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.

Anonymous20231211		99.83 1	99.80 1	99.86 1	99.97 1	99.87 1	99.90 1	99.92 1	99.76 1	99.82 2	99.79 37	99.98 1	99.63 12	99.84 3	99.78 3	99.94 1	99.61 6

v7n		99.68 5	99.61 4	99.76 9	99.89 14	99.74 8	99.87 2	99.82 14	99.20 6	99.71 6	99.96 1	99.73 12	99.76 5	99.58 17	99.59 15	99.52 43	99.46 17

SixPastTwentyTwo		99.70 4	99.59 7	99.82 3	99.93 3	99.80 2	99.86 3	99.87 7	98.87 14	99.79 5	99.85 27	99.33 63	99.74 7	99.85 2	99.82 1	99.74 22	99.63 4

v52		99.67 6	99.59 7	99.76 9	99.91 9	99.69 11	99.85 4	99.79 16	99.12 9	99.68 12	99.95 2	99.72 14	99.77 2	99.58 17	99.61 11	99.54 38	99.50 13

V4		99.67 6	99.60 6	99.76 9	99.91 9	99.69 11	99.85 4	99.79 16	99.13 8	99.68 12	99.95 2	99.72 14	99.77 2	99.58 17	99.61 11	99.54 38	99.50 13

v748		99.67 6	99.61 4	99.75 13	99.87 17	99.68 13	99.84 6	99.79 16	99.14 7	99.64 17	99.89 12	99.88 5	99.72 8	99.58 17	99.57 17	99.62 30	99.50 13

anonymousdsp		99.64 9	99.55 9	99.74 14	99.87 17	99.56 22	99.82 7	99.73 28	98.54 19	99.71 6	99.92 6	99.84 7	99.61 13	99.70 6	99.63 6	99.69 26	99.64 2

PEN-MVS		99.54 11	99.30 18	99.83 2	99.92 5	99.76 5	99.80 8	99.88 4	97.60 66	99.71 6	99.59 55	99.52 43	99.75 6	99.64 12	99.51 19	99.90 3	99.46 17

WR-MVS		99.61 10	99.44 11	99.82 3	99.92 5	99.80 2	99.80 8	99.89 2	98.54 19	99.66 15	99.78 40	99.16 86	99.68 10	99.70 6	99.63 6	99.94 1	99.49 16

DTE-MVSNet		99.52 13	99.27 19	99.82 3	99.93 3	99.77 4	99.79 10	99.87 7	97.89 45	99.70 11	99.55 62	99.21 78	99.77 2	99.65 10	99.43 23	99.90 3	99.36 21

PS-CasMVS		99.50 14	99.23 21	99.82 3	99.92 5	99.75 7	99.78 11	99.89 2	97.30 85	99.71 6	99.60 53	99.23 74	99.71 9	99.65 10	99.55 18	99.90 3	99.56 8

TDRefinement		99.54 11	99.50 10	99.60 19	99.70 62	99.35 45	99.77 12	99.58 55	99.40 5	99.28 57	99.66 45	99.41 51	99.55 20	99.74 5	99.65 5	99.70 23	99.25 27

LTVRE_ROB	98.82 1	99.76 2	99.75 2	99.77 8	99.87 17	99.71 9	99.77 12	99.76 22	99.52 3	99.80 3	99.79 37	99.91 2	99.56 18	99.83 4	99.75 4	99.86 9	99.75 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
pmmvs6		99.74 3	99.75 2	99.73 15	99.92 5	99.67 15	99.76 14	99.84 11	99.59 2	99.52 27	99.87 18	99.91 2	99.43 39	99.87 1	99.81 2	99.89 6	99.52 10

TransMVSNet (Re)		99.45 18	99.32 16	99.61 17	99.88 16	99.60 18	99.75 15	99.63 47	99.11 10	99.28 57	99.83 31	98.35 140	99.27 62	99.70 6	99.62 10	99.84 10	99.03 48

CP-MVSNet		99.39 20	99.04 29	99.80 7	99.91 9	99.70 10	99.75 15	99.88 4	96.82 107	99.68 12	99.32 76	98.86 120	99.68 10	99.57 21	99.47 21	99.89 6	99.52 10

WR-MVS_H		99.48 15	99.23 21	99.76 9	99.91 9	99.76 5	99.75 15	99.88 4	97.27 88	99.58 20	99.56 59	99.24 72	99.56 18	99.60 15	99.60 14	99.88 8	99.58 7

MIMVSNet1		99.46 17	99.34 13	99.60 19	99.83 23	99.68 13	99.74 18	99.71 33	98.20 27	99.41 35	99.86 22	99.66 27	99.41 42	99.50 23	99.39 25	99.50 49	99.10 41

ACMH	97.81 6	99.44 19	99.33 14	99.56 23	99.81 28	99.42 38	99.73 19	99.58 55	99.02 11	99.10 81	99.41 73	99.69 19	99.60 14	99.45 27	99.26 35	99.55 37	99.05 45
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
FC-MVSNet-train		99.13 37	99.05 28	99.21 74	99.87 17	99.57 21	99.67 20	99.60 54	96.75 114	98.28 155	99.48 67	99.52 43	98.10 135	99.47 26	99.37 27	99.76 21	99.21 32

v13		99.22 29	98.99 31	99.49 31	99.68 66	99.58 20	99.67 20	99.77 21	98.10 29	99.36 38	99.88 13	99.37 57	99.54 22	98.50 82	98.51 89	98.92 121	99.03 48

v12		99.19 31	98.95 32	99.48 32	99.67 69	99.56 22	99.66 22	99.76 22	98.06 31	99.33 43	99.88 13	99.34 62	99.53 23	98.42 89	98.43 94	98.91 124	98.97 55

v1240		98.86 70	98.41 86	99.38 51	99.59 98	99.05 89	99.65 23	99.14 150	97.68 61	99.66 15	99.93 5	98.72 127	99.45 34	97.38 174	97.72 142	98.79 147	98.35 113

v11		99.19 31	98.95 32	99.47 33	99.66 73	99.54 28	99.65 23	99.73 28	98.06 31	99.38 37	99.92 6	99.40 54	99.55 20	98.29 102	98.50 90	98.88 129	98.92 64

V9		99.16 35	98.90 39	99.46 34	99.66 73	99.54 28	99.65 23	99.75 25	98.01 34	99.31 47	99.87 18	99.31 66	99.51 25	98.34 96	98.34 97	98.90 126	98.91 65

v1921920		98.89 63	98.46 73	99.39 46	99.58 100	99.04 93	99.64 26	99.17 146	97.91 42	99.64 17	99.92 6	98.99 116	99.44 37	97.44 170	97.57 154	98.84 138	98.35 113

v1192		98.91 59	98.48 72	99.41 42	99.61 97	99.03 98	99.64 26	99.25 134	97.91 42	99.58 20	99.92 6	99.07 109	99.45 34	97.55 163	97.68 144	98.93 118	98.23 125

EU-MVSNet		98.68 87	98.94 36	98.37 158	99.14 176	98.74 135	99.64 26	98.20 204	98.21 26	99.17 68	99.66 45	99.18 82	99.08 79	99.11 40	98.86 57	95.00 213	98.83 70

FC-MVSNet-test		99.32 23	99.33 14	99.31 65	99.87 17	99.65 17	99.63 29	99.75 25	97.76 50	97.29 202	99.87 18	99.63 33	99.52 24	99.66 9	99.63 6	99.77 19	99.12 37

pm-mvs1		99.47 16	99.38 12	99.57 22	99.82 25	99.49 32	99.63 29	99.65 43	98.88 13	99.31 47	99.85 27	99.02 112	99.23 65	99.60 15	99.58 16	99.80 15	99.22 31

V14		99.13 37	98.85 44	99.45 35	99.65 79	99.52 30	99.63 29	99.74 27	97.97 36	99.30 50	99.87 18	99.27 70	99.49 29	98.23 108	98.24 100	98.88 129	98.83 70

Vis-MVSNet		99.25 26	99.32 16	99.17 79	99.65 79	99.55 26	99.63 29	99.33 119	98.16 28	99.29 52	99.65 49	99.77 8	97.56 153	99.44 28	99.14 39	99.58 35	99.51 12
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
v144192		98.88 65	98.46 73	99.37 53	99.56 105	99.03 98	99.61 33	99.26 131	97.79 49	99.58 20	99.88 13	99.11 100	99.43 39	97.38 174	97.61 150	98.80 145	98.43 108

v15		99.09 40	98.79 47	99.43 39	99.64 87	99.50 31	99.61 33	99.73 28	97.92 40	99.28 57	99.86 22	99.24 72	99.47 31	98.12 119	98.14 105	98.87 131	98.76 81

tfpnnormal		99.19 31	98.90 39	99.54 26	99.81 28	99.55 26	99.60 35	99.54 66	98.53 21	99.23 61	98.40 120	98.23 143	99.40 43	99.29 33	99.36 28	99.63 29	98.95 61

ACMH+	97.53 7	99.29 25	99.20 24	99.40 45	99.81 28	99.22 62	99.59 36	99.50 76	98.64 18	98.29 154	99.21 85	99.69 19	99.57 16	99.53 22	99.33 30	99.66 28	98.81 74

v1144		98.94 53	98.53 67	99.42 41	99.62 94	99.03 98	99.58 37	99.36 111	97.99 35	99.49 30	99.91 11	99.20 80	99.51 25	97.61 159	97.85 127	98.95 116	98.10 138

v7		98.91 59	98.53 67	99.36 55	99.53 117	98.99 104	99.57 38	99.36 111	97.58 69	99.32 45	99.88 13	99.23 74	99.50 27	97.77 147	97.98 115	98.91 124	98.26 122

v10		99.01 45	98.66 57	99.41 42	99.52 122	99.39 41	99.57 38	99.66 41	97.59 67	99.32 45	99.88 13	99.23 74	99.50 27	97.77 147	97.98 115	98.92 121	98.78 79

v1141		98.87 66	98.45 77	99.36 55	99.65 79	99.04 93	99.56 40	99.38 100	97.83 46	99.29 52	99.86 22	99.16 86	99.40 43	97.68 153	97.78 130	98.86 134	97.82 150

divwei89l23v2f112		98.87 66	98.45 77	99.36 55	99.65 79	99.04 93	99.56 40	99.38 100	97.83 46	99.29 52	99.86 22	99.15 90	99.40 43	97.68 153	97.78 130	98.86 134	97.82 150

COLMAP_ROB	98.29 2	99.37 21	99.25 20	99.51 30	99.74 50	99.12 80	99.56 40	99.39 93	98.96 12	99.17 68	99.44 70	99.63 33	99.58 15	99.48 25	99.27 33	99.60 34	98.81 74
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
v17		98.96 51	98.63 58	99.35 60	99.54 110	99.41 39	99.55 43	99.70 34	97.40 80	99.10 81	99.79 37	99.10 101	99.40 43	97.96 126	97.99 113	98.80 145	98.77 80

v8		98.94 53	98.60 60	99.35 60	99.54 110	99.39 41	99.55 43	99.67 40	97.48 73	99.13 76	99.81 32	99.10 101	99.39 53	97.86 136	97.89 121	98.81 140	98.66 91

v2v482		98.85 72	98.40 88	99.38 51	99.65 79	98.98 105	99.55 43	99.39 93	97.92 40	99.35 41	99.85 27	99.14 92	99.39 53	97.50 165	97.78 130	98.98 113	97.60 157

v1		98.87 66	98.45 77	99.36 55	99.65 79	99.04 93	99.55 43	99.38 100	97.83 46	99.30 50	99.86 22	99.17 83	99.40 43	97.68 153	97.77 137	98.86 134	97.82 150

v1neww		98.84 73	98.45 77	99.29 69	99.54 110	98.98 105	99.54 47	99.37 108	97.48 73	99.10 81	99.80 35	99.12 96	99.40 43	97.85 139	97.89 121	98.81 140	98.04 141

v7new		98.84 73	98.45 77	99.29 69	99.54 110	98.98 105	99.54 47	99.37 108	97.48 73	99.10 81	99.80 35	99.12 96	99.40 43	97.85 139	97.89 121	98.81 140	98.04 141

v16		98.95 52	98.62 59	99.34 62	99.53 117	99.41 39	99.54 47	99.70 34	97.34 84	99.07 87	99.76 41	99.10 101	99.40 43	97.96 126	98.00 112	98.79 147	98.76 81

v6		98.84 73	98.46 73	99.30 66	99.54 110	98.98 105	99.54 47	99.37 108	97.49 72	99.11 80	99.81 32	99.13 95	99.40 43	97.86 136	97.89 121	98.81 140	98.04 141

APDe-MVS		99.15 36	98.95 32	99.39 46	99.77 37	99.28 54	99.52 51	99.54 66	97.22 93	99.06 88	99.20 86	99.64 31	99.05 82	99.14 38	99.02 52	99.39 69	99.17 35

HSP-MVS		98.50 106	98.05 113	99.03 97	99.67 69	99.33 48	99.51 52	99.26 131	95.28 159	98.51 136	98.19 129	99.74 11	98.29 125	97.69 152	96.70 179	98.96 114	99.41 20

TSAR-MVS + MP.		99.02 44	98.95 32	99.11 87	99.23 167	98.79 129	99.51 52	98.73 179	97.50 71	98.56 131	99.03 99	99.59 39	99.16 73	99.29 33	99.17 37	99.50 49	99.24 30

v18		98.89 63	98.54 65	99.30 66	99.50 125	99.37 44	99.51 52	99.68 37	97.25 92	99.00 96	99.76 41	99.04 110	99.36 55	97.81 143	97.86 126	98.77 150	98.68 90

V42		98.81 81	98.49 71	99.18 78	99.52 122	98.92 120	99.50 55	99.29 127	97.43 78	98.97 98	99.81 32	99.00 115	99.30 59	97.93 129	98.01 111	98.51 170	98.34 117

v148		98.77 84	98.45 77	99.15 81	99.68 66	98.94 118	99.49 56	99.31 126	97.95 38	98.91 108	99.65 49	99.62 35	99.18 68	97.99 125	97.64 148	98.33 175	97.38 166

DeepC-MVS	97.88 4	99.33 22	99.15 25	99.53 29	99.73 55	99.05 89	99.49 56	99.40 91	98.42 22	99.55 24	99.71 43	99.89 4	99.49 29	99.14 38	98.81 63	99.54 38	99.02 51
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
TranMVSNet+NR-MVSNet		99.23 27	98.91 38	99.61 17	99.81 28	99.45 36	99.47 58	99.68 37	97.28 87	99.39 36	99.54 63	99.08 107	99.45 34	99.09 43	98.84 61	99.83 11	99.04 46

ACMM	96.66 11	98.90 61	98.44 83	99.44 36	99.74 50	98.95 114	99.47 58	99.55 60	97.66 62	99.09 85	96.43 179	99.41 51	99.35 58	98.95 58	98.67 78	99.45 54	99.03 48
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
XVS					99.77 37	99.07 85	99.46 60			98.95 102		99.37 57				99.33 77

X-MVStestdata					99.77 37	99.07 85	99.46 60			98.95 102		99.37 57				99.33 77

pmmvs-eth3d		98.68 87	98.14 106	99.29 69	99.49 128	98.45 158	99.45 62	99.38 100	97.21 94	99.50 29	99.65 49	99.21 78	99.16 73	97.11 182	97.56 155	98.79 147	97.82 150

EPP-MVSNet		98.61 95	98.19 104	99.11 87	99.86 22	99.60 18	99.44 63	99.53 70	97.37 83	96.85 210	98.69 112	93.75 186	99.18 68	99.22 36	99.35 29	99.82 13	99.32 23

conf0.05thres1000		97.44 160	95.93 181	99.20 77	99.82 25	99.56 22	99.41 64	99.61 52	97.42 79	98.01 170	94.34 205	82.73 220	98.68 101	99.33 32	99.42 24	99.67 27	98.74 84

HyFIR lowres test		98.08 130	97.16 149	99.14 84	99.72 58	98.91 121	99.41 64	99.58 55	97.93 39	98.82 117	99.24 80	95.81 179	98.73 99	95.16 211	95.13 204	98.60 162	97.94 147

gm-plane-assit		94.62 208	91.39 217	98.39 156	99.90 13	99.47 35	99.40 66	99.65 43	97.44 77	99.56 23	99.68 44	59.40 241	94.23 209	96.17 196	94.77 207	97.61 191	92.79 216

LGP-MVS_train		98.84 73	98.33 94	99.44 36	99.78 35	98.98 105	99.39 67	99.55 60	95.41 157	98.90 109	97.51 151	99.68 22	99.44 37	99.03 53	98.81 63	99.57 36	98.91 65

CHOSEN 1792x2688		98.31 117	98.02 115	98.66 139	99.55 107	98.57 149	99.38 68	99.25 134	98.42 22	98.48 142	99.58 57	99.85 6	98.31 124	95.75 202	95.71 195	96.96 202	98.27 121

no-one		99.01 45	98.94 36	99.09 91	98.97 190	98.55 150	99.37 69	99.04 161	97.59 67	99.36 38	99.66 45	99.75 9	99.57 16	98.47 83	99.27 33	98.21 181	99.30 25

ACMMP		98.82 80	98.33 94	99.39 46	99.77 37	99.14 77	99.37 69	99.54 66	96.47 132	99.03 95	96.26 183	99.52 43	99.28 61	98.92 62	98.80 66	99.37 71	99.16 36
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
Baseline_NR-MVSNet		99.18 34	98.87 41	99.54 26	99.74 50	99.56 22	99.36 71	99.62 51	96.53 128	99.29 52	99.85 27	98.64 134	99.40 43	99.03 53	99.63 6	99.83 11	98.86 69

ACMMPR		99.05 42	98.72 51	99.44 36	99.79 33	99.12 80	99.35 72	99.56 58	97.74 55	99.21 62	97.72 144	99.55 41	99.29 60	98.90 64	98.81 63	99.41 64	99.19 33

NR-MVSNet		99.10 39	98.68 56	99.58 21	99.89 14	99.23 59	99.35 72	99.63 47	96.58 122	99.36 38	99.05 96	98.67 132	99.46 32	99.63 13	98.73 73	99.80 15	98.88 68

PVSNet_Blended_VisFu		98.98 48	98.79 47	99.21 74	99.76 43	99.34 46	99.35 72	99.35 116	97.12 99	99.46 32	99.56 59	98.89 118	98.08 138	99.05 48	98.58 85	99.27 87	98.98 54

N_pmnet		96.68 178	95.70 186	97.84 183	99.42 137	98.00 180	99.35 72	98.21 202	98.40 24	98.13 162	99.42 72	99.30 67	97.44 160	94.00 221	88.79 219	94.47 217	91.96 219

SteuartSystems-ACMMP		98.94 53	98.52 69	99.43 39	99.79 33	99.13 78	99.33 76	99.55 60	96.17 142	99.04 93	97.53 150	99.65 30	99.46 32	99.04 52	98.76 69	99.44 56	99.35 22
Skip Steuart: Steuart Systems R&D Blog.
MDA-MVSNet-bldmvs		97.75 141	97.26 142	98.33 159	99.35 147	98.45 158	99.32 77	97.21 220	97.90 44	99.05 90	99.01 101	96.86 170	99.08 79	99.36 30	92.97 214	95.97 210	96.25 189

PGM-MVS		98.69 86	98.09 110	99.39 46	99.76 43	99.07 85	99.30 78	99.51 73	94.76 173	99.18 67	96.70 172	99.51 46	99.20 66	98.79 71	98.71 76	99.39 69	99.11 38

UniMVSNet_NR-MVSNet		98.97 49	98.46 73	99.56 23	99.76 43	99.34 46	99.29 79	99.61 52	96.55 126	99.55 24	99.05 96	97.96 154	99.36 55	98.84 66	98.50 90	99.81 14	98.97 55

DU-MVS		99.04 43	98.59 61	99.56 23	99.74 50	99.23 59	99.29 79	99.63 47	96.58 122	99.55 24	99.05 96	98.68 130	99.36 55	99.03 53	98.60 83	99.77 19	98.97 55

UniMVSNet (Re)		99.08 41	98.69 55	99.54 26	99.75 46	99.33 48	99.29 79	99.64 46	96.75 114	99.48 31	99.30 78	98.69 128	99.26 63	98.94 59	98.76 69	99.78 18	99.02 51

ACMP	96.54 13	98.87 66	98.40 88	99.41 42	99.74 50	98.88 125	99.29 79	99.50 76	96.85 103	98.96 100	97.05 163	99.66 27	99.43 39	98.98 57	98.60 83	99.52 43	98.81 74
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
UGNet		98.52 105	99.00 30	97.96 180	99.58 100	99.26 55	99.27 83	99.40 91	98.07 30	98.28 155	98.76 110	99.71 18	92.24 223	98.94 59	98.85 59	99.00 112	99.43 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
ACMMP_Plus		98.94 53	98.72 51	99.21 74	99.67 69	99.08 84	99.26 84	99.39 93	96.84 104	98.88 114	98.22 127	99.68 22	98.82 92	99.06 47	98.90 55	99.25 89	99.25 27

MP-MVS		98.78 83	98.30 96	99.34 62	99.75 46	98.95 114	99.26 84	99.46 85	95.78 153	99.17 68	96.98 167	99.72 14	99.06 81	98.84 66	98.74 72	99.33 77	99.11 38

IS_MVSNet		98.20 125	98.00 116	98.44 152	99.82 25	99.48 33	99.25 86	99.56 58	95.58 155	93.93 232	97.56 149	96.52 172	98.27 127	99.08 46	99.20 36	99.80 15	98.56 100

zzz-MVS		98.94 53	98.57 64	99.37 53	99.77 37	99.15 76	99.24 87	99.55 60	97.38 82	99.16 71	96.64 174	99.69 19	99.15 75	99.09 43	98.92 54	99.37 71	99.11 38

MDTV_nov1_ep13_2view		97.12 167	96.19 175	98.22 168	99.13 178	98.05 177	99.24 87	99.47 82	97.61 65	99.15 74	99.59 55	99.01 113	98.40 119	94.87 213	90.14 217	93.91 218	94.04 210

LS3D		98.79 82	98.52 69	99.12 85	99.64 87	99.09 82	99.24 87	99.46 85	97.75 53	98.93 106	97.47 152	98.23 143	97.98 141	99.36 30	99.30 32	99.46 53	98.42 109

DI_MVS_plusplus_trai		97.57 155	96.55 165	98.77 126	99.55 107	98.76 131	99.22 90	99.00 165	97.08 100	97.95 173	97.78 143	91.35 193	98.02 140	96.20 195	96.81 178	98.87 131	97.87 149

FMVSNet1		98.90 61	99.10 27	98.67 137	99.54 110	99.48 33	99.22 90	99.66 41	98.39 25	97.50 190	99.66 45	99.04 110	96.58 173	99.05 48	99.03 49	99.52 43	99.08 43

tfpn_n400		97.59 152	96.36 170	99.01 101	99.66 73	99.19 68	99.21 92	99.55 60	97.62 63	97.77 177	94.60 200	87.78 199	98.27 127	98.44 85	98.72 74	99.62 30	98.21 128

tfpnconf		97.59 152	96.36 170	99.01 101	99.66 73	99.19 68	99.21 92	99.55 60	97.62 63	97.77 177	94.60 200	87.78 199	98.27 127	98.44 85	98.72 74	99.62 30	98.21 128

test20.03		98.84 73	98.74 50	98.95 107	99.77 37	99.33 48	99.21 92	99.46 85	97.29 86	98.88 114	99.65 49	99.10 101	97.07 167	99.11 40	98.76 69	99.32 80	97.98 146

CP-MVS		98.86 70	98.43 85	99.36 55	99.68 66	98.97 112	99.19 95	99.46 85	96.60 121	99.20 63	97.11 162	99.51 46	99.15 75	98.92 62	98.82 62	99.45 54	99.08 43

X-MVS		98.59 98	97.99 117	99.30 66	99.75 46	99.07 85	99.17 96	99.50 76	96.62 119	98.95 102	93.95 206	99.37 57	99.11 78	98.94 59	98.86 57	99.35 75	99.09 42

Gipuma		99.22 29	98.86 42	99.64 16	99.70 62	99.24 57	99.17 96	99.63 47	99.52 3	99.89 1	96.54 178	99.14 92	99.93 1	99.42 29	99.15 38	99.52 43	99.04 46
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
Anonymous20231206		98.50 106	98.03 114	99.05 95	99.50 125	99.01 102	99.15 98	99.26 131	96.38 134	99.12 78	99.50 66	99.12 96	98.60 105	97.68 153	97.24 168	98.66 156	97.30 168

new-patchmatchnet		97.26 164	96.12 176	98.58 145	99.55 107	98.63 143	99.14 99	97.04 222	98.80 16	99.19 65	99.92 6	99.19 81	98.92 86	95.51 206	87.04 222	97.66 190	93.73 211

OPM-MVS		98.84 73	98.59 61	99.12 85	99.52 122	98.50 155	99.13 100	99.22 137	97.76 50	98.76 120	98.70 111	99.61 36	98.90 87	98.67 75	98.37 96	99.19 93	98.57 97

HFP-MVS		98.97 49	98.70 53	99.29 69	99.67 69	98.98 105	99.13 100	99.53 70	97.76 50	98.90 109	98.07 134	99.50 48	99.14 77	98.64 77	98.78 67	99.37 71	99.18 34

tfpnview11		97.49 157	96.22 174	98.97 105	99.63 92	99.24 57	99.12 102	99.54 66	96.76 112	97.77 177	94.60 200	87.78 199	98.25 130	97.93 129	99.14 39	99.52 43	98.08 140

tfpn		94.97 204	91.60 216	98.90 110	99.73 55	99.33 48	99.11 103	99.51 73	95.05 163	97.19 207	89.03 221	62.62 238	98.37 120	98.53 80	98.97 53	99.48 52	97.70 154

3Dnovator+	97.85 5	98.61 95	98.14 106	99.15 81	99.62 94	98.37 161	99.10 104	99.51 73	98.04 33	98.98 97	96.07 187	98.75 126	98.55 110	98.51 81	98.40 95	99.17 94	98.82 72

IterMVS-LS		98.23 122	97.66 129	98.90 110	99.63 92	99.38 43	99.07 105	99.48 81	97.75 53	98.81 118	99.37 75	94.57 184	97.88 145	96.54 192	97.04 173	98.53 167	98.97 55
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
view800		96.48 180	94.42 192	98.87 114	99.70 62	99.26 55	99.05 106	99.45 89	94.77 172	97.32 199	88.21 222	83.40 218	98.28 126	98.37 92	99.33 30	99.44 56	97.58 159

pmmvs5		98.37 114	97.81 123	99.03 97	99.46 130	98.97 112	99.03 107	98.96 168	95.85 151	99.05 90	99.45 69	98.66 133	98.79 94	96.02 199	97.52 156	98.87 131	98.21 128

1111		94.22 215	92.26 212	96.51 216	99.71 60	98.75 133	99.03 107	99.83 12	95.01 165	93.39 234	99.54 63	60.23 239	89.58 229	97.90 132	97.62 149	97.50 194	96.75 182

.test1245		74.10 231	68.09 233	81.11 232	99.71 60	98.75 133	99.03 107	99.83 12	95.01 165	93.39 234	99.54 63	60.23 239	89.58 229	97.90 132	10.38 235	5.14 239	14.81 235

PM-MVS		98.57 100	98.24 101	98.95 107	99.26 161	98.59 146	99.03 107	98.74 178	96.84 104	99.44 34	99.13 90	98.31 142	98.75 97	98.03 123	98.21 101	98.48 171	98.58 95

HPM-MVS++		98.56 103	98.08 111	99.11 87	99.53 117	98.61 145	99.02 111	99.32 124	96.29 139	99.06 88	97.23 157	99.50 48	98.77 95	98.15 115	97.90 119	98.96 114	98.90 67

DeepC-MVS_fast	97.38 8	98.65 90	98.34 93	99.02 100	99.33 148	98.29 163	98.99 112	98.71 181	97.40 80	99.31 47	98.20 128	99.40 54	98.54 112	98.33 99	98.18 104	99.23 92	98.58 95
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
view600		96.39 184	94.30 193	98.82 122	99.65 79	99.16 75	98.98 113	99.36 111	94.46 179	97.39 195	87.28 223	84.16 216	98.16 134	98.16 112	99.48 20	99.40 66	97.42 164

CSCG		99.23 27	99.15 25	99.32 64	99.83 23	99.45 36	98.97 114	99.21 139	98.83 15	99.04 93	99.43 71	99.64 31	99.26 63	98.85 65	98.20 103	99.62 30	99.62 5

DELS-MVS		98.63 93	98.70 53	98.55 148	99.24 166	99.04 93	98.96 115	98.52 191	96.83 106	98.38 147	99.58 57	99.68 22	97.06 168	98.74 73	98.44 93	99.10 97	98.59 94
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
CPTT-MVS		98.28 118	97.51 136	99.16 80	99.54 110	98.78 130	98.96 115	99.36 111	96.30 138	98.89 112	93.10 211	99.30 67	99.20 66	98.35 95	97.96 118	99.03 110	98.82 72

Fast-Effi-MVS+		98.42 112	97.79 124	99.15 81	99.69 65	98.66 141	98.94 117	99.68 37	94.49 177	99.05 90	98.06 136	98.86 120	98.48 115	98.18 111	97.78 130	99.05 108	98.54 101

thres600view7		96.35 185	94.27 194	98.79 125	99.66 73	99.18 70	98.94 117	99.38 100	94.37 185	97.21 204	87.19 225	84.10 217	98.10 135	98.16 112	99.47 21	99.42 61	97.43 163

APD-MVS		98.47 109	97.97 118	99.05 95	99.64 87	98.91 121	98.94 117	99.45 89	94.40 183	98.77 119	97.26 156	99.41 51	98.21 132	98.67 75	98.57 87	99.31 81	98.57 97
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
MVS_0304		98.57 100	98.36 91	98.82 122	99.72 58	98.94 118	98.92 120	99.14 150	96.76 112	99.33 43	98.30 124	99.73 12	96.74 170	98.05 122	97.79 129	99.08 102	98.97 55

TinyColmap		98.27 119	97.62 133	99.03 97	99.29 157	97.79 190	98.92 120	98.95 169	97.48 73	99.52 27	98.65 114	97.86 156	98.90 87	98.34 96	97.27 166	98.64 159	95.97 193

train_agg		97.99 131	97.26 142	98.83 119	99.43 136	98.22 171	98.91 122	99.07 158	94.43 181	97.96 172	96.42 180	99.30 67	98.81 93	97.39 172	96.62 182	98.82 139	98.47 103

UA-Net		99.30 24	99.22 23	99.39 46	99.94 2	99.66 16	98.91 122	99.86 9	97.74 55	98.74 123	99.00 102	99.60 38	99.17 71	99.50 23	99.39 25	99.70 23	99.64 2

PatchT		95.49 199	93.29 207	98.06 175	98.65 205	96.20 211	98.91 122	99.73 28	92.00 220	98.50 137	96.67 173	83.25 219	96.34 179	94.40 217	95.50 198	96.21 206	95.04 202

3Dnovator	98.16 3	98.65 90	98.35 92	99.00 103	99.59 98	98.70 137	98.90 125	99.36 111	97.97 36	99.09 85	96.55 177	99.09 105	97.97 142	98.70 74	98.65 81	99.12 96	98.81 74

tfpn1000		97.10 169	95.97 179	98.41 154	99.64 87	99.30 52	98.89 126	99.49 80	96.49 129	95.97 217	95.31 193	85.62 213	96.92 169	97.86 136	99.13 41	99.53 42	98.11 137

pmmvs4		97.87 138	97.02 153	98.86 117	99.20 168	97.68 195	98.89 126	99.03 162	96.57 124	99.12 78	99.03 99	97.26 165	98.42 118	95.16 211	96.34 186	98.53 167	97.10 178

thresconf0.02		95.49 199	92.74 210	98.70 133	99.32 151	98.70 137	98.87 128	99.21 139	95.95 148	97.57 187	90.63 218	73.55 232	97.86 147	96.09 198	97.03 174	99.40 66	97.22 174

IterMVS		97.40 161	96.67 159	98.25 163	99.45 131	98.66 141	98.87 128	98.73 179	96.40 133	98.94 105	99.56 59	95.26 181	97.58 152	95.38 207	94.70 208	95.90 211	96.72 183
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
thres400		96.22 190	94.08 197	98.72 130	99.58 100	99.05 89	98.83 130	99.22 137	94.01 192	97.40 193	86.34 231	84.91 215	97.93 143	97.85 139	99.08 43	99.37 71	97.28 169

CDPH-MVS		97.99 131	97.23 145	98.87 114	99.58 100	98.29 163	98.83 130	99.20 142	93.76 195	98.11 163	96.11 185	99.16 86	98.23 131	97.80 144	97.22 169	99.29 85	98.28 119

USDC		98.26 120	97.57 134	99.06 92	99.42 137	97.98 183	98.83 130	98.85 172	97.57 70	99.59 19	99.15 89	98.59 135	98.99 84	97.42 171	96.08 194	98.69 155	96.23 190

CANet		98.47 109	98.30 96	98.67 137	99.65 79	98.87 126	98.82 133	99.01 164	96.14 143	99.29 52	98.86 107	99.01 113	96.54 174	98.36 94	98.08 108	98.72 153	98.80 78

EG-PatchMatch MVS		99.01 45	98.77 49	99.28 73	99.64 87	98.90 124	98.81 134	99.27 130	96.55 126	99.71 6	99.31 77	99.66 27	99.17 71	99.28 35	99.11 42	99.10 97	98.57 97

Vis-MVSNet (Re-imp)		98.46 111	98.23 102	98.73 129	99.81 28	99.29 53	98.79 135	99.50 76	96.20 141	96.03 215	98.29 125	96.98 168	98.54 112	99.11 40	99.08 43	99.70 23	98.62 93

SMA-MVS		98.94 53	98.80 46	99.11 87	99.73 55	99.09 82	98.78 136	99.18 144	96.32 137	98.89 112	99.19 88	99.72 14	98.75 97	99.09 43	98.89 56	99.31 81	99.27 26

testgi		98.18 128	98.44 83	97.89 181	99.78 35	99.23 59	98.78 136	99.21 139	97.26 90	97.41 192	97.39 154	99.36 61	92.85 219	98.82 68	98.66 80	99.31 81	98.35 113

CHOSEN 280x420		96.80 175	96.30 172	97.39 191	99.09 181	96.52 206	98.76 138	99.29 127	93.88 194	97.65 184	98.34 121	93.66 187	96.29 183	98.28 105	97.73 141	93.27 222	95.70 195

LP		95.33 203	93.45 206	97.54 190	98.68 203	97.40 198	98.73 139	98.41 196	96.33 136	98.92 107	97.84 142	88.30 197	95.92 186	92.98 222	89.38 218	94.56 216	91.90 220

CVMVSNet		97.38 162	97.39 138	97.37 192	98.58 208	97.72 194	98.70 140	97.42 217	97.21 94	95.95 218	99.46 68	93.31 189	97.38 161	97.60 160	97.78 130	96.18 207	98.66 91

TSAR-MVS + ACMM		98.64 92	98.58 63	98.72 130	99.17 173	98.63 143	98.69 141	99.10 157	97.69 60	98.30 153	99.12 92	99.38 56	98.70 100	98.45 84	97.51 157	98.35 174	99.25 27

tfpn111		96.48 180	94.67 191	98.59 143	99.37 140	99.18 70	98.68 142	99.39 93	92.02 214	97.21 204	90.63 218	86.34 207	97.45 155	98.15 115	99.08 43	99.43 58	97.28 169

conf0.01		94.53 211	91.09 219	98.53 150	99.29 157	99.05 89	98.68 142	99.35 116	92.02 214	97.04 208	84.45 234	68.52 234	97.45 155	97.79 146	99.08 43	99.41 64	96.70 184

conf0.002		93.97 216	90.06 223	98.52 151	99.26 161	99.02 101	98.68 142	99.33 119	92.02 214	97.01 209	83.82 235	63.41 237	97.45 155	97.73 150	97.98 115	99.40 66	96.47 186

conf200view11		96.16 193	94.08 197	98.59 143	99.37 140	99.18 70	98.68 142	99.39 93	92.02 214	97.21 204	86.53 228	86.34 207	97.45 155	98.15 115	99.08 43	99.43 58	97.28 169

tfpn200view9		96.17 191	94.08 197	98.60 142	99.37 140	99.18 70	98.68 142	99.39 93	92.02 214	97.30 200	86.53 228	86.34 207	97.45 155	98.15 115	99.08 43	99.43 58	97.28 169

CDS-MVSNet		97.75 141	97.68 128	97.83 184	99.08 183	98.20 172	98.68 142	98.61 187	95.63 154	97.80 176	99.24 80	96.93 169	94.09 210	97.96 126	97.82 128	98.71 154	97.99 144
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
PHI-MVS		98.57 100	98.20 103	99.00 103	99.48 129	98.91 121	98.68 142	99.17 146	94.97 168	99.27 60	98.33 122	99.33 63	98.05 139	98.82 68	98.62 82	99.34 76	98.38 111

ESAPD		98.60 97	98.41 86	98.83 119	99.56 105	99.21 63	98.66 149	99.47 82	95.22 160	98.35 149	98.48 118	99.67 26	97.84 148	98.80 70	98.57 87	99.10 97	98.93 63

MCST-MVS		98.25 121	97.57 134	99.06 92	99.53 117	98.24 169	98.63 150	99.17 146	95.88 150	98.58 130	96.11 185	99.09 105	99.18 68	97.58 162	97.31 165	99.25 89	98.75 83

QAPM		98.62 94	98.40 88	98.89 112	99.57 104	98.80 128	98.63 150	99.35 116	96.82 107	98.60 128	98.85 109	99.08 107	98.09 137	98.31 100	98.21 101	99.08 102	98.72 85

GBi-Net		97.69 145	97.75 126	97.62 187	98.71 199	99.21 63	98.62 152	99.33 119	94.09 189	95.60 221	98.17 131	95.97 176	94.39 205	99.05 48	99.03 49	99.08 102	98.70 87

test1		97.69 145	97.75 126	97.62 187	98.71 199	99.21 63	98.62 152	99.33 119	94.09 189	95.60 221	98.17 131	95.97 176	94.39 205	99.05 48	99.03 49	99.08 102	98.70 87

FMVSNet2		97.94 134	98.08 111	97.77 186	98.71 199	99.21 63	98.62 152	99.47 82	96.62 119	96.37 214	99.20 86	97.70 158	94.39 205	97.39 172	97.75 138	99.08 102	98.70 87

tfpn_ndepth		96.69 177	95.49 188	98.09 174	99.17 173	99.13 78	98.61 155	99.38 100	94.90 171	95.85 219	92.85 213	88.19 198	96.07 184	97.28 179	98.67 78	99.49 51	97.44 162

PMVS	92.51 17	98.66 89	98.86 42	98.43 153	99.26 161	98.98 105	98.60 156	98.59 188	97.73 57	99.45 33	99.38 74	98.54 137	95.24 194	99.62 14	99.61 11	99.42 61	98.17 134
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
testmv		97.48 159	96.83 158	98.24 166	99.37 140	97.79 190	98.59 157	99.07 158	92.40 208	97.59 185	99.24 80	98.11 147	97.66 150	97.64 157	97.11 171	97.17 198	95.54 198

test1235678		97.49 157	96.84 157	98.24 166	99.37 140	97.79 190	98.59 157	99.07 158	92.41 207	97.59 185	99.24 80	98.15 146	97.66 150	97.64 157	97.12 170	97.17 198	95.55 197

thres200		96.23 189	94.13 195	98.69 135	99.44 134	99.18 70	98.58 159	99.38 100	93.52 198	97.35 197	86.33 232	85.83 212	97.93 143	98.16 112	98.78 67	99.42 61	97.10 178

OMC-MVS		98.35 115	98.10 109	98.64 141	98.85 194	97.99 181	98.56 160	98.21 202	97.26 90	98.87 116	98.54 117	99.27 70	98.43 117	98.34 96	97.66 145	98.92 121	97.65 156

MIMVSNet		97.24 165	97.15 150	97.36 193	99.03 186	98.52 154	98.55 161	99.73 28	94.94 170	94.94 229	97.98 139	97.37 163	93.66 214	97.60 160	97.34 163	98.23 180	96.29 188

CLD-MVS		98.48 108	98.15 105	98.86 117	99.53 117	98.35 162	98.55 161	97.83 214	96.02 147	98.97 98	99.08 93	99.75 9	99.03 83	98.10 121	97.33 164	99.28 86	98.44 107
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
RPSCF		98.84 73	98.81 45	98.89 112	99.37 140	98.95 114	98.51 163	98.85 172	97.73 57	98.33 151	98.97 104	99.14 92	98.95 85	99.18 37	98.68 77	99.31 81	98.99 53

CNVR-MVS		98.22 124	97.76 125	98.76 127	99.33 148	98.26 167	98.48 164	98.88 171	96.22 140	98.47 144	95.79 189	99.33 63	98.35 122	98.37 92	97.99 113	99.03 110	98.38 111

MVS_111021_HR		98.58 99	98.26 99	98.96 106	99.32 151	98.81 127	98.48 164	98.99 166	96.81 109	99.16 71	98.07 134	99.23 74	98.89 89	98.43 88	98.27 99	98.90 126	98.24 124

SD-MVS		98.73 85	98.54 65	98.95 107	99.14 176	98.76 131	98.46 166	99.14 150	97.71 59	98.56 131	98.06 136	99.61 36	98.85 91	98.56 79	97.74 139	99.54 38	99.32 23

NCCC		97.84 139	96.96 155	98.87 114	99.39 139	98.27 166	98.46 166	99.02 163	96.78 110	98.73 124	91.12 217	98.91 117	98.57 108	97.83 142	97.49 158	99.04 109	98.33 118

DeepPCF-MVS	96.68 10	98.20 125	98.26 99	98.12 172	97.03 234	98.11 175	98.44 168	97.70 215	96.77 111	98.52 135	98.91 105	99.17 83	98.58 107	98.41 90	98.02 110	98.46 172	98.46 104

CANet_DTU		97.65 148	97.50 137	97.82 185	99.19 171	98.08 176	98.41 169	98.67 183	94.40 183	99.16 71	98.32 123	98.69 128	93.96 212	97.87 135	97.61 150	97.51 193	97.56 160

EPMVS		93.67 219	90.82 221	96.99 204	98.62 207	96.39 210	98.40 170	99.11 155	95.54 156	97.87 175	97.14 160	81.27 225	94.97 198	88.54 230	86.80 223	92.95 224	90.06 227

TAPA-MVS	96.65 12	98.23 122	97.96 119	98.55 148	98.81 196	98.16 173	98.40 170	97.94 211	96.68 117	98.49 140	98.61 115	98.89 118	98.57 108	97.45 168	97.59 152	99.09 101	98.35 113
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
ADS-MVSNet		94.41 214	92.13 214	97.07 198	98.86 193	96.60 205	98.38 172	98.47 195	96.13 145	98.02 167	96.98 167	87.50 203	95.87 187	89.89 226	87.58 221	92.79 226	90.27 225

canonicalmvs		98.34 116	97.92 120	98.83 119	99.45 131	99.21 63	98.37 173	99.53 70	97.06 101	97.74 181	96.95 169	95.05 182	98.36 121	98.77 72	98.85 59	99.51 48	99.53 9

diffmvs		97.29 163	96.67 159	98.01 177	99.00 188	97.82 187	98.37 173	99.18 144	96.73 116	97.74 181	99.08 93	94.26 185	96.50 175	94.86 215	95.67 196	97.29 196	98.25 123

PCF-MVS	95.58 16	97.60 150	96.67 159	98.69 135	99.44 134	98.23 170	98.37 173	98.81 175	93.01 205	98.22 157	97.97 140	99.59 39	98.20 133	95.72 204	95.08 205	99.08 102	97.09 180
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
AdaColmap		97.57 155	96.57 164	98.74 128	99.25 164	98.01 179	98.36 176	98.98 167	94.44 180	98.47 144	92.44 215	97.91 155	98.62 104	98.19 110	97.74 139	98.73 152	97.28 169

OpenMVS	97.26 9	97.88 137	97.17 148	98.70 133	99.50 125	98.55 150	98.34 177	99.11 155	93.92 193	98.90 109	95.04 196	98.23 143	97.38 161	98.11 120	98.12 106	98.95 116	98.23 125

PVSNet_BlendedMVS		97.93 135	97.66 129	98.25 163	99.30 154	98.67 139	98.31 178	97.95 209	94.30 186	98.75 121	97.63 146	98.76 124	96.30 181	98.29 102	97.78 130	98.93 118	98.18 132

PVSNet_Blended		97.93 135	97.66 129	98.25 163	99.30 154	98.67 139	98.31 178	97.95 209	94.30 186	98.75 121	97.63 146	98.76 124	96.30 181	98.29 102	97.78 130	98.93 118	98.18 132

TSAR-MVS + COLMAP		97.62 149	97.31 140	97.98 178	98.47 214	97.39 199	98.29 180	98.25 200	96.68 117	97.54 189	98.87 106	98.04 151	97.08 166	96.78 187	96.26 187	98.26 178	97.12 177

MSDG		98.20 125	97.88 122	98.56 147	99.33 148	97.74 193	98.27 181	98.10 205	97.20 96	98.06 165	98.59 116	99.16 86	98.76 96	98.39 91	97.71 143	98.86 134	96.38 187

MVS_Test		97.69 145	97.15 150	98.33 159	99.27 160	98.43 160	98.25 182	99.29 127	95.00 167	97.39 195	98.86 107	98.00 152	97.14 165	95.38 207	96.22 188	98.62 160	98.15 136

PatchmatchNet		93.88 218	91.08 220	97.14 197	98.75 198	96.01 218	98.25 182	99.39 93	94.95 169	98.96 100	96.32 181	85.35 214	95.50 191	88.89 228	85.89 226	91.99 230	90.15 226
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
MVS_111021_LR		98.39 113	98.11 108	98.71 132	99.08 183	98.54 153	98.23 184	98.56 190	96.57 124	99.13 76	98.41 119	98.86 120	98.65 103	98.23 108	97.87 125	98.65 158	98.28 119

PMMVS2		96.29 188	97.05 152	95.40 225	98.32 220	96.16 212	98.18 185	97.46 216	97.20 96	84.51 239	99.60 53	98.68 130	96.37 178	98.59 78	97.38 161	97.58 192	91.76 221

Effi-MVS+		98.11 129	97.29 141	99.06 92	99.62 94	98.55 150	98.16 186	99.80 15	94.64 174	99.15 74	96.59 175	97.43 161	98.44 116	97.46 167	97.90 119	99.17 94	98.45 106

thres100view900		95.74 197	93.66 205	98.17 169	99.37 140	98.59 146	98.10 187	98.33 198	92.02 214	97.30 200	86.53 228	86.34 207	96.69 171	96.77 188	98.47 92	99.24 91	96.89 181

MDTV_nov1_ep13		94.47 212	92.15 213	97.17 196	98.54 212	96.42 209	98.10 187	98.89 170	94.49 177	98.02 167	97.41 153	86.49 204	95.56 190	90.85 225	87.95 220	93.91 218	91.45 223

abl_6				98.38 157	99.03 186	98.04 178	98.08 189	98.65 184	93.23 201	98.56 131	94.58 203	98.57 136	97.17 164			98.81 140	97.42 164

DWT-MVSNet_training		91.07 229	86.55 231	96.35 217	98.28 221	95.82 224	98.00 190	95.03 229	91.24 227	97.99 171	90.35 220	63.43 236	95.25 193	86.06 232	86.62 224	93.55 220	92.30 218

FMVSNet3		96.85 173	96.67 159	97.06 199	97.56 230	99.01 102	97.99 191	99.33 119	94.09 189	95.60 221	98.17 131	95.97 176	93.26 217	94.76 216	96.22 188	98.59 163	98.46 104

TAMVS		96.95 172	96.94 156	96.97 205	99.07 185	97.67 196	97.98 192	97.12 221	95.04 164	95.41 224	99.27 79	95.57 180	94.09 210	97.32 176	97.11 171	98.16 183	96.59 185

pmmvs3		96.30 187	95.87 182	96.80 209	97.66 229	96.48 207	97.93 193	93.80 231	93.40 200	98.54 134	98.27 126	97.50 160	97.37 163	97.49 166	93.11 213	95.52 212	94.85 204

MS-PatchMatch		97.60 150	97.22 146	98.04 176	98.67 204	97.18 201	97.91 194	98.28 199	95.82 152	98.34 150	97.66 145	98.38 139	97.77 149	97.10 183	97.25 167	97.27 197	97.18 176

GA-MVS		96.84 174	95.86 183	97.98 178	99.16 175	98.29 163	97.91 194	98.64 186	95.14 162	97.71 183	98.04 138	88.90 196	96.50 175	96.41 193	96.61 183	97.97 187	97.60 157

Effi-MVS+-dtu		97.78 140	97.37 139	98.26 162	99.25 164	98.50 155	97.89 196	99.19 143	94.51 176	98.16 160	95.93 188	98.80 123	95.97 185	98.27 107	97.38 161	99.10 97	98.23 125

HQP-MVS		97.58 154	96.65 163	98.66 139	99.30 154	97.99 181	97.88 197	98.65 184	94.58 175	98.66 125	94.65 199	99.15 90	98.59 106	96.10 197	95.59 197	98.90 126	98.50 102

CNLPA		97.75 141	97.26 142	98.32 161	98.58 208	97.86 186	97.80 198	98.09 206	96.49 129	98.49 140	96.15 184	98.08 148	98.35 122	98.00 124	97.03 174	98.61 161	97.21 175

ambc			97.89 121		99.45 131	97.88 185	97.78 199		97.27 88	99.80 3	98.99 103	98.48 138	98.55 110	97.80 144	96.68 180	98.54 166	98.10 138

RPMNet		94.72 207	92.01 215	97.88 182	98.56 210	95.85 221	97.78 199	99.70 34	91.27 225	98.33 151	93.69 207	81.88 222	94.91 199	92.60 224	94.34 210	98.01 186	94.46 206

dps		92.35 225	88.78 228	96.52 215	98.21 224	95.94 220	97.78 199	98.38 197	89.88 233	96.81 211	95.07 195	75.31 229	94.70 202	88.62 229	86.21 225	93.21 223	90.41 224

TSAR-MVS + GP.		98.54 104	98.29 98	98.82 122	99.28 159	98.59 146	97.73 202	99.24 136	95.93 149	98.59 129	99.07 95	99.17 83	98.86 90	98.44 85	98.10 107	99.26 88	98.72 85

PLC	95.63 15	97.73 144	97.01 154	98.57 146	99.10 180	97.80 189	97.72 203	98.77 177	96.34 135	98.38 147	93.46 210	98.06 149	98.66 102	97.90 132	97.65 147	98.77 150	97.90 148
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
FPMVS		96.97 171	97.20 147	96.70 211	97.75 227	96.11 215	97.72 203	95.47 226	97.13 98	98.02 167	97.57 148	96.67 171	92.97 218	99.00 56	98.34 97	98.28 177	95.58 196

new_pmnet		96.59 179	96.40 169	96.81 208	98.24 223	95.46 225	97.71 205	94.75 230	96.92 102	96.80 212	99.23 84	97.81 157	96.69 171	96.58 191	95.16 203	96.69 203	93.64 212

EPNet		96.44 183	96.08 177	96.86 206	99.32 151	97.15 202	97.69 206	99.32 124	93.67 196	98.11 163	95.64 191	93.44 188	89.07 231	96.86 186	96.83 177	97.67 189	98.97 55
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
MAR-MVS		97.12 167	96.28 173	98.11 173	98.94 191	97.22 200	97.65 207	99.38 100	90.93 229	98.15 161	95.17 194	97.13 166	96.48 177	97.71 151	97.40 160	98.06 184	98.40 110
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
CR-MVSNet		95.38 201	93.01 208	98.16 171	98.63 206	95.85 221	97.64 208	99.78 19	91.27 225	98.50 137	96.84 170	82.16 221	96.34 179	94.40 217	95.50 198	98.05 185	95.04 202

Patchmtry						96.05 217	97.64 208	99.78 19		98.50 137

tpmp4_e23		92.43 224	88.82 227	96.64 213	98.46 215	95.17 228	97.61 210	98.85 172	92.42 206	98.18 158	93.03 212	74.92 230	93.80 213	88.91 227	84.60 228	92.95 224	92.66 217

CostFormer		92.75 221	89.49 224	96.55 214	98.78 197	95.83 223	97.55 211	98.59 188	91.83 222	97.34 198	96.31 182	78.53 226	94.50 204	86.14 231	84.92 227	92.54 227	92.84 215

Fast-Effi-MVS+-dtu		96.99 170	96.46 167	97.61 189	98.98 189	97.89 184	97.54 212	99.76 22	93.43 199	96.55 213	94.93 197	98.06 149	94.32 208	96.93 185	96.50 185	98.53 167	97.47 161

gg-mvs-nofinetune		96.77 176	96.52 166	97.06 199	99.66 73	97.82 187	97.54 212	99.86 9	98.69 17	98.61 127	99.94 4	89.62 194	88.37 233	97.55 163	96.67 181	98.30 176	95.35 199

FMVSNet5		94.57 210	92.77 209	96.67 212	97.88 225	98.72 136	97.54 212	98.70 182	88.64 235	95.11 227	86.90 226	81.77 223	93.27 216	97.92 131	98.07 109	97.50 194	97.34 167

tpm		93.89 217	91.21 218	97.03 201	98.36 218	96.07 216	97.53 215	99.65 43	92.24 209	98.64 126	97.23 157	74.67 231	94.64 203	92.68 223	90.73 216	93.37 221	94.82 205

test0.0.03 1		95.81 196	95.77 185	95.85 224	99.20 168	98.15 174	97.49 216	98.50 192	92.24 209	92.74 237	96.82 171	92.70 190	88.60 232	97.31 178	97.01 176	98.57 165	96.19 191

PatchMatch-RL		97.24 165	96.45 168	98.17 169	98.70 202	97.57 197	97.31 217	98.48 194	94.42 182	98.39 146	95.74 190	96.35 175	97.88 145	97.75 149	97.48 159	98.24 179	95.87 194

PMMVS		96.47 182	95.81 184	97.23 195	97.38 232	95.96 219	97.31 217	96.91 223	93.21 202	97.93 174	97.14 160	97.64 159	95.70 188	95.24 209	96.18 191	98.17 182	95.33 200

IB-MVS	95.85 14	95.87 195	94.88 190	97.02 202	99.09 181	98.25 168	97.16 219	97.38 218	91.97 221	97.77 177	83.61 236	97.29 164	92.03 226	97.16 181	97.66 145	98.66 156	98.20 131
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
tpmrst		92.45 223	89.48 225	95.92 222	98.43 217	95.03 229	97.14 220	97.92 212	94.16 188	97.56 188	97.86 141	81.63 224	93.56 215	85.89 233	82.86 229	90.91 234	88.95 232

MSLP-MVS++		97.99 131	97.64 132	98.40 155	98.91 192	98.47 157	97.12 221	98.78 176	96.49 129	98.48 142	93.57 209	99.12 96	98.51 114	98.31 100	98.58 85	98.58 164	98.95 61

tpm cat1		91.52 228	87.70 230	95.97 221	98.33 219	94.98 230	97.06 222	98.03 208	92.11 213	98.03 166	94.77 198	77.19 228	92.71 220	83.56 234	82.24 231	91.67 231	89.04 231

EPNet_dtu		96.31 186	95.96 180	96.72 210	99.18 172	95.39 226	97.03 223	99.13 154	93.02 204	99.35 41	97.23 157	97.07 167	90.70 228	95.74 203	95.08 205	94.94 214	98.16 135
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
test12356		95.71 198	95.55 187	95.89 223	98.27 222	96.48 207	96.90 224	97.35 219	92.13 212	95.64 220	99.13 90	97.97 153	92.34 222	96.94 184	96.55 184	94.87 215	89.61 228

testus		96.13 194	95.13 189	97.28 194	99.13 178	97.00 203	96.84 225	97.89 213	90.48 230	97.40 193	93.60 208	96.47 173	95.39 192	96.21 194	96.19 190	97.05 200	95.99 192

MVSTER		95.38 201	93.99 201	97.01 203	98.83 195	98.95 114	96.62 226	99.14 150	92.17 211	97.44 191	97.29 155	77.88 227	91.63 227	97.45 168	96.18 191	98.41 173	97.99 144

test-LLR		94.79 206	93.71 203	96.06 220	99.20 168	96.16 212	96.31 227	98.50 192	89.98 231	94.08 230	97.01 164	86.43 205	92.20 224	96.76 189	95.31 200	96.05 208	94.31 207

TESTMET0.1,1		94.44 213	93.71 203	95.30 227	97.84 226	96.16 212	96.31 227	95.32 228	89.98 231	94.08 230	97.01 164	86.43 205	92.20 224	96.76 189	95.31 200	96.05 208	94.31 207

test-mter		94.62 208	94.02 200	95.32 226	97.72 228	96.75 204	96.23 229	95.67 225	89.83 234	93.23 236	96.99 166	85.94 211	92.66 221	97.32 176	96.11 193	96.44 204	95.22 201

test2356		92.46 222	88.72 229	96.82 207	98.48 213	95.34 227	96.22 230	98.09 206	87.46 236	96.01 216	92.82 214	64.42 235	95.10 196	94.08 219	94.05 211	97.02 201	92.87 214

EMVS		91.84 227	89.39 226	94.70 229	98.44 216	90.84 235	95.27 231	93.53 233	95.18 161	95.26 226	95.62 192	87.59 202	94.77 201	94.87 213	80.72 233	90.95 233	80.88 234

E-PMN		92.28 226	90.12 222	94.79 228	98.56 210	90.90 234	95.16 232	93.68 232	95.36 158	95.10 228	96.56 176	89.05 195	95.24 194	95.21 210	81.84 232	90.98 232	81.94 233

CMPMVS	74.71 19	96.17 191	96.06 178	96.30 218	97.41 231	94.52 231	94.83 233	95.46 227	91.57 223	97.26 203	94.45 204	98.33 141	94.98 197	98.28 105	97.59 152	97.86 188	97.68 155
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
MVS-HIRNet		94.86 205	93.83 202	96.07 219	97.07 233	94.00 232	94.31 234	99.17 146	91.23 228	98.17 159	98.69 112	97.43 161	95.66 189	94.05 220	91.92 215	92.04 229	89.46 229

MVE	82.47 18	93.12 220	94.09 196	91.99 231	90.79 236	82.50 238	93.93 235	96.30 224	96.06 146	88.81 238	98.19 129	96.38 174	97.56 153	97.24 180	95.18 202	84.58 236	93.07 213
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
testpf		87.81 230	83.90 232	92.37 230	96.76 235	88.65 236	93.04 236	98.24 201	85.20 237	95.28 225	86.82 227	72.43 233	82.35 234	82.62 235	82.30 230	88.55 235	89.29 230

DeepMVS_CX						87.86 237	92.27 237	61.98 234	93.64 197	93.62 233	91.17 216	91.67 192	94.90 200	95.99 200		92.48 228	94.18 209

GG-mvs-BLEND		65.66 232	92.62 211	34.20 234	1.45 240	93.75 233	85.40 238	1.64 238	91.37 224	17.21 241	87.25 224	94.78 183	3.25 238	95.64 205	93.80 212	96.27 205	91.74 222

tmp_tt				65.28 233	82.24 237	71.50 239	70.81 239	23.21 235	96.14 143	81.70 240	85.98 233	92.44 191	49.84 235	95.81 201	94.36 209	83.86 237

Patchmatch-RL test							32.47 240

testmvs		9.73 233	13.38 234	5.48 236	3.62 238	4.12 240	6.40 241	3.19 237	14.92 238	7.68 243	22.10 237	13.89 243	6.83 236	13.47 236	10.38 235	5.14 239	14.81 235

test123		9.37 234	12.26 235	6.00 235	3.32 239	4.06 241	6.39 242	3.41 236	13.20 239	10.48 242	16.43 238	16.22 242	6.76 237	11.37 237	10.40 234	5.62 238	14.10 237

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										99.19 65		99.68 22

MTMP										99.20 63		99.54 42

mPP-MVS					99.75 46							99.49 50

NP-MVS									93.07 203

High-resolution multi-view benchmark