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.

WR-MVS		99.22 4	99.15 5	99.30 2	99.54 11	99.62 1	99.63 4	99.45 2	97.75 17	98.47 25	99.71 8	99.05 43	98.88 7	99.54 6	99.49 3	99.81 1	98.87 10

Anonymous20231211		99.36 1	99.64 1	99.03 9	99.22 34	99.53 6	99.38 15	99.55 1	99.70 1	98.74 19	99.74 6	99.96 1	97.48 71	99.75 1	99.63 1	99.80 2	99.19 3

PS-CasMVS		99.08 5	98.90 13	99.28 3	99.65 3	99.56 4	99.59 6	99.39 4	96.36 35	98.83 16	99.46 38	99.09 35	98.62 14	99.51 7	99.36 8	99.63 3	98.97 7

CP-MVSNet		98.91 14	98.61 21	99.25 4	99.63 5	99.50 7	99.55 10	99.36 5	95.53 68	98.77 18	99.11 58	98.64 89	98.57 17	99.42 11	99.28 11	99.61 4	98.78 15

PEN-MVS		99.08 5	98.95 10	99.23 5	99.65 3	99.59 2	99.64 2	99.34 6	96.68 28	98.65 20	99.43 41	99.33 17	98.47 21	99.50 8	99.32 9	99.60 5	98.79 12

pmmvs6		98.77 16	99.35 3	98.09 49	98.32 95	98.92 20	98.57 80	99.03 12	99.36 2	96.86 95	99.77 5	99.86 2	96.20 110	99.56 5	99.39 7	99.59 6	98.61 22

EPP-MVSNet		97.29 93	96.88 99	97.76 84	98.70 73	99.10 12	98.92 50	98.36 39	95.12 84	93.36 192	97.39 113	91.00 186	97.65 59	98.72 37	98.91 23	99.58 7	97.92 51

TranMVSNet+NR-MVSNet		98.45 21	98.22 32	98.72 20	99.32 30	99.06 13	98.99 39	98.89 14	95.52 69	97.53 61	99.42 43	98.83 69	98.01 36	98.55 48	98.34 49	99.57 8	97.80 55

DTE-MVSNet		99.03 7	98.88 14	99.21 6	99.66 2	99.59 2	99.62 5	99.34 6	96.92 25	98.52 22	99.36 48	98.98 50	98.57 17	99.49 9	99.23 12	99.56 9	98.55 24

UniMVSNet_NR-MVSNet		98.12 37	97.56 64	98.78 17	99.13 46	98.89 22	98.76 63	98.78 19	93.81 131	98.50 23	98.81 70	97.64 128	97.99 38	98.18 65	97.92 75	99.53 10	97.64 61

WR-MVS_H		98.97 12	98.82 16	99.14 8	99.56 9	99.56 4	99.54 11	99.42 3	96.07 42	98.37 27	99.34 49	99.09 35	98.43 22	99.45 10	99.41 6	99.53 10	98.86 11

NR-MVSNet		98.00 44	97.88 43	98.13 45	98.33 92	98.77 33	98.83 58	98.88 15	94.10 121	97.46 66	98.87 66	98.58 95	95.78 117	99.13 24	98.16 61	99.52 12	97.53 68

IS_MVSNet		96.62 121	96.48 116	96.78 128	98.46 84	98.68 51	98.61 78	98.24 51	92.23 156	89.63 215	95.90 145	94.40 169	96.23 108	98.65 44	98.77 29	99.52 12	96.76 107

LTVRE_ROB	97.71 1	99.33 2	99.47 2	99.16 7	99.16 40	99.11 10	99.39 14	99.16 11	99.26 3	99.22 4	99.51 32	99.75 4	98.54 19	99.71 2	99.47 4	99.52 12	99.46 1
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
UniMVSNet (Re)		98.23 27	97.85 45	98.67 22	99.15 41	98.87 23	98.74 72	98.84 17	94.27 120	97.94 48	99.01 60	98.39 103	97.82 48	98.35 59	98.29 53	99.51 15	97.78 56

DU-MVS		98.23 27	97.74 55	98.81 16	99.23 32	98.77 33	98.76 63	98.88 15	94.10 121	98.50 23	98.87 66	98.32 106	97.99 38	98.40 54	98.08 70	99.49 16	97.64 61

UA-Net		98.66 19	98.60 23	98.73 19	99.83 1	99.28 9	98.56 82	99.24 8	96.04 43	97.12 81	98.44 85	98.95 56	98.17 30	99.15 23	99.00 18	99.48 17	99.33 2

TDRefinement		99.00 9	99.13 6	98.86 12	98.99 57	99.05 15	99.58 7	98.29 48	98.96 5	97.96 47	99.40 45	98.67 86	98.87 8	99.60 4	99.46 5	99.46 18	98.74 18

TransMVSNet (Re)		98.23 27	98.72 18	97.66 86	98.22 107	98.73 45	98.66 77	98.03 73	98.60 7	96.40 112	99.60 21	98.24 109	95.26 126	99.19 21	99.05 17	99.36 19	97.64 61

pm-mvs1		98.14 35	98.66 20	97.53 94	97.93 139	98.49 66	98.14 101	98.19 58	97.95 14	96.17 124	99.63 18	98.85 67	95.41 124	98.91 30	98.89 25	99.34 20	97.86 53

anonymousdsp		98.85 15	98.88 14	98.83 15	98.69 76	98.20 79	99.68 1	97.35 135	97.09 24	98.98 12	99.86 1	99.43 11	98.94 6	99.28 16	99.19 13	99.33 21	99.08 5

ACMH+	94.90 8	98.40 24	98.71 19	98.04 60	98.93 59	98.84 25	99.30 19	97.86 85	97.78 16	94.19 174	98.77 73	99.39 14	98.61 15	99.33 13	99.07 14	99.33 21	97.81 54

SixPastTwentyTwo		99.25 3	99.20 4	99.32 1	99.53 14	99.32 8	99.64 2	99.19 10	98.05 13	99.19 5	99.74 6	98.96 55	99.03 5	99.69 3	99.58 2	99.32 23	99.06 6

CSCG		98.45 21	98.61 21	98.26 38	99.11 48	99.06 13	98.17 100	97.49 112	97.93 15	97.37 70	98.88 64	99.29 18	98.10 31	98.40 54	97.51 83	99.32 23	99.16 4

COLMAP_ROB	96.84 2	98.75 17	98.82 16	98.66 23	99.14 44	98.79 31	99.30 19	97.67 97	98.33 8	97.82 50	99.20 55	99.18 32	98.76 9	99.27 17	98.96 21	99.29 25	98.03 46
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
Baseline_NR-MVSNet		98.17 32	97.90 42	98.48 30	99.23 32	98.59 56	98.83 58	98.73 21	93.97 128	96.95 89	99.66 12	98.23 111	97.90 44	98.40 54	99.06 16	99.25 26	97.42 76

LGP-MVS_train		97.96 51	97.53 65	98.45 32	99.45 22	98.64 53	99.09 29	98.27 49	92.99 149	96.04 128	96.57 130	99.29 18	98.66 12	98.73 35	98.42 44	99.19 27	98.09 44

ACMP	94.03 12	97.97 50	97.61 59	98.39 34	99.43 24	98.51 64	98.97 43	98.06 70	94.63 99	96.10 126	96.12 139	99.20 30	98.63 13	98.68 41	98.20 59	99.14 28	97.93 50
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
ACMM	94.29 11	98.12 37	97.71 57	98.59 25	99.51 16	98.58 57	99.24 21	98.25 50	96.22 40	96.90 90	95.01 159	98.89 61	98.52 20	98.66 43	98.32 52	99.13 29	98.28 40
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
Vis-MVSNet (Re-imp)		96.29 126	96.50 114	96.05 156	97.96 138	97.83 124	97.30 153	97.86 85	93.14 144	88.90 219	96.80 125	95.28 160	95.15 129	98.37 58	98.25 55	99.12 30	95.84 128

SD-MVS		97.84 56	97.78 51	97.90 67	98.33 92	98.06 95	97.95 112	97.80 90	96.03 45	96.72 97	97.57 108	99.18 32	97.50 70	97.88 68	97.08 97	99.11 31	98.68 20

LS3D		97.93 53	97.80 47	98.08 54	99.20 37	98.77 33	98.89 54	97.92 78	96.59 30	96.99 87	96.71 127	97.14 139	96.39 106	99.04 25	98.96 21	99.10 32	97.39 77

CP-MVS		98.00 44	97.57 62	98.50 28	99.47 20	98.56 60	98.91 51	98.38 37	94.71 95	97.01 86	95.20 155	99.06 40	98.20 28	98.61 46	98.46 41	99.02 33	98.40 35

FMVSNet1		97.40 84	98.09 35	96.60 136	97.80 153	98.76 38	98.26 97	98.50 25	96.79 27	93.13 196	99.28 52	98.64 89	92.90 162	97.67 77	97.86 78	99.02 33	97.64 61

v748		98.92 13	98.95 10	98.87 11	98.54 80	98.69 49	99.33 17	98.64 22	98.07 12	99.06 9	99.66 12	99.76 3	98.68 11	99.25 18	98.72 32	99.01 35	98.54 25

ACMMPR		98.31 25	98.07 37	98.60 24	99.58 6	98.83 26	99.09 29	98.48 26	96.25 38	97.03 85	96.81 124	99.09 35	98.39 24	98.55 48	98.45 42	99.01 35	98.53 28

TSAR-MVS + MP.		98.15 34	98.23 31	98.06 58	98.47 83	98.16 85	99.23 22	96.87 150	95.58 63	96.72 97	98.41 86	99.06 40	98.05 34	98.99 27	98.90 24	99.00 37	98.51 29

ACMMP		97.99 46	97.60 60	98.45 32	99.53 14	98.83 26	99.13 28	98.30 44	94.57 101	96.39 116	95.32 153	98.95 56	98.37 25	98.61 46	98.47 39	99.00 37	98.45 32
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
conf0.05thres1000		95.91 136	94.67 152	97.37 100	98.54 80	98.73 45	98.41 90	98.07 69	96.10 41	94.93 160	92.83 187	80.67 212	95.26 126	98.68 41	98.65 35	98.99 39	97.02 93

ACMH	95.26 7	98.75 17	98.93 12	98.54 27	98.86 64	99.01 18	99.58 7	98.10 67	98.67 6	97.30 73	99.18 56	99.42 12	98.40 23	99.19 21	98.86 26	98.99 39	98.19 42
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
PMVS	90.51 17	97.77 59	97.98 40	97.53 94	98.68 77	98.14 89	97.67 125	97.03 145	96.43 31	98.38 26	98.72 76	97.03 142	94.44 142	99.37 12	99.30 10	98.98 41	96.86 102
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
SteuartSystems-ACMMP		98.06 40	97.78 51	98.39 34	99.54 11	98.79 31	98.94 48	98.42 34	93.98 127	95.85 133	96.66 129	99.25 25	98.61 15	98.71 39	98.38 46	98.97 42	98.67 21
Skip Steuart: Steuart Systems R&D Blog.
DeepC-MVS	96.08 5	98.58 20	98.49 25	98.68 21	99.37 26	98.52 63	99.01 37	98.17 62	97.17 23	98.25 31	99.56 25	99.62 5	98.29 26	98.40 54	98.09 63	98.97 42	98.08 45
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
FC-MVSNet-train		97.65 63	98.16 33	97.05 116	98.85 65	98.85 24	99.34 16	98.08 68	94.50 109	94.41 168	99.21 54	98.80 73	92.66 164	98.98 28	98.85 27	98.96 44	97.94 49

X-MVS		97.60 65	97.00 91	98.29 37	99.50 17	98.76 38	98.90 52	98.37 38	94.67 98	96.40 112	91.47 197	98.78 75	97.60 65	98.55 48	98.50 38	98.96 44	98.29 37

HFP-MVS		98.17 32	98.02 38	98.35 36	99.36 27	98.62 54	98.79 60	98.46 31	96.24 39	96.53 105	97.13 121	98.98 50	98.02 35	98.20 62	98.42 44	98.95 46	98.54 25

XVS					99.48 18	98.76 38	99.22 24			96.40 112		98.78 75				98.94 47

X-MVStestdata					99.48 18	98.76 38	99.22 24			96.40 112		98.78 75				98.94 47

zzz-MVS		98.14 35	97.78 51	98.55 26	99.58 6	98.58 57	98.98 41	98.48 26	95.98 46	97.39 68	94.73 163	99.27 22	97.98 40	98.81 32	98.64 36	98.90 49	98.46 31

canonicalmvs		97.11 100	96.88 99	97.38 99	98.34 91	98.72 47	97.52 136	97.94 77	95.60 61	95.01 158	94.58 166	94.50 168	96.59 97	97.84 69	98.03 72	98.90 49	98.91 8

MP-MVS		97.98 47	97.53 65	98.50 28	99.56 9	98.58 57	98.97 43	98.39 36	93.49 136	97.14 78	96.08 140	99.23 28	98.06 33	98.50 51	98.38 46	98.90 49	98.44 33

PGM-MVS		97.82 58	97.25 71	98.48 30	99.54 11	98.75 42	99.02 33	98.35 41	92.41 154	96.84 96	95.39 152	98.99 48	98.24 27	98.43 52	98.34 49	98.90 49	98.41 34

RPSCF		97.83 57	98.27 29	97.31 104	98.23 104	98.06 95	97.44 146	95.79 182	96.90 26	95.81 135	98.76 74	98.61 93	97.70 55	98.90 31	98.36 48	98.90 49	98.29 37

Gipuma		98.43 23	98.15 34	98.76 18	99.00 56	98.29 76	97.91 115	98.06 70	99.02 4	99.50 1	96.33 134	98.67 86	99.22 1	99.02 26	98.02 73	98.88 54	97.66 60
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
tfpnnormal		97.66 62	97.79 48	97.52 96	98.32 95	98.53 62	98.45 87	97.69 96	97.59 19	96.12 125	97.79 104	96.70 144	95.69 120	98.35 59	98.34 49	98.85 55	97.22 89

APDe-MVS		98.29 26	98.42 26	98.14 44	99.45 22	98.90 21	99.18 26	98.30 44	95.96 48	95.13 153	98.79 71	99.25 25	97.92 43	98.80 33	98.71 33	98.85 55	98.54 25

SMA-MVS		98.22 30	98.31 28	98.11 47	99.46 21	98.77 33	98.34 92	97.92 78	95.27 79	96.97 88	98.82 69	99.39 14	97.10 84	98.69 40	98.47 39	98.84 57	98.77 16

V4		98.98 10	99.10 7	98.85 13	98.91 60	99.03 16	99.41 12	97.77 93	98.12 10	99.06 9	99.85 2	99.60 6	99.15 2	99.30 14	98.99 19	98.80 58	98.79 12

APD-MVS		97.47 78	97.16 78	97.84 75	99.32 30	98.39 72	98.47 86	98.21 55	92.08 159	95.23 150	96.68 128	98.90 60	96.99 87	98.20 62	98.21 56	98.80 58	97.67 59
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
v52		98.98 10	99.10 7	98.85 13	98.91 60	99.03 16	99.41 12	97.77 93	98.12 10	99.07 8	99.84 3	99.60 6	99.15 2	99.29 15	98.99 19	98.79 60	98.79 12

PHI-MVS		97.44 80	97.17 77	97.74 85	98.14 117	98.41 71	98.03 106	97.50 109	92.07 160	98.01 45	97.33 115	98.62 92	96.02 114	98.34 61	98.21 56	98.76 61	97.24 86

tfpn_n400		95.11 151	93.86 162	96.57 138	98.16 115	97.92 113	97.59 132	97.90 80	95.90 51	92.83 201	89.94 206	83.01 203	94.23 148	97.50 85	97.43 86	98.73 62	95.30 144

tfpnconf		95.11 151	93.86 162	96.57 138	98.16 115	97.92 113	97.59 132	97.90 80	95.90 51	92.83 201	89.94 206	83.01 203	94.23 148	97.50 85	97.43 86	98.73 62	95.30 144

FC-MVSNet-test		97.54 68	98.26 30	96.70 131	98.87 63	97.79 129	98.49 84	98.56 23	96.04 43	90.39 209	99.65 14	98.67 86	95.15 129	99.23 19	99.07 14	98.73 62	97.39 77

tfpn1000		94.36 166	93.33 173	95.56 171	98.09 123	98.07 94	97.08 165	97.78 92	94.02 126	89.16 218	91.38 198	80.56 213	92.54 172	96.76 118	98.09 63	98.69 65	94.40 166

ACMMP_Plus		98.12 37	98.08 36	98.18 42	99.34 28	98.74 43	98.97 43	98.00 74	95.13 83	96.90 90	97.54 110	99.27 22	97.18 82	98.72 37	98.45 42	98.68 66	98.69 19

tfpn_ndepth		93.27 184	92.11 187	94.61 186	96.96 187	97.93 112	96.87 170	97.49 112	90.91 173	87.89 226	85.98 216	83.53 200	89.77 194	95.91 152	97.31 93	98.67 67	93.25 178

v7n		99.03 7	99.03 9	99.02 10	99.09 51	99.11 10	99.57 9	98.82 18	98.21 9	99.25 2	99.84 3	99.59 8	98.76 9	99.23 19	98.83 28	98.63 68	98.40 35

MIMVSNet1		98.22 30	98.51 24	97.87 73	99.40 25	98.82 28	99.31 18	98.53 24	97.39 20	96.59 103	99.31 51	99.23 28	94.76 136	98.93 29	98.67 34	98.63 68	97.25 84

Vis-MVSNet		98.01 42	98.42 26	97.54 93	96.89 189	98.82 28	99.14 27	97.59 101	96.30 36	97.04 84	99.26 53	98.83 69	96.01 115	98.73 35	98.21 56	98.58 70	98.75 17
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
tfpn		92.86 186	89.37 203	96.93 121	98.40 87	98.34 74	98.02 108	97.80 90	92.54 152	93.99 177	86.54 215	57.58 236	94.82 134	97.66 80	97.99 74	98.56 71	94.95 153

ESAPD		97.71 61	97.79 48	97.62 87	99.21 35	98.80 30	98.31 95	98.30 44	93.60 134	94.74 162	97.94 99	99.24 27	96.58 98	98.42 53	98.27 54	98.56 71	98.28 40

CDPH-MVS		96.68 117	95.99 126	97.48 97	99.13 46	97.64 132	98.08 103	97.46 117	90.56 177	95.13 153	94.87 161	98.27 108	96.56 100	97.09 101	96.45 123	98.54 73	97.08 91

CPTT-MVS		97.08 102	96.25 118	98.05 59	99.21 35	98.30 75	98.54 83	97.98 75	94.28 118	95.89 132	89.57 209	98.54 97	98.18 29	97.82 70	97.32 91	98.54 73	97.91 52

3Dnovator+	96.20 4	97.58 66	97.14 80	98.10 48	98.98 58	97.85 123	98.60 79	98.33 42	96.41 33	97.23 77	94.66 165	97.26 135	96.91 89	97.91 67	97.87 77	98.53 75	98.03 46

tfpn111		93.73 180	91.63 192	96.17 152	97.52 164	98.15 86	97.48 139	97.48 114	87.65 201	93.42 188	82.19 228	84.12 195	92.62 165	97.04 103	98.09 63	98.52 76	94.17 167

conf200view11		93.79 179	91.75 190	96.17 152	97.52 164	98.15 86	97.48 139	97.48 114	87.65 201	93.42 188	83.03 225	84.12 195	92.62 165	97.04 103	98.09 63	98.52 76	94.17 167

tfpn200view9		93.80 178	91.75 190	96.20 150	97.52 164	98.15 86	97.48 139	97.47 116	87.65 201	93.56 186	83.03 225	84.12 195	92.62 165	97.04 103	98.09 63	98.52 76	94.17 167

tfpnview11		94.92 156	93.56 167	96.50 142	98.12 121	97.99 103	97.48 139	97.86 85	94.50 109	92.83 201	89.94 206	83.01 203	94.19 150	96.91 114	98.07 71	98.50 79	94.53 159

view800		94.54 162	92.55 178	96.86 126	98.28 100	98.22 78	97.97 111	97.62 100	92.10 158	94.19 174	85.52 218	81.33 211	94.61 138	97.41 88	98.51 37	98.50 79	94.72 156

thres600view7		94.34 168	92.31 184	96.70 131	98.19 110	98.12 90	97.85 121	97.45 122	91.49 165	93.98 178	84.27 221	82.02 209	94.24 146	97.04 103	98.76 30	98.49 81	94.47 162

thres200		93.98 177	91.90 189	96.40 147	97.66 157	98.12 90	97.20 160	97.45 122	90.16 182	93.82 179	83.08 224	83.74 199	93.80 155	97.04 103	97.48 85	98.49 81	93.70 174

conf0.01		91.86 195	88.22 206	96.10 154	97.40 173	97.94 110	97.48 139	97.41 129	87.65 201	93.22 194	80.39 230	63.83 232	92.62 165	96.63 124	98.09 63	98.47 83	93.03 183

conf0.002		91.12 203	86.87 217	96.08 155	97.35 176	97.89 119	97.48 139	97.38 131	87.65 201	93.19 195	79.38 232	57.48 237	92.62 165	96.56 126	96.64 115	98.46 84	92.50 186

Effi-MVS+-dtu		95.94 135	95.08 144	96.94 120	98.54 80	97.38 144	96.66 176	97.89 83	88.68 190	95.92 130	92.90 186	97.28 134	94.18 151	96.68 122	96.13 138	98.45 85	96.51 116

MCST-MVS		96.79 113	96.08 123	97.62 87	98.78 71	97.52 141	98.01 109	97.32 136	93.20 142	95.84 134	93.97 176	98.12 113	97.34 78	96.34 133	95.88 150	98.45 85	97.51 71

TAPA-MVS	93.96 13	96.79 113	96.70 111	96.90 124	97.64 158	97.58 135	97.54 135	94.50 209	95.14 82	96.64 102	96.76 126	97.90 120	96.63 94	95.98 150	96.14 136	98.45 85	97.39 77
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
view600		94.36 166	92.33 183	96.73 129	98.14 117	98.03 99	97.88 118	97.36 134	91.61 162	94.29 171	84.38 220	82.08 208	94.31 145	97.05 102	98.75 31	98.42 88	94.41 164

OPM-MVS		98.01 42	98.01 39	98.00 63	99.11 48	98.12 90	98.68 76	97.72 95	96.65 29	96.68 101	98.40 87	99.28 21	97.44 73	98.20 62	97.82 81	98.40 89	97.58 66

thres400		94.04 175	91.94 188	96.50 142	97.98 137	97.82 126	97.66 127	96.96 146	90.96 171	94.20 172	83.24 223	82.82 206	93.80 155	96.50 127	98.09 63	98.38 90	94.15 170

CLD-MVS		96.73 116	96.92 95	96.51 141	98.70 73	97.57 137	97.64 128	92.07 216	93.10 147	96.31 117	98.29 89	99.02 46	95.99 116	97.20 96	96.47 122	98.37 91	96.81 106
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
GBi-Net		95.21 149	95.35 136	95.04 179	96.77 192	98.18 81	97.28 154	97.58 102	88.43 195	90.28 211	96.01 141	92.43 178	90.04 190	97.67 77	97.86 78	98.28 92	96.90 97

test1		95.21 149	95.35 136	95.04 179	96.77 192	98.18 81	97.28 154	97.58 102	88.43 195	90.28 211	96.01 141	92.43 178	90.04 190	97.67 77	97.86 78	98.28 92	96.90 97

FMVSNet2		95.77 138	96.20 121	95.27 176	96.77 192	98.18 81	97.28 154	97.90 80	93.12 145	91.37 206	98.25 91	96.05 155	90.04 190	94.96 174	95.94 147	98.28 92	96.90 97

OMC-MVS		97.23 96	97.21 73	97.25 108	97.85 144	97.52 141	97.92 114	95.77 183	95.83 55	97.09 83	97.86 102	98.52 98	96.62 95	97.51 83	96.65 114	98.26 95	96.57 112

3Dnovator	96.31 3	97.22 97	97.19 75	97.25 108	98.14 117	97.95 107	98.03 106	96.77 156	96.42 32	97.14 78	95.11 156	97.59 129	95.14 131	97.79 71	97.72 82	98.26 95	97.76 58

PVSNet_Blended_VisFu		97.44 80	97.14 80	97.79 78	99.15 41	98.44 69	98.32 94	97.66 98	93.74 133	97.73 52	98.79 71	96.93 143	95.64 123	97.69 75	96.91 103	98.25 97	97.50 72

DeepC-MVS_fast	95.38 6	97.53 71	97.30 69	97.79 78	98.83 69	97.64 132	98.18 98	97.14 141	95.57 64	97.83 49	97.10 122	98.80 73	96.53 102	97.41 88	97.32 91	98.24 98	97.26 83
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
thresconf0.02		91.75 197	88.21 207	95.87 160	97.38 174	97.14 153	97.27 157	96.85 152	93.04 148	92.39 204	82.19 228	63.31 233	93.10 159	94.43 184	95.06 165	98.23 99	92.32 187

HPM-MVS++		97.56 67	97.11 85	98.09 49	99.18 39	97.95 107	98.57 80	98.20 56	94.08 123	97.25 76	95.96 144	98.81 72	97.13 83	97.51 83	97.30 94	98.21 100	98.15 43

NCCC		96.56 122	95.68 132	97.59 89	99.04 54	97.54 140	97.67 125	97.56 105	94.84 92	96.10 126	87.91 212	98.09 114	96.98 88	97.20 96	96.80 109	98.21 100	97.38 80

TSAR-MVS + GP.		97.26 95	97.33 68	97.18 110	98.21 108	98.06 95	96.38 183	97.66 98	93.92 130	95.23 150	98.48 83	98.33 105	97.41 74	97.63 82	97.35 89	98.18 102	97.57 67

thres100view900		92.93 185	90.89 196	95.31 174	97.52 164	96.82 166	96.41 181	95.08 194	87.65 201	93.56 186	83.03 225	84.12 195	91.12 179	94.53 178	96.91 103	98.17 103	93.21 180

MVS_0304		97.18 98	96.84 105	97.58 90	99.15 41	98.19 80	98.11 102	97.81 89	92.36 155	98.06 42	97.43 112	99.06 40	94.24 146	96.80 117	96.54 120	98.12 104	97.52 70

CNVR-MVS		97.03 105	96.77 109	97.34 101	98.89 62	97.67 131	97.64 128	97.17 140	94.40 114	95.70 141	94.02 174	98.76 79	96.49 104	97.78 72	97.29 95	98.12 104	97.47 73

HQP-MVS		95.97 134	95.01 147	97.08 113	98.72 72	97.19 150	97.07 166	96.69 160	91.49 165	95.77 137	92.19 192	97.93 119	96.15 112	94.66 176	94.16 175	98.10 106	97.45 74

PCF-MVS	92.69 14	95.98 133	95.05 145	97.06 115	98.43 86	97.56 138	97.76 122	96.65 162	89.95 185	95.70 141	96.18 138	98.48 101	95.74 118	93.64 192	93.35 188	98.09 107	96.18 121
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
MVS_111021_HR		97.27 94	97.11 85	97.46 98	98.46 84	97.82 126	97.50 137	96.86 151	94.97 88	97.13 80	96.99 123	98.39 103	96.82 91	97.65 81	97.38 88	98.02 108	96.56 114

EG-PatchMatch MVS		97.98 47	97.92 41	98.04 60	98.84 66	98.04 98	97.90 116	96.83 154	95.07 85	98.79 17	99.07 59	99.37 16	97.88 46	98.74 34	98.16 61	98.01 109	96.96 95

DELS-MVS		96.90 107	97.24 72	96.50 142	97.85 144	98.18 81	97.88 118	95.92 175	93.48 137	95.34 148	98.86 68	98.94 59	94.03 152	97.33 92	97.04 98	98.00 110	96.85 104
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
OpenMVS	94.63 9	95.75 139	95.04 146	96.58 137	97.85 144	97.55 139	96.71 175	96.07 170	90.15 183	96.47 107	90.77 205	95.95 156	94.41 143	97.01 109	96.95 100	98.00 110	96.90 97

Effi-MVS+		96.46 123	95.28 138	97.85 74	98.64 78	97.16 151	97.15 164	98.75 20	90.27 180	98.03 44	93.93 177	96.21 151	96.55 101	96.34 133	96.69 113	97.97 112	96.33 119

train_agg		96.68 117	95.93 129	97.56 91	99.08 52	97.16 151	98.44 89	97.37 133	91.12 170	95.18 152	95.43 151	98.48 101	97.36 76	96.48 128	95.52 157	97.95 113	97.34 82

PLC	92.55 15	96.10 129	95.36 135	96.96 118	98.13 120	96.88 162	96.49 180	96.67 161	94.07 124	95.71 140	91.14 200	96.09 154	96.84 90	96.70 121	96.58 119	97.92 114	96.03 125
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
Fast-Effi-MVS+		96.80 112	95.92 130	97.84 75	98.57 79	97.46 143	98.06 104	98.24 51	89.64 187	97.57 60	96.45 132	97.35 133	96.73 92	97.22 95	96.64 115	97.86 115	96.65 110

UGNet		96.79 113	97.82 46	95.58 169	97.57 162	98.39 72	98.48 85	97.84 88	95.85 54	94.68 163	97.91 101	99.07 39	87.12 210	97.71 74	97.51 83	97.80 116	98.29 37
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
CNLPA		96.24 128	95.97 127	96.57 138	97.48 170	97.10 158	96.75 173	94.95 199	94.92 90	96.20 122	94.81 162	96.61 146	96.25 107	96.94 112	95.64 154	97.79 117	95.74 134

QAPM		97.04 104	97.14 80	96.93 121	97.78 156	98.02 100	97.36 151	96.72 157	94.68 97	96.23 119	97.21 119	97.68 126	95.70 119	97.37 90	97.24 96	97.78 118	97.77 57

PVSNet_BlendedMVS		95.44 145	95.09 142	95.86 161	97.31 177	97.13 154	96.31 187	95.01 196	88.55 193	96.23 119	94.55 169	97.75 123	92.56 170	96.42 130	95.44 159	97.71 119	95.81 129

PVSNet_Blended		95.44 145	95.09 142	95.86 161	97.31 177	97.13 154	96.31 187	95.01 196	88.55 193	96.23 119	94.55 169	97.75 123	92.56 170	96.42 130	95.44 159	97.71 119	95.81 129

AdaColmap		95.85 137	94.65 153	97.26 105	98.70 73	97.20 149	97.33 152	97.30 137	91.28 168	95.90 131	88.16 211	96.17 153	96.60 96	97.34 91	96.82 105	97.71 119	95.60 137

test20.03		96.08 130	96.80 107	95.25 178	99.19 38	97.58 135	97.24 159	97.56 105	94.95 89	91.91 205	98.58 80	98.03 117	87.88 206	97.43 87	96.94 101	97.69 122	94.05 171

MVS_111021_LR		96.86 108	96.72 110	97.03 117	97.80 153	97.06 159	97.04 167	95.51 188	94.55 102	97.47 64	97.35 114	97.68 126	96.66 93	97.11 99	96.73 110	97.69 122	96.57 112

MSDG		96.27 127	96.17 122	96.38 148	97.85 144	96.27 178	96.55 179	94.41 210	94.55 102	95.62 143	97.56 109	97.80 122	96.22 109	97.17 98	96.27 129	97.67 124	93.60 175

HSP-MVS		97.44 80	97.13 83	97.79 78	99.34 28	98.99 19	99.23 22	98.12 65	93.43 138	95.95 129	97.45 111	99.50 9	96.44 105	96.35 132	95.33 162	97.65 125	98.89 9

DI_MVS_plusplus_trai		95.48 143	94.51 155	96.61 135	97.13 183	97.30 145	98.05 105	96.79 155	93.75 132	95.08 156	96.38 133	89.76 189	94.95 132	93.97 191	94.82 171	97.64 126	95.63 136

FMVSNet3		94.06 174	93.85 164	94.31 192	95.46 217	97.80 128	96.34 184	97.58 102	88.43 195	90.28 211	96.01 141	92.43 178	88.67 202	91.82 207	93.96 180	97.53 127	96.50 117

MSLP-MVS++		96.66 119	96.46 117	96.89 125	98.02 127	97.71 130	95.57 201	96.96 146	94.36 116	96.19 123	91.37 199	98.24 109	97.07 85	97.69 75	97.89 76	97.52 128	97.95 48

abl_6				96.45 145	97.79 155	97.28 146	97.16 163	96.16 168	89.92 186	95.72 139	91.59 196	97.16 138	94.37 144			97.51 129	95.49 139

v1192		97.52 72	97.03 89	98.09 49	98.31 98	98.01 101	98.96 46	97.25 138	95.22 80	98.89 14	99.64 15	98.83 69	97.68 57	95.63 159	95.91 148	97.47 130	95.97 127

v1144		97.51 73	97.05 87	98.04 60	98.26 102	97.98 104	98.88 55	97.42 125	95.38 74	98.56 21	99.59 24	99.01 47	97.65 59	95.77 157	96.06 142	97.47 130	95.56 138

v144192		97.49 77	96.99 93	98.07 56	98.11 122	97.95 107	99.02 33	97.21 139	94.90 91	98.88 15	99.53 31	98.89 61	97.75 51	95.59 160	95.90 149	97.43 132	96.16 122

v7		97.45 79	97.01 90	97.97 64	98.07 124	97.96 105	98.86 56	97.50 109	94.46 112	98.24 32	99.56 25	98.98 50	97.72 53	96.05 147	96.26 130	97.42 133	95.79 131

v10		97.64 64	97.26 70	98.08 54	98.07 124	98.56 60	98.86 56	98.18 61	94.48 111	98.24 32	99.56 25	98.98 50	97.72 53	96.05 147	96.26 130	97.42 133	96.93 96

CANet		96.81 111	96.50 114	97.17 111	99.10 50	97.96 105	97.86 120	97.51 107	91.30 167	97.75 51	97.64 106	97.89 121	93.39 158	96.98 110	96.73 110	97.40 135	96.99 94

v13		98.04 41	97.86 44	98.24 39	98.36 90	98.77 33	99.04 31	98.47 28	95.93 49	98.20 35	99.67 11	99.11 34	98.00 37	97.11 99	96.93 102	97.40 135	97.53 68

v1921920		97.50 76	97.00 91	98.07 56	98.20 109	97.94 110	99.03 32	97.06 143	95.29 78	99.01 11	99.62 19	98.73 83	97.74 52	95.52 162	95.78 153	97.39 137	96.12 124

v12		97.98 47	97.78 51	98.21 40	98.33 92	98.74 43	99.01 37	98.44 33	95.82 56	98.13 36	99.64 15	99.08 38	97.95 41	96.97 111	96.82 105	97.39 137	97.38 80

IB-MVS	92.44 16	93.33 183	92.15 186	94.70 185	97.42 172	96.39 174	95.57 201	94.67 204	86.40 218	93.59 185	78.28 234	95.76 158	89.59 196	95.88 153	95.98 143	97.39 137	96.34 118
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
V9		97.91 54	97.70 58	98.17 43	98.30 99	98.70 48	98.98 41	98.40 35	95.72 58	98.07 40	99.64 15	99.04 44	97.90 44	96.82 115	96.71 112	97.37 140	97.23 87

v2v482		97.33 89	96.84 105	97.90 67	98.19 110	97.83 124	98.74 72	97.44 124	95.42 73	98.23 34	99.46 38	98.84 68	97.46 72	95.51 163	96.10 140	97.36 141	94.72 156

DeepPCF-MVS	94.55 10	97.05 103	97.13 83	96.95 119	96.06 202	97.12 156	98.01 109	95.44 189	95.18 81	97.50 62	97.86 102	98.08 115	97.31 80	97.23 94	97.00 99	97.36 141	97.45 74

V14		97.85 55	97.60 60	98.13 45	98.27 101	98.66 52	98.94 48	98.36 39	95.62 60	98.04 43	99.62 19	98.99 48	97.84 47	96.65 123	96.59 118	97.34 143	97.07 92

v1240		97.43 83	96.87 104	98.09 49	98.25 103	97.92 113	99.02 33	97.06 143	94.77 94	99.09 7	99.68 10	98.51 99	97.78 49	95.25 167	95.81 151	97.32 144	96.13 123

v15		97.77 59	97.50 67	98.09 49	98.23 104	98.62 54	98.90 52	98.32 43	95.51 71	98.01 45	99.60 21	98.95 56	97.78 49	96.47 129	96.45 123	97.32 144	96.90 97

v1141		97.36 88	96.92 95	97.88 72	98.18 112	97.90 117	98.76 63	97.42 125	95.38 74	98.07 40	99.56 25	98.87 64	97.59 67	95.78 154	95.98 143	97.29 146	94.97 151

divwei89l23v2f112		97.37 86	96.92 95	97.89 69	98.18 112	97.90 117	98.76 63	97.42 125	95.38 74	98.09 38	99.56 25	98.87 64	97.59 67	95.78 154	95.98 143	97.29 146	94.97 151

v1		97.37 86	96.92 95	97.89 69	98.18 112	97.91 116	98.76 63	97.42 125	95.38 74	98.09 38	99.55 30	98.88 63	97.59 67	95.78 154	95.98 143	97.29 146	94.98 150

v11		97.94 52	97.72 56	98.20 41	98.37 89	98.69 49	98.96 46	98.30 44	95.68 59	98.35 28	99.70 9	99.19 31	97.93 42	96.76 118	96.82 105	97.28 149	97.23 87

CDS-MVSNet		94.91 157	95.17 141	94.60 187	97.85 144	96.21 179	96.90 169	96.39 165	90.81 174	93.40 190	97.24 118	94.54 167	85.78 216	96.25 138	96.15 132	97.26 150	95.01 149
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
pmmvs-eth3d		96.84 110	96.22 120	97.56 91	97.63 160	96.38 175	98.74 72	96.91 149	94.63 99	98.26 30	99.43 41	98.28 107	96.58 98	94.52 180	95.54 156	97.24 151	94.75 155

pmmvs5		95.70 140	95.22 139	96.26 149	96.55 197	97.24 147	97.50 137	94.99 198	90.95 172	96.87 92	98.47 84	97.40 131	94.45 141	92.86 200	94.98 167	97.23 152	94.64 158

v6		97.30 90	96.88 99	97.78 81	97.99 131	97.87 120	98.75 69	97.46 117	94.54 105	97.61 58	99.48 34	98.77 78	97.65 59	96.09 144	96.15 132	97.21 153	95.28 146

Anonymous20231206		95.69 141	95.68 132	95.70 165	98.32 95	96.95 160	97.37 149	96.65 162	93.33 139	93.61 184	98.70 78	98.03 117	91.04 180	95.07 170	94.59 174	97.20 154	93.09 182

v1neww		97.30 90	96.88 99	97.78 81	97.99 131	97.87 120	98.75 69	97.46 117	94.54 105	97.62 56	99.48 34	98.76 79	97.65 59	96.09 144	96.15 132	97.20 154	95.28 146

v7new		97.30 90	96.88 99	97.78 81	97.99 131	97.87 120	98.75 69	97.46 117	94.54 105	97.62 56	99.48 34	98.76 79	97.65 59	96.09 144	96.15 132	97.20 154	95.28 146

v8		97.51 73	97.16 78	97.91 66	97.99 131	98.48 68	98.76 63	98.17 62	94.54 105	97.69 53	99.48 34	98.76 79	97.63 64	96.10 143	96.14 136	97.20 154	96.64 111

v17		97.54 68	97.21 73	97.92 65	98.02 127	98.50 65	98.79 60	98.24 51	94.39 115	97.60 59	99.45 40	98.72 84	97.68 57	96.29 136	96.28 128	97.19 158	96.86 102

v16		97.51 73	97.19 75	97.89 69	97.99 131	98.49 66	98.77 62	98.23 54	94.29 117	97.48 63	99.42 43	98.68 85	97.69 56	96.28 137	96.29 127	97.18 159	96.85 104

IterMVS-LS		96.35 124	95.85 131	96.93 121	97.53 163	98.00 102	97.37 149	97.97 76	95.49 72	96.71 100	98.94 63	93.23 175	94.82 134	93.15 198	95.05 166	97.17 160	97.12 90
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
MVSTER		91.97 193	90.31 197	93.91 194	96.81 190	96.91 161	94.22 219	95.64 185	84.98 221	92.98 200	93.42 180	72.56 225	86.64 214	95.11 169	93.89 182	97.16 161	95.31 142

v18		97.40 84	97.04 88	97.81 77	97.90 142	98.42 70	98.71 75	98.17 62	94.06 125	97.34 72	99.40 45	98.59 94	97.60 65	96.05 147	96.12 139	97.14 162	96.67 109

PM-MVS		96.85 109	96.62 113	97.11 112	97.13 183	96.51 168	98.29 96	94.65 205	94.84 92	98.12 37	98.59 79	97.20 136	97.41 74	96.24 139	96.41 125	97.09 163	96.56 114

USDC		96.30 125	95.64 134	97.07 114	97.62 161	96.35 177	97.17 162	95.71 184	95.52 69	99.17 6	98.11 96	97.46 130	95.67 121	95.44 165	93.60 184	97.09 163	92.99 184

TinyColmap		96.64 120	96.07 124	97.32 103	97.84 149	96.40 172	97.63 130	96.25 166	95.86 53	98.98 12	97.94 99	96.34 150	96.17 111	97.30 93	95.38 161	97.04 165	93.24 179

MVS_Test		95.34 148	94.88 149	95.89 159	96.93 188	96.84 165	96.66 176	97.08 142	90.06 184	94.02 176	97.61 107	96.64 145	93.59 157	92.73 202	94.02 179	97.03 166	96.24 120

Fast-Effi-MVS+-dtu		94.34 168	93.26 174	95.62 168	97.82 150	95.97 182	95.86 194	99.01 13	86.88 210	93.39 191	90.83 203	95.46 159	90.61 185	94.46 183	94.68 172	97.01 167	94.51 160

PatchMatch-RL		94.79 160	93.75 166	96.00 157	96.80 191	95.00 189	95.47 205	95.25 193	90.68 176	95.80 136	92.97 185	93.64 173	95.67 121	96.13 142	95.81 151	96.99 168	92.01 188

testgi		94.81 159	96.05 125	93.35 199	99.06 53	96.87 164	97.57 134	96.70 159	95.77 57	88.60 221	93.19 184	98.87 64	81.21 226	97.03 108	96.64 115	96.97 169	93.99 173

MAR-MVS		95.51 142	94.49 156	96.71 130	97.92 140	96.40 172	96.72 174	98.04 72	86.74 212	96.72 97	92.52 190	95.14 162	94.02 153	96.81 116	96.54 120	96.85 170	97.25 84
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
FPMVS		94.70 161	94.99 148	94.37 189	95.84 209	93.20 200	96.00 192	91.93 217	95.03 86	94.64 165	94.68 164	93.29 174	90.95 181	98.07 66	97.34 90	96.85 170	93.29 177

TSAR-MVS + ACMM		97.54 68	97.79 48	97.26 105	98.23 104	98.10 93	97.71 124	97.88 84	95.97 47	95.57 146	98.71 77	98.57 96	97.36 76	97.74 73	96.81 108	96.83 172	98.59 23

PMMVS		91.67 198	91.47 194	91.91 213	89.43 235	88.61 230	94.99 215	85.67 229	87.50 207	93.80 181	94.42 172	94.88 163	90.71 184	92.26 205	92.96 191	96.83 172	89.65 199

V42		97.10 101	96.97 94	97.26 105	97.64 158	97.60 134	98.45 87	95.99 172	94.44 113	97.35 71	99.40 45	98.63 91	97.34 78	96.33 135	96.38 126	96.82 174	96.00 126

pmmvs4		95.37 147	94.25 157	96.67 134	97.01 186	95.28 188	97.60 131	96.07 170	93.11 146	97.29 74	98.09 97	94.23 171	95.21 128	91.56 209	93.91 181	96.82 174	93.59 176

ambc			96.78 108		99.01 55	97.11 157	95.73 199		95.91 50	99.25 2	98.56 81	97.17 137	97.04 86	96.76 118	95.22 164	96.72 176	96.73 108

gg-mvs-nofinetune		94.13 173	93.93 161	94.37 189	97.99 131	95.86 183	95.45 208	99.22 9	97.61 18	95.10 155	99.50 33	84.50 194	81.73 225	95.31 166	94.12 177	96.71 177	90.59 194

TSAR-MVS + COLMAP		96.05 131	95.94 128	96.18 151	97.46 171	96.41 171	97.26 158	95.83 179	94.69 96	95.30 149	98.31 88	96.52 147	94.71 137	95.48 164	94.87 168	96.54 178	95.33 141

v148		96.99 106	96.70 111	97.34 101	97.89 143	97.23 148	98.33 93	96.96 146	95.57 64	97.12 81	98.99 61	99.40 13	97.23 81	96.22 140	95.45 158	96.50 179	94.02 172

test0.0.03 1		91.17 202	91.50 193	90.80 219	98.01 129	95.46 186	94.22 219	95.80 180	86.55 216	81.75 236	90.83 203	87.93 190	78.48 229	94.51 182	94.11 178	96.50 179	91.08 192

MIMVSNet		93.68 181	93.96 160	93.35 199	97.82 150	96.08 181	96.34 184	98.46 31	91.28 168	86.67 230	94.95 160	94.87 164	84.39 221	94.53 178	94.65 173	96.45 181	91.34 191

CR-MVSNet		91.94 194	88.50 205	95.94 158	96.14 201	92.08 209	95.23 212	98.47 28	84.30 225	96.44 108	94.58 166	75.57 219	92.92 160	90.22 217	92.22 194	96.43 182	90.56 195

RPMNet		90.52 205	86.27 221	95.48 172	95.95 206	92.08 209	95.55 204	98.12 65	84.30 225	95.60 145	87.49 214	72.78 224	91.24 177	87.93 221	89.34 208	96.41 183	89.98 198

HyFIR lowres test		95.05 153	93.54 168	96.81 127	97.81 152	96.88 162	98.18 98	97.46 117	94.28 118	94.98 159	96.57 130	92.89 177	96.15 112	90.90 214	91.87 199	96.28 184	91.35 190

GA-MVS		94.18 172	92.98 175	95.58 169	97.36 175	96.42 170	96.21 190	95.86 176	90.29 179	95.08 156	96.19 137	85.37 193	92.82 163	94.01 190	94.14 176	96.16 185	94.41 164

EPNet		94.33 171	93.52 169	95.27 176	98.81 70	94.71 192	96.77 172	98.20 56	88.12 198	96.53 105	92.53 189	91.19 184	85.25 220	95.22 168	95.26 163	96.09 186	97.63 65
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
no-one		97.16 99	97.57 62	96.68 133	96.30 200	95.74 184	98.40 91	94.04 212	96.28 37	96.30 118	97.95 98	99.45 10	99.06 4	96.93 113	98.19 60	95.99 187	98.48 30

gm-plane-assit		91.85 196	87.91 209	96.44 146	99.14 44	98.25 77	99.02 33	97.38 131	95.57 64	98.31 29	99.34 49	51.00 241	88.93 199	93.16 197	91.57 200	95.85 188	86.50 216

new-patchmatchnet		94.48 163	94.02 159	95.02 181	97.51 169	95.00 189	95.68 200	94.26 211	97.32 21	95.73 138	99.60 21	98.22 112	91.30 176	94.13 188	84.41 217	95.65 189	89.45 201

CANet_DTU		94.96 155	94.62 154	95.35 173	98.03 126	96.11 180	96.92 168	95.60 186	88.59 192	97.27 75	95.27 154	96.50 148	88.77 201	95.53 161	95.59 155	95.54 190	94.78 154

FMVSNet5		89.65 212	87.60 212	92.04 212	95.63 213	96.61 167	94.82 217	94.75 201	80.11 234	87.72 227	77.73 235	73.81 223	83.81 222	95.64 158	96.08 141	95.49 191	93.21 180

TAMVS		92.46 187	93.34 171	91.44 216	97.03 185	93.84 198	94.68 218	90.60 220	90.44 178	85.31 231	97.14 120	93.03 176	85.78 216	94.34 185	93.67 183	95.22 192	90.93 193

MS-PatchMatch		94.84 158	94.76 150	94.94 182	96.38 199	94.69 193	95.90 193	94.03 213	92.49 153	93.81 180	95.79 146	96.38 149	94.54 139	94.70 175	94.85 169	94.97 193	94.43 163

CMPMVS	71.81 19	92.34 191	92.85 176	91.75 214	92.70 229	90.43 222	88.84 234	88.56 223	85.87 219	94.35 170	90.98 201	95.89 157	91.14 178	96.14 141	94.83 170	94.93 194	95.78 132
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
diffmvs		94.34 168	93.83 165	94.93 183	96.41 198	94.88 191	96.41 181	96.09 169	93.24 141	93.79 182	98.12 95	92.20 181	91.98 173	90.79 215	92.20 196	94.91 195	95.35 140

CVMVSNet		94.01 176	94.25 157	93.73 196	94.36 223	92.44 205	97.45 145	88.56 223	95.59 62	93.06 199	98.88 64	90.03 188	94.84 133	94.08 189	93.45 185	94.09 196	95.31 142

MDA-MVSNet-bldmvs		95.45 144	95.20 140	95.74 164	94.24 224	96.38 175	97.93 113	94.80 200	95.56 67	96.87 92	98.29 89	95.24 161	96.50 103	98.65 44	90.38 205	94.09 196	91.93 189

IterMVS		94.48 163	93.46 170	95.66 166	97.52 164	96.43 169	97.20 160	94.73 203	92.91 151	96.44 108	98.75 75	91.10 185	94.53 140	92.10 206	90.10 207	93.51 198	92.84 185
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
testmv		92.35 190	92.53 180	92.13 210	97.16 181	92.68 204	96.31 187	94.61 208	86.68 214	88.16 224	97.27 117	97.09 141	83.28 223	94.52 180	93.39 187	93.26 199	86.10 218

test1235678		92.36 189	92.55 178	92.13 210	97.16 181	92.69 203	96.32 186	94.62 206	86.69 213	88.16 224	97.28 116	97.13 140	83.28 223	94.54 177	93.40 186	93.26 199	86.11 217

new_pmnet		90.85 204	92.26 185	89.21 225	93.68 227	89.05 228	93.20 228	84.16 232	92.99 149	84.25 232	97.72 105	94.60 166	86.80 213	93.20 196	91.30 201	93.21 201	86.94 215

test-mter		89.16 214	88.14 208	90.37 220	94.79 220	91.05 219	93.60 225	85.26 230	81.65 230	88.32 223	92.22 191	79.35 217	87.03 211	92.28 203	90.12 206	93.19 202	90.29 197

EU-MVSNet		96.03 132	96.23 119	95.80 163	95.48 216	94.18 194	98.99 39	91.51 218	97.22 22	97.66 54	99.15 57	98.51 99	98.08 32	95.92 151	92.88 192	93.09 203	95.72 135

GG-mvs-BLEND		61.03 232	87.02 216	30.71 234	0.74 240	90.01 223	78.90 238	0.74 238	84.56 223	9.46 241	79.17 233	90.69 187	1.37 238	91.74 208	89.13 211	93.04 204	83.83 226

testus		90.01 207	90.03 200	89.98 221	95.89 207	91.43 217	93.88 222	89.30 222	83.54 227	89.68 214	87.81 213	94.62 165	78.31 230	92.87 199	92.01 197	92.85 205	87.91 210

EPNet_dtu		93.45 182	92.51 181	94.55 188	98.39 88	91.67 215	95.46 206	97.50 109	86.56 215	97.38 69	93.52 179	94.20 172	85.82 215	93.31 195	92.53 193	92.72 206	95.76 133
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
test-LLR		89.77 211	87.47 213	92.45 207	98.01 129	89.77 224	93.25 226	95.80 180	81.56 231	89.19 216	92.08 193	79.59 215	85.77 218	91.47 211	89.04 212	92.69 207	88.75 204

TESTMET0.1,1		88.60 219	87.47 213	89.93 223	94.23 225	89.77 224	93.25 226	84.47 231	81.56 231	89.19 216	92.08 193	79.59 215	85.77 218	91.47 211	89.04 212	92.69 207	88.75 204

PatchT		91.40 200	88.54 204	94.74 184	91.48 234	92.18 208	97.42 147	97.51 107	84.96 222	96.44 108	94.16 173	75.47 220	92.92 160	90.22 217	92.22 194	92.66 209	90.56 195

test2356		85.48 228	81.66 230	89.94 222	95.36 218	88.71 229	91.69 230	92.78 215	78.28 236	86.79 229	85.80 217	58.29 235	80.44 227	89.39 219	89.17 210	92.60 210	81.98 230

pmmvs3		91.20 201	91.40 195	90.96 218	91.71 233	91.08 218	95.41 209	81.34 233	87.36 208	94.57 166	95.02 158	94.30 170	90.42 186	94.28 186	89.26 209	92.30 211	88.49 207

CHOSEN 1792x2688		94.98 154	94.69 151	95.31 174	97.27 179	95.58 185	97.90 116	95.56 187	95.03 86	93.77 183	95.65 148	99.29 18	95.30 125	91.51 210	91.28 202	92.05 212	94.50 161

LP		92.03 192	90.19 199	94.17 193	94.52 222	93.87 197	96.79 171	95.05 195	93.58 135	95.62 143	95.68 147	83.37 202	91.78 174	90.73 216	86.99 214	91.27 213	87.09 214

PMMVS2		86.47 227	92.62 177	79.29 231	92.01 230	85.63 233	93.74 224	86.37 226	93.95 129	54.18 240	98.19 92	97.39 132	58.46 234	96.57 125	93.07 189	90.99 214	83.55 227

MDTV_nov1_ep13_2view		94.39 165	93.34 171	95.63 167	97.23 180	95.33 187	97.76 122	96.84 153	94.55 102	97.47 64	98.96 62	97.70 125	93.88 154	92.27 204	86.81 215	90.56 215	87.73 211

MDTV_nov1_ep13		90.30 206	87.32 215	93.78 195	96.00 204	92.97 201	95.46 206	95.39 190	88.61 191	95.41 147	94.45 171	80.39 214	89.87 193	86.58 224	83.54 221	90.56 215	84.71 221

DWT-MVSNet_training		86.69 225	81.24 231	93.05 202	95.31 219	92.06 211	95.75 198	91.51 218	84.32 224	94.49 167	83.46 222	55.37 238	90.81 183	82.76 232	83.19 223	90.45 217	87.52 212

dps		88.36 220	84.32 227	93.07 201	93.86 226	92.29 207	94.89 216	95.93 174	83.50 228	93.13 196	91.87 195	67.79 230	90.32 188	85.99 226	83.22 222	90.28 218	85.56 219

CostFormer		89.06 215	85.65 223	93.03 204	95.88 208	92.40 206	95.30 211	95.86 176	86.49 217	93.12 198	93.40 182	74.18 222	88.25 204	82.99 231	81.46 225	89.77 219	88.66 206

tpmp4_e23		88.68 217	84.61 225	93.43 197	96.00 204	91.46 216	95.40 210	96.60 164	87.71 200	94.67 164	88.54 210	69.81 227	88.41 203	85.50 227	81.08 227	89.52 220	88.18 209

N_pmnet		92.46 187	92.38 182	92.55 206	97.91 141	93.47 199	97.42 147	94.01 214	96.40 34	88.48 222	98.50 82	98.07 116	88.14 205	91.04 213	84.30 218	89.35 221	84.85 220

EPMVS		89.28 213	86.28 220	92.79 205	96.01 203	92.00 212	95.83 195	95.85 178	90.78 175	91.00 208	94.58 166	74.65 221	88.93 199	85.00 228	82.88 224	89.09 222	84.09 224

1111		88.65 218	87.69 211	89.78 224	98.84 66	94.02 195	95.79 196	98.19 58	91.57 163	82.27 233	98.19 92	53.19 239	74.80 231	94.98 172	93.04 190	88.80 223	88.82 203

PatchmatchNet		89.98 208	86.23 222	94.36 191	96.56 196	91.90 214	96.07 191	96.72 157	90.18 181	96.87 92	93.36 183	78.06 218	91.46 175	84.71 230	81.40 226	88.45 224	83.97 225
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
ADS-MVSNet		89.89 209	87.70 210	92.43 208	95.52 214	90.91 220	95.57 201	95.33 191	93.19 143	91.21 207	93.41 181	82.12 207	89.05 197	86.21 225	83.77 220	87.92 225	84.31 222

MVS-HIRNet		88.72 216	86.49 219	91.33 217	91.81 232	85.66 232	87.02 236	96.25 166	81.48 233	94.82 161	96.31 136	92.14 182	90.32 188	87.60 222	83.82 219	87.74 226	78.42 231

test12356		88.21 221	89.73 201	86.43 229	91.94 231	89.52 227	91.79 229	86.07 228	85.51 220	81.97 235	95.56 150	96.20 152	79.11 228	94.14 187	90.94 203	87.70 227	76.23 232

CHOSEN 280x420		91.55 199	90.27 198	93.05 202	94.61 221	88.01 231	96.56 178	94.62 206	88.04 199	94.20 172	92.66 188	86.60 191	90.82 182	95.06 171	91.89 198	87.49 228	89.61 200

tpm cat1		87.19 223	82.78 229	92.33 209	95.66 211	90.61 221	94.19 221	95.27 192	86.97 209	94.38 169	90.91 202	69.40 229	87.21 209	79.57 234	77.82 230	87.25 229	84.18 223

tpm		89.84 210	86.81 218	93.36 198	96.60 195	91.92 213	95.02 214	97.39 130	86.79 211	96.54 104	95.03 157	69.70 228	87.66 207	88.79 220	86.19 216	86.95 230	89.27 202

E-PMN		86.94 224	85.10 224	89.09 227	95.77 210	83.54 235	89.89 232	86.55 225	92.18 157	87.34 228	94.02 174	83.42 201	89.63 195	93.32 194	77.11 231	85.33 231	72.09 233

EMVS		86.63 226	84.48 226	89.15 226	95.51 215	83.66 234	90.19 231	86.14 227	91.78 161	88.68 220	93.83 178	81.97 210	89.05 197	92.76 201	76.09 232	85.31 232	71.28 234

tpmrst		87.60 222	84.13 228	91.66 215	95.65 212	89.73 226	93.77 223	94.74 202	88.85 189	93.35 193	95.60 149	72.37 226	87.40 208	81.24 233	78.19 229	85.02 233	82.90 229

testpf		81.59 230	76.31 232	87.75 228	93.50 228	83.16 236	89.19 233	95.94 173	73.85 237	90.39 209	80.32 231	61.17 234	73.99 233	76.52 235	75.82 233	83.50 234	83.33 228

DeepMVS_CX						72.99 237	80.14 237	37.34 234	83.46 229	60.13 239	84.40 219	85.48 192	86.93 212	87.22 223		79.61 235	87.32 213

MVE	72.99 18	85.37 229	89.43 202	80.63 230	74.43 236	71.94 238	88.25 235	89.81 221	93.27 140	67.32 238	96.32 135	91.83 183	90.40 187	93.36 193	90.79 204	73.55 236	88.49 207
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
tmp_tt				45.72 233	60.00 237	38.74 239	45.50 239	12.18 235	79.58 235	68.42 237	67.62 236	65.04 231	22.12 235	84.83 229	78.72 228	66.08 237

.test1245		69.06 231	63.57 233	75.47 232	98.84 66	94.02 195	95.79 196	98.19 58	91.57 163	82.27 233	98.19 92	53.19 239	74.80 231	94.98 172	5.51 235	2.94 238	7.51 235

testmvs		4.99 233	6.88 234	2.78 236	1.73 238	2.04 241	3.10 241	1.71 236	7.27 238	3.92 243	12.18 237	6.71 242	3.31 237	6.94 236	5.51 235	2.94 238	7.51 235

test123		4.41 234	5.71 235	2.88 235	1.28 239	2.21 240	3.09 242	1.65 237	6.35 239	4.98 242	8.53 238	3.88 243	3.46 236	5.79 237	5.71 234	2.85 240	7.50 237

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

MTAPA										97.43 67		99.27 22

MTMP										97.63 55		99.03 45

Patchmatch-RL test							17.42 240

mPP-MVS					99.58 6							98.98 50

NP-MVS									89.27 188

Patchmtry						92.70 202	95.23 212	98.47 28		96.44 108

High-resolution multi-view benchmark