99.46 26	99.28 21

DeepC-MVS_fast	96.13 1	98.13 17	98.27 22	97.97 21	99.16 22	99.03 35	99.05 16	97.24 22	98.22 7	94.17 28	95.82 33	98.07 33	98.69 16	98.83 8	98.80 2	99.52 13	99.10 42
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
TSAR-MVS + ACMM		97.71 25	98.60 9	96.66 36	98.64 36	99.05 30	98.85 23	97.23 23	98.45 2	89.40 74	97.51 20	99.27 10	96.88 57	98.53 12	97.81 31	98.96 113	99.59 4

TSAR-MVS + MP.		98.49 6	98.78 3	98.15 16	98.14 46	99.17 26	99.34 3	97.18 24	98.44 3	95.72 15	97.84 13	99.28 9	98.87 6	99.05 1	98.05 21	99.66 1	99.60 3

PCF-MVS	93.95 6	95.65 47	95.14 63	96.25 40	97.73 52	98.73 65	97.59 47	97.13 25	92.50 126	89.09 78	89.85 89	96.65 44	96.90 56	94.97 127	94.89 113	99.08 102	98.38 105
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
train_agg		97.65 26	98.06 26	97.18 30	98.94 28	98.91 55	98.98 22	97.07 26	96.71 44	90.66 56	97.43 22	99.08 18	98.20 24	97.96 36	97.14 49	99.22 83	99.19 35

MSLP-MVS++		98.04 20	97.93 29	98.18 13	99.10 23	99.09 29	98.34 33	96.99 27	97.54 25	96.60 9	94.82 43	98.45 30	98.89 5	97.46 48	98.77 4	99.17 89	99.37 14

CPTT-MVS		97.78 23	97.54 30	98.05 18	98.91 30	99.05 30	99.00 19	96.96 28	97.14 35	95.92 14	95.50 36	98.78 23	98.99 4	97.20 53	96.07 75	98.54 164	99.04 53

ACMMP		97.37 30	97.48 32	97.25 28	98.88 32	99.28 17	98.47 31	96.86 29	97.04 39	92.15 44	97.57 19	96.05 52	97.67 34	97.27 51	95.99 79	99.46 26	99.14 41
Qingshan Xu, Weihang Kong, Wenbing Tao, Marc Pollefeys: Multi-Scale Geometric Consistency Guided and Planar Prior Assisted Multi-View Stereo. IEEE Transactions on Pattern Analysis and Machine Intelligence
PLC	94.95 3	97.37 30	96.77 43	98.07 17	98.97 27	98.21 86	97.94 41	96.85 30	97.66 21	97.58 1	93.33 52	96.84 42	98.01 30	97.13 57	96.20 74	99.09 101	98.01 122
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
CNLPA		96.90 38	96.28 49	97.64 25	98.56 38	98.63 72	96.85 58	96.60 31	97.73 15	97.08 4	89.78 90	96.28 49	97.80 33	96.73 72	96.63 59	98.94 114	98.14 118

MSDG		94.82 60	93.73 90	96.09 43	98.34 42	97.43 99	97.06 52	96.05 32	95.84 68	90.56 57	86.30 122	89.10 84	95.55 75	96.13 103	95.61 98	99.00 110	95.73 184

DeepPCF-MVS	95.28 2	97.00 36	98.35 17	95.42 53	97.30 56	98.94 43	94.82 111	96.03 33	98.24 6	92.11 45	95.80 34	98.64 27	95.51 76	98.95 4	98.66 6	96.78 195	99.20 34

PHI-MVS		97.78 23	98.44 14	97.02 33	98.73 33	99.25 21	98.11 36	95.54 34	96.66 46	92.79 40	98.52 4	99.38 6	97.50 38	97.84 39	98.39 12	99.45 30	99.03 54

CSCG		97.44 29	97.18 36	97.75 24	99.47 6	99.52 3	98.55 29	95.41 35	97.69 19	95.72 15	94.29 47	95.53 54	98.10 27	96.20 100	97.38 44	99.24 77	99.62 2

CDPH-MVS		96.84 39	97.49 31	96.09 43	98.92 29	98.85 59	98.61 26	95.09 36	96.00 62	87.29 97	95.45 38	97.42 37	97.16 43	97.83 40	97.94 26	99.44 40	98.92 66

OMC-MVS		97.00 36	96.92 41	97.09 31	98.69 34	98.66 68	97.85 42	95.02 37	98.09 9	94.47 24	93.15 53	96.90 40	97.38 39	97.16 56	96.82 57	99.13 96	97.65 146

TAPA-MVS	94.18 5	96.38 43	96.49 47	96.25 40	98.26 43	98.66 68	98.00 39	94.96 38	97.17 34	89.48 71	92.91 56	96.35 46	97.53 37	96.59 79	95.90 83	99.28 70	97.82 135
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
TSAR-MVS + COLMAP		94.79 62	94.51 73	95.11 56	96.50 64	97.54 95	97.99 40	94.54 39	97.81 13	85.88 100	96.73 27	81.28 127	96.99 55	96.29 96	95.21 107	98.76 148	96.73 173

PGM-MVS		97.81 22	98.11 25	97.46 26	99.55 3	99.34 13	99.32 6	94.51 40	96.21 55	93.07 34	98.05 10	97.95 36	98.82 11	98.22 27	97.89 29	99.48 21	99.09 44

OPM-MVS		93.61 87	92.43 111	95.00 60	96.94 61	97.34 100	97.78 43	94.23 41	89.64 161	85.53 101	88.70 96	82.81 120	96.28 66	96.28 97	95.00 112	99.24 77	97.22 157

HQP-MVS		94.43 68	94.57 71	94.27 76	96.41 67	97.23 102	96.89 56	93.98 42	95.94 64	83.68 109	95.01 42	84.46 108	95.58 74	95.47 118	94.85 115	99.07 104	99.00 58

MVS_111021_LR		97.16 33	98.01 28	96.16 42	98.47 39	98.98 40	96.94 55	93.89 43	97.64 22	91.44 48	98.89 1	96.41 45	97.20 42	98.02 35	97.29 48	99.04 109	98.85 73

abl_6				96.82 35	98.60 37	98.74 63	97.74 44	93.73 44	96.25 53	94.37 25	94.55 46	98.60 29	97.25 41			99.27 72	98.61 85

EPNet		96.27 45	96.97 39	95.46 52	98.47 39	98.28 81	97.41 49	93.67 45	95.86 67	92.86 39	97.51 20	93.79 60	91.76 130	97.03 58	97.03 50	98.61 160	99.28 21
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
ACMM	92.75 10	94.41 70	93.84 87	95.09 57	96.41 67	96.80 112	94.88 110	93.54 46	96.41 49	90.16 63	92.31 64	83.11 119	96.32 64	96.22 99	94.65 117	99.22 83	97.35 153
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
CANet		96.84 39	97.20 34	96.42 37	97.92 48	99.24 23	98.60 27	93.51 47	97.11 36	93.07 34	91.16 74	97.24 39	96.21 67	98.24 26	98.05 21	99.22 83	99.35 16

3Dnovator	93.79 8	97.08 34	97.20 34	96.95 34	99.09 24	99.03 35	98.20 35	93.33 48	97.99 11	93.82 29	90.61 82	96.80 43	97.82 31	97.90 38	98.78 3	99.47 24	99.26 26

TSAR-MVS + GP.		97.45 28	98.36 15	96.39 38	95.56 77	98.93 49	97.74 44	93.31 49	97.61 23	94.24 27	98.44 7	99.19 11	98.03 29	97.60 44	97.41 43	99.44 40	99.33 18

DELS-MVS		96.06 46	96.04 52	96.07 45	97.77 50	99.25 21	98.10 37	93.26 50	94.42 92	92.79 40	88.52 99	93.48 63	95.06 82	98.51 13	98.83 1	99.45 30	99.28 21
Christian Sormann, Emanuele Santellani, Mattia Rossi, Andreas Kuhn, Friedrich Fraundorfer: DELS-MVS: Deep Epipolar Line Search for Multi-View Stereo. Winter Conference on Applications of Computer Vision (WACV), 2023
QAPM		96.78 41	97.14 37	96.36 39	99.05 25	99.14 28	98.02 38	93.26 50	97.27 32	90.84 54	91.16 74	97.31 38	97.64 36	97.70 42	98.20 15	99.33 61	99.18 38

3Dnovator+	93.91 7	97.23 32	97.22 33	97.24 29	98.89 31	98.85 59	98.26 34	93.25 52	97.99 11	95.56 18	90.01 88	98.03 35	98.05 28	97.91 37	98.43 10	99.44 40	99.35 16

MVS_111021_HR		97.04 35	98.20 23	95.69 48	98.44 41	99.29 16	96.59 72	93.20 53	97.70 17	89.94 66	98.46 6	96.89 41	96.71 60	98.11 32	97.95 25	99.27 72	99.01 57

COLMAP_ROB	90.49 14	93.27 98	92.71 102	93.93 81	97.75 51	97.44 98	96.07 83	93.17 54	95.40 76	83.86 108	83.76 136	88.72 86	93.87 96	94.25 139	94.11 132	98.87 119	95.28 190
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
MVS_0304		96.31 44	96.91 42	95.62 49	97.21 58	99.20 24	98.55 29	93.10 55	97.04 39	89.73 68	90.30 84	96.35 46	95.71 71	98.14 29	97.93 28	99.38 54	99.40 13

PVSNet_BlendedMVS		95.41 52	95.28 60	95.57 50	97.42 54	99.02 37	95.89 86	93.10 55	96.16 56	93.12 32	91.99 66	85.27 103	94.66 84	98.09 33	97.34 45	99.24 77	99.08 45

PVSNet_Blended		95.41 52	95.28 60	95.57 50	97.42 54	99.02 37	95.89 86	93.10 55	96.16 56	93.12 32	91.99 66	85.27 103	94.66 84	98.09 33	97.34 45	99.24 77	99.08 45

OpenMVS	92.33 11	95.50 48	95.22 62	95.82 47	98.98 26	98.97 41	97.67 46	93.04 58	94.64 89	89.18 76	84.44 131	94.79 56	96.79 58	97.23 52	97.61 35	99.24 77	98.88 70

EPNet_dtu		92.45 107	95.02 66	89.46 136	98.02 47	95.47 154	94.79 112	92.62 59	94.97 86	70.11 200	94.76 45	92.61 67	84.07 206	95.94 106	95.56 99	97.15 192	95.82 183
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
thresconf0.02		93.57 88	93.84 87	93.25 94	95.03 103	98.16 88	95.80 91	92.46 60	96.12 58	83.88 107	92.61 59	80.39 128	92.83 115	96.11 104	96.21 73	99.49 20	97.28 156

conf0.002		93.20 99	91.63 124	95.02 58	95.31 89	98.94 43	96.82 59	92.43 61	92.63 118	88.99 79	88.16 102	70.49 198	97.12 46	96.77 69	96.30 63	99.44 40	98.16 117

conf0.01		93.33 96	91.89 121	95.00 60	95.32 88	98.94 43	96.82 59	92.41 62	92.63 118	88.91 80	88.02 103	72.75 179	97.12 46	96.78 68	95.85 87	99.44 40	98.27 111

thres400		93.56 89	92.43 111	94.87 67	95.40 82	98.91 55	96.70 68	92.38 63	92.93 115	88.19 88	86.69 113	77.35 143	97.13 44	96.75 71	95.85 87	99.42 47	98.56 87

tfpn111		94.05 74	93.34 99	94.88 64	95.33 84	98.94 43	96.82 59	92.31 64	92.63 118	88.26 85	92.61 59	78.01 138	97.12 46	96.82 62	95.85 87	99.45 30	98.56 87

conf200view11		93.64 83	92.57 103	94.88 64	95.33 84	98.94 43	96.82 59	92.31 64	92.63 118	88.26 85	87.21 105	78.01 138	97.12 46	96.82 62	95.85 87	99.45 30	98.56 87

tfpn200view9		93.64 83	92.57 103	94.89 63	95.33 84	98.94 43	96.82 59	92.31 64	92.63 118	88.29 82	87.21 105	78.01 138	97.12 46	96.82 62	95.85 87	99.45 30	98.56 87

thres600view7		93.49 94	92.37 115	94.79 70	95.42 78	98.93 49	96.58 73	92.31 64	93.04 113	87.88 90	86.62 115	76.94 146	97.09 52	96.82 62	95.63 96	99.45 30	98.63 83

thres200		93.62 86	92.54 105	94.88 64	95.36 83	98.93 49	96.75 67	92.31 64	92.84 116	88.28 84	86.99 108	77.81 142	97.13 44	96.82 62	95.92 81	99.45 30	98.49 97

thres100view900		93.55 92	92.47 110	94.81 68	95.33 84	98.74 63	96.78 66	92.30 69	92.63 118	88.29 82	87.21 105	78.01 138	96.78 59	96.38 92	95.92 81	99.38 54	98.40 103

view600		93.50 93	92.39 114	94.80 69	95.41 81	98.93 49	96.60 71	92.30 69	93.09 112	87.96 89	86.67 114	76.97 145	97.12 46	96.83 61	95.64 95	99.43 46	98.62 84

CLD-MVS		94.79 62	94.36 77	95.30 55	95.21 95	97.46 97	97.23 51	92.24 71	96.43 48	91.77 47	92.69 58	84.31 109	96.06 68	95.52 117	95.03 109	99.31 66	99.06 49
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
view800		93.45 95	92.37 115	94.71 71	95.42 78	98.92 53	96.51 75	92.19 72	93.14 111	87.62 92	86.72 112	76.54 149	97.08 53	96.86 60	95.74 92	99.45 30	98.70 80

tfpn		92.91 101	91.44 128	94.63 73	95.42 78	98.92 53	96.41 78	92.10 73	93.19 109	87.34 96	86.85 109	69.20 206	97.01 54	96.88 59	96.28 67	99.47 24	98.75 79

tfpn_ndepth		94.36 71	94.64 69	94.04 79	95.16 97	98.51 76	95.58 94	92.09 74	95.78 70	88.52 81	92.38 63	85.74 100	93.34 107	96.39 90	95.90 83	99.54 11	97.79 137

ACMP	92.88 9	94.43 68	94.38 76	94.50 74	96.01 72	97.69 94	95.85 89	92.09 74	95.74 71	89.12 77	95.14 40	82.62 122	94.77 83	95.73 113	94.67 116	99.14 95	99.06 49
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
tfpnview11		93.63 85	94.42 74	92.71 99	95.08 100	98.26 84	95.58 94	92.06 76	96.32 50	81.88 116	93.44 49	83.43 115	92.14 122	96.58 81	95.88 85	99.52 13	97.07 164

PatchMatch-RL		94.69 65	94.41 75	95.02 58	97.63 53	98.15 89	94.50 117	91.99 77	95.32 78	91.31 49	95.47 37	83.44 114	96.02 70	96.56 82	95.23 106	98.69 155	96.67 174

tfpn_n400		93.56 89	94.36 77	92.63 100	95.07 101	98.28 81	95.50 98	91.98 78	95.48 74	81.88 116	93.44 49	83.43 115	92.01 125	96.60 77	96.27 68	99.34 59	97.04 165

tfpnconf		93.56 89	94.36 77	92.63 100	95.07 101	98.28 81	95.50 98	91.98 78	95.48 74	81.88 116	93.44 49	83.43 115	92.01 125	96.60 77	96.27 68	99.34 59	97.04 165

tfpn1000		94.14 72	94.54 72	93.67 87	95.27 92	98.50 77	95.36 100	91.84 80	96.31 51	87.38 95	92.98 55	84.04 110	92.60 117	96.49 89	95.62 97	99.55 9	97.82 135

LGP-MVS_train		94.12 73	94.62 70	93.53 89	96.44 66	97.54 95	97.40 50	91.84 80	94.66 88	81.09 125	95.70 35	83.36 118	95.10 81	96.36 94	95.71 93	99.32 63	99.03 54

PVSNet_Blended_VisFu		94.77 64	95.54 58	93.87 83	96.48 65	98.97 41	94.33 119	91.84 80	94.93 87	90.37 61	85.04 127	94.99 55	90.87 146	98.12 31	97.30 47	99.30 68	99.45 11

conf0.05thres1000		92.47 106	91.39 129	93.73 86	95.21 95	98.52 75	95.66 92	91.56 83	90.87 147	84.27 105	82.79 142	76.12 150	96.29 65	96.59 79	95.68 94	99.39 52	99.19 35

RPSCF		94.05 74	94.00 83	94.12 78	96.20 69	96.41 126	96.61 70	91.54 84	95.83 69	89.73 68	96.94 26	92.80 66	95.35 79	91.63 191	90.44 199	95.27 211	93.94 200

UGNet		94.92 57	96.63 44	92.93 97	96.03 71	98.63 72	94.53 116	91.52 85	96.23 54	90.03 64	92.87 57	96.10 51	86.28 193	96.68 74	96.60 60	99.16 92	99.32 19
Wanjuan Su, Qingshan Xu, Wenbing Tao: Uncertainty-guided Multi-view Stereo Network for Depth Estimation. IEEE Transactions on Circuits and Systems for Video Technology, 2022
IB-MVS	89.56 15	91.71 113	92.50 107	90.79 119	95.94 73	98.44 78	87.05 205	91.38 86	93.15 110	92.98 38	84.78 128	85.14 106	78.27 213	92.47 166	94.44 128	99.10 100	99.08 45
Christian Sormann, Mattia Rossi, Andreas Kuhn and Friedrich Fraundorfer: IB-MVS: An Iterative Algorithm for Deep Multi-View Stereo based on Binary Decisions. BMVC 2021
FC-MVSNet-train		93.85 79	93.91 84	93.78 85	94.94 104	96.79 115	94.29 120	91.13 87	93.84 102	88.26 85	90.40 83	85.23 105	94.65 86	96.54 84	95.31 104	99.38 54	99.28 21

IS_MVSNet		95.28 54	96.43 48	93.94 80	95.30 90	99.01 39	95.90 84	91.12 88	94.13 98	87.50 94	91.23 73	94.45 58	94.17 93	98.45 16	98.50 7	99.65 2	99.23 29

Vis-MVSNet (Re-imp)		94.46 67	96.24 50	92.40 103	95.23 94	98.64 70	95.56 96	90.99 89	94.42 92	85.02 103	90.88 80	94.65 57	88.01 183	98.17 28	98.37 14	99.57 8	98.53 92

EPP-MVSNet		95.27 55	96.18 51	94.20 77	94.88 105	98.64 70	94.97 106	90.70 90	95.34 77	89.67 70	91.66 70	93.84 59	95.42 78	97.32 50	97.00 51	99.58 6	99.47 9

DI_MVS_plusplus_trai		94.01 76	93.63 92	94.44 75	94.54 111	98.26 84	97.51 48	90.63 91	95.88 66	89.34 75	80.54 152	89.36 81	95.48 77	96.33 95	96.27 68	99.17 89	98.78 77

MVSTER		94.89 58	95.07 65	94.68 72	94.71 108	96.68 118	97.00 53	90.57 92	95.18 84	93.05 36	95.21 39	86.41 95	93.72 100	97.59 45	95.88 85	99.00 110	98.50 96

UniMVSNet_NR-MVSNet		90.35 131	89.96 140	90.80 118	89.66 164	95.83 143	92.48 143	90.53 93	90.96 146	79.57 130	79.33 156	77.14 144	93.21 111	92.91 160	94.50 127	99.37 57	99.05 51

TranMVSNet+NR-MVSNet		89.23 146	88.48 154	90.11 131	89.07 194	95.25 169	92.91 137	90.43 94	90.31 154	77.10 140	76.62 163	71.57 193	91.83 129	92.12 177	94.59 120	99.32 63	98.92 66

TDRefinement		89.07 149	88.15 157	90.14 129	95.16 97	96.88 107	95.55 97	90.20 95	89.68 159	76.42 145	76.67 162	74.30 161	84.85 200	93.11 156	91.91 190	98.64 159	94.47 194

tfpnnormal		88.50 154	87.01 190	90.23 125	91.36 144	95.78 145	92.74 138	90.09 96	83.65 213	76.33 146	71.46 207	69.58 204	91.84 128	95.54 116	94.02 135	99.06 107	99.03 54

UA-Net		93.96 77	95.95 53	91.64 109	96.06 70	98.59 74	95.29 101	90.00 97	91.06 144	82.87 112	90.64 81	98.06 34	86.06 194	98.14 29	98.20 15	99.58 6	96.96 167

DU-MVS		89.67 140	88.84 150	90.63 121	89.26 188	95.61 148	92.48 143	89.91 98	91.22 142	79.57 130	77.72 160	71.18 195	93.21 111	92.53 164	94.57 121	99.35 58	99.05 51

NR-MVSNet		89.34 143	88.66 151	90.13 130	90.40 154	95.61 148	93.04 136	89.91 98	91.22 142	78.96 133	77.72 160	68.90 208	89.16 177	94.24 140	93.95 136	99.32 63	98.99 59

canonicalmvs		95.25 56	95.45 59	95.00 60	95.27 92	98.72 66	96.89 56	89.82 100	96.51 47	90.84 54	93.72 48	86.01 98	97.66 35	95.78 112	97.94 26	99.54 11	99.50 7

MAR-MVS		95.50 48	95.60 56	95.39 54	98.67 35	98.18 87	95.89 86	89.81 101	94.55 91	91.97 46	92.99 54	90.21 77	97.30 40	96.79 67	97.49 38	98.72 151	98.99 59
Zhenyu Xu, Yiguang Liu, Xuelei Shi, Ying Wang, Yunan Zheng: MARMVS: Matching Ambiguity Reduced Multiple View Stereo for Efficient Large Scale Scene Reconstruction. CVPR 2020
Baseline_NR-MVSNet		89.27 145	88.01 160	90.73 120	89.26 188	93.71 197	92.71 140	89.78 102	90.73 149	81.28 124	73.53 194	72.85 174	92.30 121	92.53 164	93.84 141	99.07 104	98.88 70

ACMH	90.77 13	91.51 118	91.63 124	91.38 111	95.62 76	96.87 109	91.76 176	89.66 103	91.58 139	78.67 134	86.73 111	78.12 136	93.77 99	94.59 130	94.54 124	98.78 142	98.98 61
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
TransMVSNet (Re)		87.73 176	86.79 192	88.83 145	90.76 150	94.40 192	91.33 182	89.62 104	84.73 209	75.41 153	72.73 198	71.41 194	86.80 190	94.53 132	93.93 137	99.06 107	95.83 182

PMMVS		94.61 66	95.56 57	93.50 90	94.30 114	96.74 116	94.91 109	89.56 105	95.58 73	87.72 91	96.15 31	92.86 65	96.06 68	95.47 118	95.02 110	98.43 172	97.09 160

ACMH+	90.88 12	91.41 119	91.13 131	91.74 108	95.11 99	96.95 106	93.13 134	89.48 106	92.42 128	79.93 129	85.13 126	78.02 137	93.82 98	93.49 151	93.88 138	98.94 114	97.99 123

UniMVSNet (Re)		90.03 136	89.61 144	90.51 122	89.97 161	96.12 131	92.32 151	89.26 107	90.99 145	80.95 126	78.25 159	75.08 158	91.14 137	93.78 144	93.87 139	99.41 48	99.21 33

CVMVSNet		89.77 139	91.66 123	87.56 181	93.21 131	95.45 155	91.94 175	89.22 108	89.62 162	69.34 205	83.99 134	85.90 99	84.81 201	94.30 138	95.28 105	96.85 194	97.09 160

CDS-MVSNet		92.77 102	93.60 93	91.80 107	92.63 135	96.80 112	95.24 102	89.14 109	90.30 155	84.58 104	86.76 110	90.65 74	90.42 162	95.89 107	96.49 61	98.79 135	98.32 109
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
diffmvs		94.83 59	95.64 55	93.89 82	94.73 107	97.96 93	96.49 76	89.13 110	96.82 42	89.47 72	91.66 70	93.63 61	95.15 80	94.76 128	95.93 80	98.36 174	98.69 81

MVS_Test		94.82 60	95.66 54	93.84 84	94.79 106	98.35 80	96.49 76	89.10 111	96.12 58	87.09 98	92.58 61	90.61 75	96.48 63	96.51 88	96.89 54	99.11 99	98.54 91

GBi-Net		93.81 80	94.18 80	93.38 91	91.34 145	95.86 140	96.22 80	88.68 112	95.23 81	90.40 58	86.39 118	91.16 70	94.40 90	96.52 85	96.30 63	99.21 86	97.79 137

test1		93.81 80	94.18 80	93.38 91	91.34 145	95.86 140	96.22 80	88.68 112	95.23 81	90.40 58	86.39 118	91.16 70	94.40 90	96.52 85	96.30 63	99.21 86	97.79 137

FMVSNet3		93.79 82	94.17 82	93.35 93	91.21 148	95.99 133	96.62 69	88.68 112	95.23 81	90.40 58	86.39 118	91.16 70	94.11 94	95.96 105	96.67 58	99.07 104	97.79 137

CHOSEN 1792x2688		92.66 104	92.49 108	92.85 98	97.13 59	98.89 58	95.90 84	88.50 115	95.32 78	83.31 111	71.99 202	88.96 85	94.10 95	96.69 73	96.49 61	98.15 177	99.10 42

Vis-MVSNet		92.77 102	95.00 67	90.16 127	94.10 117	98.79 61	94.76 113	88.26 116	92.37 131	79.95 128	88.19 101	91.58 69	84.38 203	97.59 45	97.58 36	99.52 13	98.91 68
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
FMVSNet2		93.30 97	93.36 98	93.22 95	91.34 145	95.86 140	96.22 80	88.24 117	95.15 85	89.92 67	81.64 146	89.36 81	94.40 90	96.77 69	96.98 52	99.21 86	97.79 137

v148		87.51 179	86.79 192	88.36 153	89.39 180	95.21 170	89.84 196	88.20 118	87.61 192	77.56 137	73.38 196	70.32 201	86.80 190	90.70 197	92.31 185	98.37 173	97.98 125

pm-mvs1		89.19 147	89.02 149	89.38 138	90.40 154	95.74 146	92.05 168	88.10 119	86.13 204	77.70 136	73.72 193	79.44 132	88.97 178	95.81 111	94.51 126	99.08 102	97.78 143

v1141		88.17 163	87.69 172	88.74 147	89.44 174	95.41 161	92.25 160	87.98 120	88.38 180	73.54 179	74.43 179	72.71 183	90.45 160	92.08 184	92.72 176	98.79 135	98.09 119

v1		88.17 163	87.66 174	88.77 146	89.44 174	95.40 163	92.29 155	87.98 120	88.21 187	73.75 173	74.41 181	72.75 179	90.36 168	92.07 187	92.71 179	98.80 129	98.09 119

divwei89l23v2f112		88.17 163	87.69 172	88.74 147	89.44 174	95.41 161	92.26 158	87.97 122	88.29 184	73.57 178	74.45 178	72.75 179	90.42 162	92.08 184	92.72 176	98.81 126	98.09 119

pmmvs4		90.55 128	89.91 141	91.30 112	90.26 158	94.95 177	92.73 139	87.94 123	93.44 108	85.35 102	82.28 145	76.09 152	93.02 114	93.56 149	92.26 188	98.51 166	96.77 172

v2v482		88.25 160	87.71 171	88.88 144	89.23 192	95.28 166	92.10 166	87.89 124	88.69 175	73.31 181	75.32 166	71.64 191	91.89 127	92.10 183	92.92 156	98.86 121	97.99 123

v6		88.43 155	88.01 160	88.92 141	89.60 170	95.43 160	92.36 147	87.66 125	89.07 168	74.50 165	75.06 169	73.47 169	90.59 158	92.11 180	92.76 172	98.79 135	98.18 114

v1neww		88.41 156	88.00 163	88.89 142	89.61 168	95.44 158	92.31 152	87.65 126	89.09 166	74.30 168	75.02 171	73.42 171	90.68 153	92.12 177	92.77 168	98.79 135	98.18 114

v7new		88.41 156	88.00 163	88.89 142	89.61 168	95.44 158	92.31 152	87.65 126	89.09 166	74.30 168	75.02 171	73.42 171	90.68 153	92.12 177	92.77 168	98.79 135	98.18 114

GA-MVS		89.28 144	90.75 137	87.57 180	91.77 142	96.48 123	92.29 155	87.58 128	90.61 152	65.77 211	84.48 130	76.84 147	89.46 175	95.84 109	93.68 143	98.52 165	97.34 154

CANet_DTU		93.92 78	96.57 45	90.83 117	95.63 75	98.39 79	96.99 54	87.38 129	96.26 52	71.97 187	96.31 30	93.02 64	94.53 87	97.38 49	96.83 56	98.49 167	97.79 137

FC-MVSNet-test		91.63 114	93.82 89	89.08 140	92.02 141	96.40 127	93.26 132	87.26 130	93.72 103	77.26 139	88.61 98	89.86 79	85.50 196	95.72 115	95.02 110	99.16 92	97.44 150

V42		88.31 159	87.95 166	88.73 149	89.44 174	95.34 165	92.23 162	87.21 131	88.83 172	74.49 166	74.89 175	73.43 170	90.41 165	92.08 184	92.77 168	98.60 162	98.33 107

FMVSNet1		91.54 117	90.93 134	92.26 104	90.35 156	95.27 168	95.22 103	87.16 132	91.37 141	87.62 92	75.45 165	83.84 112	94.43 88	96.52 85	96.30 63	98.82 122	97.74 144

test-LLR		91.62 115	93.56 95	89.35 139	93.31 129	96.57 121	92.02 171	87.06 133	92.34 132	75.05 159	90.20 85	88.64 87	90.93 142	96.19 101	94.07 133	97.75 187	96.90 170

test0.0.03 1		91.97 109	93.91 84	89.72 132	93.31 129	96.40 127	91.34 181	87.06 133	93.86 100	81.67 121	91.15 76	89.16 83	86.02 195	95.08 124	95.09 108	98.91 117	96.64 176

USDC		90.69 125	90.52 138	90.88 116	94.17 116	96.43 125	95.82 90	86.76 135	93.92 99	76.27 147	86.49 117	74.30 161	93.67 102	95.04 126	93.36 147	98.61 160	94.13 198

WR-MVS		87.93 168	88.09 158	87.75 172	89.26 188	95.28 166	90.81 187	86.69 136	88.90 171	75.29 155	74.31 182	73.72 164	85.19 199	92.26 167	93.32 149	99.27 72	98.81 75

HyFIR lowres test		92.03 108	91.55 126	92.58 102	97.13 59	98.72 66	94.65 114	86.54 137	93.58 106	82.56 114	67.75 217	90.47 76	95.67 72	95.87 108	95.54 100	98.91 117	98.93 65

TinyColmap		89.42 141	88.58 152	90.40 124	93.80 124	95.45 155	93.96 124	86.54 137	92.24 134	76.49 144	80.83 150	70.44 199	93.37 106	94.45 134	93.30 150	98.26 176	93.37 208

PEN-MVS		87.22 189	86.50 199	88.07 161	88.88 197	94.44 191	90.99 186	86.21 139	86.53 199	73.66 176	74.97 174	66.56 217	89.42 176	91.20 193	93.48 146	99.24 77	98.31 110

N_pmnet		84.80 202	85.10 206	84.45 202	89.25 191	92.86 200	84.04 211	86.21 139	88.78 173	66.73 210	72.41 201	74.87 160	85.21 198	88.32 209	86.45 219	95.30 210	92.04 210

IterMVS-LS		92.56 105	93.18 100	91.84 106	93.90 120	94.97 176	94.99 105	86.20 141	94.18 97	82.68 113	85.81 124	87.36 92	94.43 88	95.31 120	96.02 78	98.87 119	98.60 86
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
DWT-MVSNet_training		91.30 120	89.73 142	93.13 96	94.64 110	96.87 109	94.93 107	86.17 142	94.22 96	93.18 31	89.11 93	73.28 173	93.59 103	88.00 210	90.73 197	96.26 198	95.87 181

TAMVS		90.54 129	90.87 136	90.16 127	91.48 143	96.61 120	93.26 132	86.08 143	87.71 190	81.66 122	83.11 141	84.04 110	90.42 162	94.54 131	94.60 119	98.04 182	95.48 188

DTE-MVSNet		86.67 192	86.09 200	87.35 184	88.45 203	94.08 195	90.65 188	86.05 144	86.13 204	72.19 186	74.58 177	66.77 215	87.61 186	90.31 199	93.12 151	99.13 96	97.62 147

Effi-MVS+		92.93 100	93.86 86	91.86 105	94.07 118	98.09 91	95.59 93	85.98 145	94.27 95	79.54 132	91.12 77	81.81 124	96.71 60	96.67 75	96.06 76	99.27 72	98.98 61

MDA-MVSNet-bldmvs		80.11 214	80.24 216	79.94 212	77.01 229	93.21 198	78.86 223	85.94 146	82.71 217	60.86 217	79.71 155	51.77 230	83.71 208	75.60 228	86.37 220	93.28 224	92.35 209

CP-MVSNet		87.89 172	87.27 181	88.62 150	89.30 184	95.06 173	90.60 189	85.78 147	87.43 194	75.98 148	74.60 176	68.14 210	90.76 147	93.07 158	93.60 144	99.30 68	98.98 61

PS-CasMVS		87.33 187	86.68 195	88.10 160	89.22 193	94.93 178	90.35 192	85.70 148	86.44 200	74.01 170	73.43 195	66.59 216	90.04 172	92.92 159	93.52 145	99.28 70	98.91 68

v1144		87.92 171	87.79 169	88.07 161	89.27 187	95.15 171	92.17 165	85.62 149	88.52 176	71.52 189	73.80 192	72.40 188	91.06 140	93.54 150	92.80 162	98.81 126	98.33 107

WR-MVS_H		87.93 168	87.85 168	88.03 166	89.62 166	95.58 152	90.47 190	85.55 150	87.20 196	76.83 142	74.42 180	72.67 185	86.37 192	93.22 155	93.04 152	99.33 61	98.83 74

CMPMVS	65.18 17	84.76 203	83.10 212	86.69 190	95.29 91	95.05 174	88.37 200	85.51 151	80.27 220	71.31 191	68.37 215	73.85 163	85.25 197	87.72 211	87.75 210	94.38 220	88.70 220
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
CHOSEN 280x420		95.46 50	97.01 38	93.66 88	97.28 57	97.98 92	96.40 79	85.39 152	96.10 60	91.07 50	96.53 29	96.34 48	95.61 73	97.65 43	96.95 53	96.21 199	97.49 148

v7		88.18 162	88.01 160	88.39 152	89.45 173	95.14 172	92.36 147	85.37 153	89.29 165	72.94 184	73.98 189	72.77 177	91.38 134	93.59 145	92.87 158	98.82 122	98.42 100

EU-MVSNet		85.62 200	87.65 175	83.24 207	88.54 202	92.77 201	87.12 204	85.32 154	86.71 197	64.54 213	78.52 158	75.11 157	78.35 212	92.25 168	92.28 187	95.58 206	95.93 179

SixPastTwentyTwo		88.37 158	89.47 145	87.08 186	90.01 160	95.93 139	87.41 202	85.32 154	90.26 156	70.26 198	86.34 121	71.95 189	90.93 142	92.89 161	91.72 192	98.55 163	97.22 157

pmmvs6		85.98 198	84.89 208	87.25 185	88.83 199	94.35 193	89.36 198	85.30 156	78.51 222	75.44 152	62.71 223	75.41 155	87.65 185	93.58 148	92.40 183	96.89 193	97.29 155

testgi		89.42 141	91.50 127	87.00 188	92.40 138	95.59 150	89.15 199	85.27 157	92.78 117	72.42 185	91.75 69	76.00 153	84.09 205	94.38 136	93.82 142	98.65 158	96.15 177

v1192		87.51 179	87.31 180	87.74 173	89.04 195	94.87 183	92.07 167	85.03 158	88.49 177	70.32 197	72.65 199	70.35 200	91.21 136	93.59 145	92.80 162	98.78 142	98.42 100

v8		88.21 161	87.94 167	88.51 151	89.62 166	95.01 175	92.31 152	84.99 159	88.94 170	74.70 163	75.03 170	73.51 168	90.67 155	92.11 180	92.74 174	98.80 129	98.24 112

Effi-MVS+-dtu		91.78 112	93.59 94	89.68 135	92.44 137	97.11 104	94.40 118	84.94 160	92.43 127	75.48 151	91.09 78	83.75 113	93.55 104	96.61 76	95.47 101	97.24 191	98.67 82

LTVRE_ROB	87.32 16	87.55 178	88.25 156	86.73 189	90.66 151	95.80 144	93.05 135	84.77 161	83.35 214	60.32 220	83.12 140	67.39 211	93.32 108	94.36 137	94.86 114	98.28 175	98.87 72
Andreas Kuhn, Heiko HirschmÃ¼ller, Daniel Scharstein, Helmut Mayer: A TV Prior for High-Quality Scalable Multi-View Stereo Reconstruction. International Journal of Computer Vision 2016
pmmvs5		87.83 174	88.09 158	87.51 183	89.59 171	95.48 153	89.75 197	84.73 162	86.07 206	71.44 190	80.57 151	70.09 202	90.74 149	94.47 133	92.87 158	98.82 122	97.10 159

v144192		87.40 184	87.20 183	87.64 177	88.89 196	94.88 182	91.65 177	84.70 163	87.80 189	71.17 194	73.20 197	70.91 196	90.75 148	92.69 162	92.49 181	98.71 152	98.43 99

Fast-Effi-MVS+		91.87 110	92.08 118	91.62 110	92.91 133	97.21 103	94.93 107	84.60 164	93.61 104	81.49 123	83.50 137	78.95 133	96.62 62	96.55 83	96.22 72	99.16 92	98.51 95

v1921920		87.31 188	87.13 187	87.52 182	88.87 198	94.72 184	91.96 174	84.59 165	88.28 185	69.86 202	72.50 200	70.03 203	91.10 138	93.33 153	92.61 180	98.71 152	98.44 98

v748		85.88 199	85.66 202	86.14 197	88.03 204	94.63 188	87.02 206	84.59 165	84.30 210	74.56 164	70.94 209	67.27 212	83.94 207	90.96 196	92.74 174	98.71 152	98.81 75

MS-PatchMatch		91.82 111	92.51 106	91.02 113	95.83 74	96.88 107	95.05 104	84.55 167	93.85 101	82.01 115	82.51 144	91.71 68	90.52 159	95.07 125	93.03 153	98.13 178	94.52 193

v1240		86.89 190	86.75 194	87.06 187	88.75 200	94.65 187	91.30 183	84.05 168	87.49 193	68.94 206	71.96 203	68.86 209	90.65 156	93.33 153	92.72 176	98.67 156	98.24 112

pmmvs-eth3d		84.33 206	82.94 213	85.96 200	84.16 218	90.94 214	86.55 207	83.79 169	84.25 211	75.85 150	70.64 211	56.43 227	87.44 188	92.20 172	90.41 200	97.97 183	95.68 185

Anonymous20231206		83.84 207	85.19 205	82.26 208	87.38 211	92.87 199	85.49 209	83.65 170	86.07 206	63.44 216	68.42 214	69.01 207	75.45 216	93.34 152	92.44 182	98.12 180	94.20 197

FMVSNet5		90.36 130	90.93 134	89.70 133	87.99 205	92.25 202	92.03 170	83.51 171	92.20 135	84.13 106	85.59 125	86.48 93	92.43 119	94.61 129	94.52 125	98.13 178	90.85 215

v52		86.57 193	86.63 196	86.50 192	87.47 210	94.89 181	89.90 194	83.39 172	86.36 201	71.17 194	71.53 205	71.65 190	88.34 181	91.14 194	92.32 184	98.74 150	98.52 93

V4		86.56 194	86.61 197	86.50 192	87.49 209	94.90 180	89.87 195	83.39 172	86.25 202	71.20 193	71.57 204	71.58 192	88.30 182	91.14 194	92.31 185	98.75 149	98.52 93

v10		88.00 166	87.96 165	88.05 164	89.44 174	94.68 185	92.36 147	83.35 174	89.37 164	72.96 182	73.98 189	72.79 176	91.35 135	93.59 145	92.88 157	98.81 126	98.42 100

v7n		86.43 195	86.52 198	86.33 195	87.91 206	94.93 178	90.15 193	83.05 175	86.57 198	70.21 199	71.48 206	66.78 214	87.72 184	94.19 142	92.96 155	98.92 116	98.76 78

test20.03		82.92 210	85.52 203	79.90 213	87.75 207	91.84 212	82.80 214	82.99 176	82.65 218	60.32 220	78.90 157	70.50 197	67.10 224	92.05 188	90.89 195	98.44 170	91.80 211

anonymousdsp		88.90 151	91.00 133	86.44 194	88.74 201	95.97 135	90.40 191	82.86 177	88.77 174	67.33 208	81.18 149	81.44 126	90.22 171	96.23 98	94.27 130	99.12 98	99.16 40

TESTMET0.1,1		91.07 122	93.56 95	88.17 158	90.43 153	96.57 121	92.02 171	82.83 178	92.34 132	75.05 159	90.20 85	88.64 87	90.93 142	96.19 101	94.07 133	97.75 187	96.90 170

v18		87.93 168	87.61 176	88.31 155	89.74 162	92.04 203	92.59 142	82.71 179	89.70 158	75.32 154	75.23 167	73.55 167	90.74 149	92.11 180	92.77 168	98.78 142	97.87 131

test-mter		90.95 123	93.54 97	87.93 169	90.28 157	96.80 112	91.44 178	82.68 180	92.15 136	74.37 167	89.57 91	88.23 90	90.88 145	96.37 93	94.31 129	97.93 184	97.37 152

v17		87.83 174	87.56 178	88.13 159	89.65 165	92.02 204	92.34 150	82.55 181	89.38 163	74.76 162	75.14 168	73.59 166	90.70 152	92.15 175	92.78 166	98.78 142	97.89 129

v16		87.87 173	87.60 177	88.19 157	89.70 163	92.01 205	92.37 146	82.54 182	89.67 160	75.00 161	75.02 171	73.65 165	90.73 151	92.14 176	92.80 162	98.77 146	97.90 128

MIMVSNet1		80.03 215	80.93 215	78.97 214	72.46 232	90.73 215	80.81 218	82.44 183	80.39 219	63.64 215	57.57 225	64.93 218	76.37 214	91.66 190	91.55 193	98.07 181	89.70 218

PM-MVS		84.72 204	84.47 209	85.03 201	84.67 215	91.57 213	86.27 208	82.31 184	87.65 191	70.62 196	76.54 164	56.41 228	88.75 180	92.59 163	89.85 202	97.54 189	96.66 175

v15		87.46 182	87.16 185	87.81 170	89.41 179	91.96 206	92.26 158	82.28 185	88.42 178	73.72 174	74.29 184	72.73 182	90.41 165	92.17 174	92.76 172	98.79 135	97.83 134

V14		87.47 181	87.19 184	87.80 171	89.37 181	91.95 207	92.25 160	82.12 186	88.39 179	73.83 172	74.31 182	72.84 175	90.44 161	92.20 172	92.78 166	98.80 129	97.84 133

V9		87.41 183	87.15 186	87.72 174	89.33 183	91.93 208	92.23 162	82.02 187	88.35 181	73.59 177	74.13 186	72.77 177	90.37 167	92.21 171	92.80 162	98.79 135	97.86 132

v12		87.38 185	87.13 187	87.68 175	89.30 184	91.92 209	92.01 173	81.94 188	88.35 181	73.69 175	74.10 188	72.57 186	90.33 170	92.23 169	92.82 160	98.80 129	97.91 127

v13		87.34 186	87.11 189	87.62 178	89.30 184	91.91 210	92.04 169	81.86 189	88.35 181	73.36 180	73.88 191	72.69 184	90.34 169	92.23 169	92.82 160	98.80 129	97.88 130

v11		87.58 177	87.50 179	87.67 176	89.34 182	91.91 210	92.22 164	81.63 190	89.01 169	72.95 183	74.11 187	72.51 187	91.08 139	94.01 143	93.00 154	98.77 146	97.93 126

EG-PatchMatch MVS		86.68 191	87.24 182	86.02 199	90.58 152	96.26 129	91.08 185	81.59 191	84.96 208	69.80 203	71.35 208	75.08 158	84.23 204	94.24 140	93.35 148	98.82 122	95.46 189

IterMVS		90.20 132	92.43 111	87.61 179	92.82 134	94.31 194	94.11 121	81.54 192	92.97 114	69.90 201	84.71 129	88.16 91	89.96 173	95.25 121	94.17 131	97.31 190	97.46 149
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
Fast-Effi-MVS+-dtu		91.19 121	93.64 91	88.33 154	92.19 140	96.46 124	93.99 123	81.52 193	92.59 124	71.82 188	92.17 65	85.54 101	91.68 131	95.73 113	94.64 118	98.80 129	98.34 106

Anonymous20231211		75.89 218	74.18 223	77.88 219	81.42 221	87.72 220	79.33 222	81.05 194	66.49 232	60.00 223	45.74 231	51.46 231	71.22 222	85.70 218	86.91 218	94.25 221	95.25 191

MDTV_nov1_ep13		91.57 116	93.18 100	89.70 133	93.39 127	96.97 105	93.53 127	80.91 195	95.70 72	81.86 119	92.40 62	89.93 78	93.25 110	91.97 189	90.80 196	95.25 212	94.46 195

testus		81.33 212	84.13 210	78.06 216	84.54 216	87.72 220	79.66 219	80.42 196	87.36 195	54.13 231	83.83 135	56.63 226	73.21 221	90.51 198	91.74 191	96.40 196	91.11 213

new-patchmatchnet		78.49 217	78.19 218	78.84 215	84.13 219	90.06 216	77.11 225	80.39 197	79.57 221	59.64 224	66.01 218	55.65 229	75.62 215	84.55 222	80.70 225	96.14 200	90.77 216

FPMVS		75.84 219	74.59 219	77.29 220	86.92 212	83.89 226	85.01 210	80.05 198	82.91 216	60.61 219	65.25 219	60.41 221	63.86 225	75.60 228	73.60 230	87.29 231	80.47 227

test2356		81.26 213	84.10 211	77.95 218	84.35 217	87.38 222	79.56 220	79.53 199	86.17 203	54.14 230	83.24 138	60.71 220	73.77 217	90.01 203	91.18 194	96.33 197	90.01 217

tpmp4_e23		89.82 137	89.31 148	90.42 123	94.01 119	95.45 155	94.63 115	78.37 200	93.59 105	87.09 98	86.62 115	76.59 148	93.06 113	88.50 207	88.52 206	95.36 208	95.88 180

new_pmnet		81.53 211	82.68 214	80.20 211	83.47 220	89.47 218	82.21 217	78.36 201	87.86 188	60.14 222	67.90 216	69.43 205	82.03 209	89.22 206	87.47 212	94.99 214	87.39 221

gm-plane-assit		83.26 209	85.29 204	80.89 210	89.52 172	89.89 217	70.26 226	78.24 202	77.11 223	58.01 225	74.16 185	66.90 213	90.63 157	97.20 53	96.05 77	98.66 157	95.68 185

CostFormer		90.69 125	90.48 139	90.93 115	94.18 115	96.08 132	94.03 122	78.20 203	93.47 107	89.96 65	90.97 79	80.30 129	93.72 100	87.66 213	88.75 205	95.51 207	96.12 178

tpm cat1		88.90 151	87.78 170	90.22 126	93.88 122	95.39 164	93.79 125	78.11 204	92.55 125	89.43 73	81.31 148	79.84 131	91.40 133	84.95 219	86.34 221	94.68 219	94.09 199

dps		90.11 135	89.37 147	90.98 114	93.89 121	96.21 130	93.49 128	77.61 205	91.95 137	92.74 42	88.85 94	78.77 135	92.37 120	87.71 212	87.71 211	95.80 202	94.38 196

PMVS	63.12 18	67.27 225	66.39 227	68.30 225	77.98 227	60.24 236	59.53 235	76.82 206	66.65 231	60.74 218	54.39 228	59.82 222	51.24 230	73.92 231	70.52 231	83.48 233	79.17 230
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
EPMVS		90.88 124	92.12 117	89.44 137	94.71 108	97.24 101	93.55 126	76.81 207	95.89 65	81.77 120	91.49 72	86.47 94	93.87 96	90.21 200	90.07 201	95.92 201	93.49 206

MDTV_nov1_ep13_2view		86.30 196	88.27 155	84.01 203	87.71 208	94.67 186	88.08 201	76.78 208	90.59 153	68.66 207	80.46 153	80.12 130	87.58 187	89.95 204	88.20 208	95.25 212	93.90 202

testmv		72.66 221	74.40 220	70.62 222	80.64 222	81.51 229	64.99 231	76.60 209	68.76 228	44.81 233	63.78 221	48.00 232	62.52 226	84.74 220	87.17 213	94.19 222	86.86 222

test1235678		72.65 222	74.40 220	70.62 222	80.64 222	81.50 230	64.99 231	76.59 210	68.74 229	44.81 233	63.78 221	47.99 233	62.51 227	84.73 221	87.17 213	94.19 222	86.85 223

PatchmatchNet		90.56 127	92.49 108	88.31 155	93.83 123	96.86 111	92.42 145	76.50 211	95.96 63	78.31 135	91.96 68	89.66 80	93.48 105	90.04 202	89.20 204	95.32 209	93.73 204
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
tpmrst		88.86 153	89.62 143	87.97 168	94.33 113	95.98 134	92.62 141	76.36 212	94.62 90	76.94 141	85.98 123	82.80 121	92.80 116	86.90 214	87.15 215	94.77 216	93.93 201

tpm		87.95 167	89.44 146	86.21 196	92.53 136	94.62 189	91.40 179	76.36 212	91.46 140	69.80 203	87.43 104	75.14 156	91.55 132	89.85 205	90.60 198	95.61 205	96.96 167

CR-MVSNet		90.16 134	91.96 120	88.06 163	93.32 128	95.95 137	93.36 130	75.99 214	92.40 129	75.19 156	83.18 139	85.37 102	92.05 123	95.21 122	94.56 122	98.47 169	97.08 162

Patchmtry						95.96 136	93.36 130	75.99 214		75.19 156

test12356		69.55 223	71.53 225	67.24 226	77.70 228	78.48 231	65.92 228	75.55 216	68.39 230	44.26 235	61.80 224	40.70 235	47.92 234	81.45 226	87.01 217	92.09 227	82.89 225

MIMVSNet		88.99 150	91.07 132	86.57 191	86.78 213	95.62 147	91.20 184	75.40 217	90.65 151	76.57 143	84.05 133	82.44 123	91.01 141	95.84 109	95.38 103	98.48 168	93.50 205

PatchT		89.13 148	91.71 122	86.11 198	92.92 132	95.59 150	83.64 212	75.09 218	91.87 138	75.19 156	82.63 143	85.06 107	92.05 123	95.21 122	94.56 122	97.76 186	97.08 162

ADS-MVSNet		89.80 138	91.33 130	88.00 167	94.43 112	96.71 117	92.29 155	74.95 219	96.07 61	77.39 138	88.67 97	86.09 97	93.26 109	88.44 208	89.57 203	95.68 204	93.81 203

RPMNet		90.19 133	92.03 119	88.05 164	93.46 125	95.95 137	93.41 129	74.59 220	92.40 129	75.91 149	84.22 132	86.41 95	92.49 118	94.42 135	93.85 140	98.44 170	96.96 167

pmmvs3		79.16 216	80.12 217	78.05 217	79.36 224	86.59 224	78.13 224	73.87 221	76.42 224	57.51 226	70.59 212	57.02 225	84.66 202	90.10 201	88.32 207	94.75 217	91.77 212

LP		84.43 205	85.10 206	83.66 204	92.31 139	93.89 196	87.13 203	72.88 222	90.81 148	67.08 209	70.65 210	75.76 154	86.87 189	86.43 217	87.15 215	95.70 203	90.98 214

MVS-HIRNet		85.36 201	86.89 191	83.57 205	90.13 159	94.51 190	83.57 213	72.61 223	88.27 186	71.22 192	68.97 213	81.81 124	88.91 179	93.08 157	91.94 189	94.97 215	89.64 219

no-one		55.96 229	55.63 230	56.35 230	68.48 233	73.29 234	43.03 236	72.52 224	44.01 236	34.80 236	32.83 233	29.11 237	35.21 235	56.63 233	75.72 228	84.04 232	77.79 231

gg-mvs-nofinetune		86.17 197	88.57 153	83.36 206	93.44 126	98.15 89	96.58 73	72.05 225	74.12 225	49.23 232	64.81 220	90.85 73	89.90 174	97.83 40	96.84 55	98.97 112	97.41 151

Gipuma		68.35 224	66.71 226	70.27 224	74.16 231	68.78 235	63.93 234	71.77 226	83.34 215	54.57 229	34.37 232	31.88 236	68.69 223	83.30 224	85.53 222	88.48 230	79.78 228
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
DeepMVS_CX						86.86 223	79.50 221	70.43 227	90.73 149	63.66 214	80.36 154	60.83 219	79.68 211	76.23 227		89.46 229	86.53 224

E-PMN		50.67 230	47.85 232	53.96 231	64.13 236	50.98 239	38.06 237	69.51 228	51.40 235	24.60 239	29.46 236	24.39 239	56.07 229	48.17 234	59.70 232	71.40 235	70.84 233

EMVS		49.98 231	46.76 233	53.74 232	64.96 235	51.29 238	37.81 238	69.35 229	51.83 234	22.69 240	29.57 235	25.06 238	57.28 228	44.81 235	56.11 233	70.32 236	68.64 234

PMMVS2		64.36 227	65.94 228	62.52 228	67.37 234	77.44 232	64.39 233	69.32 230	61.47 233	34.59 237	46.09 230	41.03 234	48.02 233	74.56 230	78.23 226	91.43 228	82.76 226

1111		73.35 220	74.40 220	72.12 221	78.22 225	82.24 227	65.06 229	65.61 231	70.28 226	55.42 227	56.30 226	57.35 223	73.66 218	86.73 215	88.16 209	94.75 217	79.76 229

.test1245		56.65 228	56.09 229	57.30 229	78.22 225	82.24 227	65.06 229	65.61 231	70.28 226	55.42 227	56.30 226	57.35 223	73.66 218	86.73 215	15.01 234	5.84 238	24.75 235

testpf		83.57 208	85.70 201	81.08 209	90.99 149	88.96 219	82.71 215	65.32 233	90.22 157	73.86 171	81.58 147	76.10 151	81.19 210	84.14 223	85.41 223	92.43 226	93.45 207

MVE	50.86 19	49.54 232	51.43 231	47.33 233	44.14 237	59.20 237	36.45 239	60.59 234	41.47 237	31.14 238	29.58 234	17.06 241	48.52 232	62.22 232	74.63 229	63.12 237	75.87 232
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
tmp_tt				66.88 227	86.07 214	73.86 233	68.22 227	33.38 235	96.88 41	80.67 127	88.23 100	78.82 134	49.78 231	82.68 225	77.47 227	83.19 234

testmvs		12.09 233	16.94 234	6.42 235	3.15 238	6.08 240	9.51 241	3.84 236	21.46 238	5.31 241	27.49 237	6.76 242	10.89 236	17.06 236	15.01 234	5.84 238	24.75 235

test123		9.58 234	13.53 235	4.97 236	1.31 240	5.47 241	8.32 242	2.95 237	18.14 239	2.03 243	20.82 238	2.34 243	10.60 237	10.00 237	14.16 236	4.60 240	23.77 237

GG-mvs-BLEND		66.17 226	94.91 68	32.63 234	1.32 239	96.64 119	91.40 179	0.85 238	94.39 94	2.20 242	90.15 87	95.70 53	2.27 238	96.39 90	95.44 102	97.78 185	95.68 185

sosnet-low-res		0.00 235	0.00 236	0.00 237	0.00 241	0.00 242	0.00 243	0.00 239	0.00 240	0.00 244	0.00 239	0.00 244	0.00 239	0.00 238	0.00 237	0.00 241	0.00 238

sosnet		0.00 235	0.00 236	0.00 237	0.00 241	0.00 242	0.00 243	0.00 239	0.00 240	0.00 244	0.00 239	0.00 244	0.00 239	0.00 238	0.00 237	0.00 241	0.00 238

ambc			73.83 224		76.23 230	85.13 225	82.27 216		84.16 212	65.58 212	52.82 229	23.31 240	73.55 220	91.41 192	85.26 224	92.97 225	94.70 192

MTAPA										96.83 5		99.12 15

MTMP										97.18 3		98.83 21

Patchmatch-RL test							34.61 240

XVS					96.60 62	99.35 10	96.82 59			90.85 51		98.72 24				99.46 26

X-MVStestdata					96.60 62	99.35 10	96.82 59			90.85 51		98.72 24				99.46 26

mPP-MVS					99.21 20							98.29 32

NP-MVS									95.32 78

High-resolution multi-view benchmark