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.

gg-mvs-nofinetune		98.40 186	98.26 178	98.57 208	99.83 98	98.86 183	98.77 211	99.97 1	99.57 34	99.99 1	99.99 1	93.81 200	93.50 231	98.91 165	98.20 182	99.33 180	98.52 195

Anonymous20231211		99.90 1	99.87 1	99.92 4	99.98 1	99.91 1	99.92 2	99.97 1	99.86 2	99.98 2	99.82 64	100.00 1	99.70 39	99.86 17	99.79 12	99.96 3	99.87 12

Gipuma		99.55 68	99.23 92	99.91 9	99.87 50	99.52 72	99.86 14	99.93 36	99.87 1	99.96 3	96.72 217	99.55 121	99.97 1	99.77 31	99.46 51	99.87 47	99.74 39
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
SixPastTwentyTwo		99.89 3	99.85 7	99.93 1	99.97 2	99.88 2	99.92 2	99.97 1	99.66 15	99.94 4	99.94 15	99.74 95	99.81 9	99.97 2	99.89 1	99.96 3	99.89 5

WR-MVS		99.79 12	99.68 17	99.91 9	99.95 12	99.83 3	99.87 13	99.96 15	99.39 57	99.93 5	99.87 50	99.29 153	99.77 17	99.83 21	99.72 21	99.97 1	99.82 17

EG-PatchMatch MVS		99.59 51	99.49 47	99.70 76	99.82 104	99.26 138	99.39 132	99.83 98	98.99 105	99.93 5	99.54 101	99.92 48	99.51 81	99.78 29	99.50 45	99.73 100	99.41 127

MIMVSNet1		99.79 12	99.75 13	99.84 25	99.89 39	99.83 3	99.84 18	99.89 56	99.31 64	99.93 5	99.92 26	99.97 10	99.68 40	99.89 9	99.64 28	99.82 74	99.66 57

LTVRE_ROB	99.39 1	99.90 1	99.87 1	99.93 1	99.97 2	99.82 7	99.91 6	99.92 43	99.75 8	99.93 5	99.89 42	100.00 1	99.87 2	99.93 4	99.82 6	99.96 3	99.90 2
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
ambc			98.83 149		99.72 148	98.52 204	98.84 205		98.96 109	99.92 9	99.34 126	99.74 95	99.04 145	98.68 192	97.57 205	99.46 161	98.99 184

v748		99.89 3	99.87 1	99.92 4	99.96 9	99.80 12	99.91 6	99.95 29	99.77 6	99.92 9	99.96 5	99.93 39	99.81 9	99.92 7	99.82 6	99.96 3	99.90 2

TransMVSNet (Re)		99.72 25	99.59 33	99.88 16	99.95 12	99.76 19	99.88 11	99.94 32	99.58 31	99.92 9	99.90 39	98.55 171	99.65 55	99.89 9	99.76 15	99.95 10	99.70 50

DTE-MVSNet		99.75 18	99.61 27	99.92 4	99.95 12	99.81 10	99.86 14	99.96 15	99.18 81	99.92 9	99.66 83	99.45 137	99.85 3	99.80 26	99.56 34	99.96 3	99.79 24

Anonymous20231206		99.48 76	99.31 81	99.69 80	99.79 116	99.57 55	99.63 78	99.79 127	98.88 118	99.91 13	99.72 76	99.93 39	99.59 66	99.24 103	98.63 158	99.43 168	99.18 160

PVSNet_Blended_VisFu		99.66 37	99.64 20	99.67 83	99.91 31	99.71 28	99.61 84	99.79 127	99.41 53	99.91 13	99.85 57	99.61 110	99.00 147	99.67 42	99.42 55	99.81 78	99.81 20

pmmvs6		99.88 8	99.87 1	99.89 13	99.97 2	99.76 19	99.89 9	99.96 15	99.82 3	99.90 15	99.92 26	99.95 23	99.68 40	99.93 4	99.88 2	99.95 10	99.86 13

v52		99.89 3	99.85 7	99.92 4	99.97 2	99.80 12	99.92 2	99.97 1	99.78 4	99.90 15	99.96 5	99.85 76	99.82 7	99.88 12	99.82 6	99.96 3	99.89 5

V4		99.89 3	99.85 7	99.92 4	99.97 2	99.80 12	99.92 2	99.97 1	99.78 4	99.90 15	99.96 5	99.84 78	99.82 7	99.88 12	99.82 6	99.96 3	99.89 5

DeepC-MVS	99.05 5	99.74 19	99.64 20	99.84 25	99.90 34	99.39 99	99.79 32	99.81 118	99.69 11	99.90 15	99.87 50	99.98 5	99.81 9	99.62 49	99.32 62	99.83 70	99.65 61
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
v7n		99.89 3	99.86 5	99.93 1	99.97 2	99.83 3	99.93 1	99.96 15	99.77 6	99.89 19	99.99 1	99.86 70	99.84 5	99.89 9	99.81 10	99.97 1	99.88 9

UniMVSNet_NR-MVSNet		99.41 89	99.12 117	99.76 57	99.86 66	99.48 83	99.50 112	99.81 118	98.84 124	99.89 19	99.45 115	98.32 177	99.59 66	99.22 109	98.89 121	99.90 28	99.63 65

PEN-MVS		99.77 14	99.65 19	99.91 9	99.95 12	99.80 12	99.86 14	99.97 1	99.08 95	99.89 19	99.69 81	99.68 103	99.84 5	99.81 25	99.64 28	99.95 10	99.81 20

DU-MVS		99.48 76	99.26 87	99.75 63	99.85 79	99.38 104	99.50 112	99.81 118	98.86 120	99.89 19	99.51 110	98.98 162	99.59 66	99.46 61	98.97 110	99.87 47	99.63 65

UniMVSNet (Re)		99.50 72	99.29 83	99.75 63	99.86 66	99.47 84	99.51 108	99.82 106	98.90 116	99.89 19	99.64 87	99.00 161	99.55 72	99.32 86	99.08 90	99.90 28	99.59 71

tfpnnormal		99.74 19	99.63 22	99.86 19	99.93 25	99.75 23	99.80 30	99.89 56	99.31 64	99.88 24	99.43 116	99.66 106	99.77 17	99.80 26	99.71 22	99.92 23	99.76 33

test20.03		99.68 34	99.60 31	99.76 57	99.91 31	99.70 33	99.68 71	99.87 62	99.05 100	99.88 24	99.92 26	99.88 64	99.50 85	99.77 31	99.42 55	99.75 92	99.49 107

Baseline_NR-MVSNet		99.62 41	99.48 48	99.78 49	99.85 79	99.76 19	99.59 89	99.82 106	98.84 124	99.88 24	99.91 35	99.04 160	99.61 63	99.46 61	99.78 13	99.94 17	99.60 70

TranMVSNet+NR-MVSNet		99.59 51	99.42 63	99.80 39	99.87 50	99.55 62	99.64 76	99.86 66	99.05 100	99.88 24	99.72 76	99.33 151	99.64 58	99.47 59	99.14 81	99.91 25	99.67 55

TDRefinement		99.81 11	99.76 12	99.86 19	99.83 98	99.53 66	99.89 9	99.91 48	99.73 9	99.88 24	99.83 62	99.96 14	99.76 19	99.91 8	99.81 10	99.86 53	99.59 71

pm-mvs1		99.77 14	99.69 16	99.86 19	99.94 22	99.68 36	99.84 18	99.93 36	99.59 29	99.87 29	99.92 26	99.21 156	99.65 55	99.88 12	99.77 14	99.93 19	99.78 27

v2v482		99.56 66	99.35 77	99.81 35	99.87 50	99.35 117	99.75 46	99.85 78	99.56 36	99.87 29	99.95 10	99.44 139	99.66 52	98.91 165	98.76 144	99.86 53	99.45 115

NR-MVSNet		99.52 69	99.29 83	99.80 39	99.96 9	99.38 104	99.55 98	99.81 118	98.86 120	99.87 29	99.51 110	98.81 167	99.72 34	99.86 17	99.04 98	99.89 33	99.54 91

USDC		99.29 117	98.98 136	99.65 86	99.72 148	98.87 182	99.47 121	99.66 175	99.35 61	99.87 29	99.58 97	99.87 69	99.51 81	98.85 177	97.93 196	99.65 116	98.38 198

ACMH	99.11 4	99.72 25	99.63 22	99.84 25	99.87 50	99.59 51	99.83 21	99.88 60	99.46 47	99.87 29	99.66 83	99.95 23	99.76 19	99.73 36	99.47 49	99.84 60	99.52 101
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
new-patchmatchnet		98.49 182	97.60 190	99.53 112	99.90 34	99.55 62	99.77 37	99.48 199	99.67 13	99.86 34	99.98 3	99.98 5	99.50 85	96.90 221	91.52 222	98.67 207	95.62 223

v144192		99.58 55	99.39 70	99.80 39	99.87 50	99.44 88	99.77 37	99.84 90	99.64 18	99.86 34	99.93 19	99.35 149	99.72 34	98.92 162	98.82 135	99.74 96	99.66 57

v1921920		99.59 51	99.40 66	99.82 31	99.88 43	99.45 86	99.81 27	99.83 98	99.65 16	99.86 34	99.95 10	99.29 153	99.75 21	98.98 154	98.86 129	99.78 83	99.59 71

v1240		99.58 55	99.38 75	99.82 31	99.89 39	99.49 82	99.82 25	99.83 98	99.63 20	99.86 34	99.96 5	98.92 165	99.75 21	99.15 128	98.96 112	99.76 89	99.56 85

PS-CasMVS		99.73 21	99.59 33	99.90 12	99.95 12	99.80 12	99.85 17	99.97 1	98.95 110	99.86 34	99.73 73	99.36 146	99.81 9	99.83 21	99.67 25	99.95 10	99.83 16

no-one		99.73 21	99.70 15	99.76 57	99.77 127	99.58 53	99.76 40	99.90 55	99.08 95	99.86 34	99.90 39	99.98 5	99.66 52	99.98 1	99.73 19	99.59 143	99.67 55

CLD-MVS		99.30 113	99.01 133	99.63 92	99.75 136	98.89 181	99.35 143	99.60 179	98.53 157	99.86 34	99.57 98	99.94 33	99.52 80	98.96 158	98.10 189	99.70 108	99.08 172
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
pmmvs-eth3d		99.61 43	99.48 48	99.75 63	99.87 50	99.30 130	99.75 46	99.89 56	99.23 71	99.85 41	99.88 48	99.97 10	99.49 89	99.46 61	99.01 102	99.68 110	99.52 101

v1192		99.60 49	99.41 64	99.82 31	99.89 39	99.43 92	99.81 27	99.84 90	99.63 20	99.85 41	99.95 10	99.35 149	99.72 34	99.01 148	98.90 120	99.82 74	99.58 80

MDA-MVSNet-bldmvs		99.11 137	99.11 119	99.12 180	99.91 31	99.38 104	99.77 37	98.72 224	99.31 64	99.85 41	99.43 116	98.26 178	99.48 93	99.85 19	98.47 167	96.99 219	99.08 172

Vis-MVSNet		99.76 16	99.78 11	99.75 63	99.92 27	99.77 18	99.83 21	99.85 78	99.43 50	99.85 41	99.84 60	100.00 1	99.13 137	99.83 21	99.66 26	99.90 28	99.90 2
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
gm-plane-assit		96.82 214	94.84 220	99.13 178	99.95 12	99.78 17	99.69 70	99.92 43	99.19 79	99.84 45	99.92 26	72.93 237	96.44 216	98.21 206	97.01 210	98.92 201	96.87 219

v1144		99.61 43	99.43 60	99.82 31	99.88 43	99.41 96	99.76 40	99.86 66	99.64 18	99.84 45	99.95 10	99.49 133	99.74 26	99.00 150	98.93 117	99.84 60	99.58 80

v1141		99.58 55	99.39 70	99.80 39	99.87 50	99.39 99	99.74 54	99.85 78	99.58 31	99.84 45	99.92 26	99.49 133	99.68 40	98.98 154	98.83 132	99.84 60	99.52 101

divwei89l23v2f112		99.58 55	99.39 70	99.80 39	99.87 50	99.39 99	99.74 54	99.85 78	99.57 34	99.84 45	99.92 26	99.48 135	99.67 44	98.98 154	98.83 132	99.84 60	99.52 101

v1		99.58 55	99.39 70	99.80 39	99.87 50	99.39 99	99.74 54	99.85 78	99.58 31	99.84 45	99.92 26	99.51 128	99.67 44	98.98 154	98.82 135	99.84 60	99.52 101

V9		99.71 29	99.59 33	99.84 25	99.86 66	99.69 35	99.78 35	99.96 15	99.61 24	99.84 45	99.93 19	99.61 110	99.73 30	99.34 83	99.17 75	99.88 37	99.78 27

CP-MVSNet		99.68 34	99.51 45	99.89 13	99.95 12	99.76 19	99.83 21	99.96 15	98.83 126	99.84 45	99.65 86	99.09 159	99.80 13	99.78 29	99.62 32	99.95 10	99.82 17

WR-MVS_H		99.73 21	99.61 27	99.88 16	99.95 12	99.82 7	99.83 21	99.96 15	99.01 103	99.84 45	99.71 79	99.41 143	99.74 26	99.77 31	99.70 23	99.95 10	99.82 17

MDTV_nov1_ep13_2view		98.73 172	98.31 177	99.22 168	99.75 136	99.24 146	99.75 46	99.93 36	99.31 64	99.84 45	99.86 56	99.81 84	99.31 113	97.40 218	94.77 216	96.73 221	97.81 211

anonymousdsp		99.87 9	99.86 5	99.88 16	99.95 12	99.75 23	99.90 8	99.96 15	99.69 11	99.83 54	99.96 5	99.99 4	99.74 26	99.95 3	99.83 3	99.91 25	99.88 9

v148		99.58 55	99.43 60	99.76 57	99.87 50	99.40 98	99.76 40	99.85 78	99.48 45	99.83 54	99.82 64	99.83 81	99.51 81	99.20 115	98.82 135	99.75 92	99.45 115

v15		99.67 36	99.54 41	99.83 30	99.86 66	99.67 39	99.76 40	99.95 29	99.59 29	99.83 54	99.93 19	99.55 121	99.71 38	99.23 106	99.05 95	99.87 47	99.75 36

v13		99.73 21	99.63 22	99.85 22	99.87 50	99.71 28	99.80 30	99.96 15	99.62 23	99.83 54	99.93 19	99.66 106	99.75 21	99.41 70	99.26 67	99.89 33	99.80 23

v12		99.72 25	99.61 27	99.85 22	99.86 66	99.70 33	99.79 32	99.96 15	99.61 24	99.83 54	99.93 19	99.61 110	99.74 26	99.38 73	99.22 69	99.89 33	99.79 24

v11		99.72 25	99.62 25	99.85 22	99.87 50	99.71 28	99.81 27	99.96 15	99.63 20	99.83 54	99.97 4	99.58 117	99.75 21	99.33 84	99.33 60	99.87 47	99.79 24

V14		99.69 32	99.56 38	99.84 25	99.86 66	99.68 36	99.78 35	99.96 15	99.60 28	99.83 54	99.93 19	99.58 117	99.72 34	99.28 97	99.11 87	99.88 37	99.77 29

LP		97.43 210	96.28 212	98.77 202	99.69 157	98.92 177	99.49 117	99.70 163	98.53 157	99.82 61	99.12 146	95.67 193	97.30 200	94.65 225	91.76 220	96.65 223	95.34 225

TinyColmap		99.21 123	98.89 144	99.59 100	99.61 177	98.61 200	99.47 121	99.67 171	99.02 102	99.82 61	99.15 142	99.74 95	99.35 103	99.17 125	98.33 177	99.63 127	98.22 204

COLMAP_ROB	99.18 2	99.70 30	99.60 31	99.81 35	99.84 87	99.37 110	99.76 40	99.84 90	99.54 40	99.82 61	99.64 87	99.95 23	99.75 21	99.79 28	99.56 34	99.83 70	99.37 140
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
APDe-MVS		99.60 49	99.48 48	99.73 71	99.85 79	99.51 79	99.75 46	99.85 78	99.17 82	99.81 64	99.56 99	99.94 33	99.44 98	99.42 69	99.22 69	99.67 112	99.54 91

test12356		99.12 136	99.03 129	99.23 163	99.78 121	98.95 176	99.10 179	99.72 158	98.26 177	99.81 64	99.87 50	99.20 157	98.06 184	99.47 59	98.80 141	98.91 202	98.67 191

SteuartSystems-ACMMP		99.47 79	99.22 94	99.76 57	99.88 43	99.36 113	99.65 75	99.84 90	98.47 160	99.80 66	98.68 174	99.96 14	99.68 40	99.37 75	99.06 92	99.72 104	99.66 57
Skip Steuart: Steuart Systems R&D Blog.
ACMH+	98.94 6	99.69 32	99.59 33	99.81 35	99.88 43	99.41 96	99.75 46	99.86 66	99.43 50	99.80 66	99.54 101	99.97 10	99.73 30	99.82 24	99.52 44	99.85 57	99.43 122

3Dnovator	99.16 3	99.42 87	99.22 94	99.65 86	99.78 121	99.13 161	99.50 112	99.85 78	99.40 55	99.80 66	98.59 177	99.79 91	99.30 115	99.20 115	99.06 92	99.71 106	99.35 143

testmv		99.39 94	99.19 102	99.62 97	99.84 87	99.38 104	99.37 138	99.86 66	98.47 160	99.79 69	99.82 64	99.39 145	99.63 60	99.30 89	98.70 151	99.21 190	99.28 151

test1235678		99.39 94	99.20 99	99.62 97	99.84 87	99.38 104	99.38 136	99.86 66	98.47 160	99.79 69	99.82 64	99.41 143	99.63 60	99.30 89	98.71 149	99.21 190	99.28 151

LGP-MVS_train		99.46 83	99.18 105	99.78 49	99.87 50	99.25 141	99.71 68	99.87 62	98.02 189	99.79 69	98.90 163	99.96 14	99.66 52	99.49 55	99.17 75	99.79 82	99.49 107

PVSNet_BlendedMVS		99.20 126	99.17 109	99.23 163	99.69 157	99.33 124	99.04 182	99.13 219	98.41 169	99.79 69	99.33 127	99.36 146	98.10 180	99.29 93	98.87 126	99.65 116	99.56 85

PVSNet_Blended		99.20 126	99.17 109	99.23 163	99.69 157	99.33 124	99.04 182	99.13 219	98.41 169	99.79 69	99.33 127	99.36 146	98.10 180	99.29 93	98.87 126	99.65 116	99.56 85

ACMP	98.32 13	99.44 85	99.18 105	99.75 63	99.83 98	99.18 154	99.64 76	99.83 98	98.81 128	99.79 69	98.42 187	99.96 14	99.64 58	99.46 61	98.98 109	99.74 96	99.44 118
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
PHI-MVS		99.33 106	99.19 102	99.49 124	99.69 157	99.25 141	99.27 157	99.59 184	98.44 166	99.78 75	99.15 142	99.92 48	98.95 156	99.39 72	99.04 98	99.64 125	99.18 160

FC-MVSNet-test		99.84 10	99.80 10	99.89 13	99.96 9	99.83 3	99.84 18	99.95 29	99.37 58	99.77 76	99.95 10	99.96 14	99.85 3	99.93 4	99.83 3	99.95 10	99.72 46

v8		99.61 43	99.45 56	99.79 46	99.80 108	99.59 51	99.73 57	99.93 36	99.48 45	99.77 76	99.90 39	99.48 135	99.67 44	99.11 134	98.89 121	99.84 60	99.73 42

DeepC-MVS_fast	98.69 9	99.32 108	99.13 115	99.53 112	99.63 171	98.78 187	99.53 102	99.33 214	99.08 95	99.77 76	99.18 141	99.89 59	99.29 116	99.00 150	98.70 151	99.65 116	99.30 149
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
v1neww		99.57 62	99.40 66	99.77 53	99.80 108	99.34 119	99.72 62	99.82 106	99.49 42	99.76 79	99.89 42	99.50 130	99.67 44	99.10 142	98.89 121	99.84 60	99.59 71

v7new		99.57 62	99.40 66	99.77 53	99.80 108	99.34 119	99.72 62	99.82 106	99.49 42	99.76 79	99.89 42	99.50 130	99.67 44	99.10 142	98.89 121	99.84 60	99.59 71

v7		99.61 43	99.46 55	99.79 46	99.83 98	99.37 110	99.75 46	99.84 90	99.56 36	99.76 79	99.94 15	99.60 114	99.73 30	99.11 134	99.01 102	99.85 57	99.63 65

v6		99.57 62	99.40 66	99.77 53	99.80 108	99.34 119	99.72 62	99.82 106	99.49 42	99.76 79	99.89 42	99.52 126	99.67 44	99.10 142	98.89 121	99.84 60	99.59 71

v10		99.65 38	99.51 45	99.81 35	99.83 98	99.61 46	99.75 46	99.94 32	99.56 36	99.76 79	99.94 15	99.60 114	99.73 30	99.11 134	99.01 102	99.85 57	99.74 39

pmmvs3		98.85 164	98.60 160	99.13 178	99.66 166	98.72 193	99.37 138	99.06 221	98.44 166	99.76 79	99.74 71	99.55 121	99.15 133	99.04 146	96.00 215	97.80 214	98.72 190

OpenMVS	98.82 8	99.17 129	98.85 148	99.53 112	99.75 136	99.06 167	99.36 140	99.82 106	98.28 176	99.76 79	98.47 183	99.61 110	98.91 157	98.80 185	98.70 151	99.60 137	99.04 180

MVS_0304		99.36 99	99.35 77	99.37 143	99.85 79	99.36 113	99.39 132	99.56 186	99.36 60	99.75 86	99.23 137	99.90 56	97.97 189	99.00 150	98.83 132	99.69 109	99.77 29

V42		99.57 62	99.41 64	99.75 63	99.84 87	99.37 110	99.73 57	99.83 98	99.41 53	99.75 86	99.89 42	99.42 141	99.60 65	99.15 128	98.96 112	99.76 89	99.65 61

EU-MVSNet		99.76 16	99.74 14	99.78 49	99.82 104	99.81 10	99.88 11	99.87 62	99.31 64	99.75 86	99.91 35	99.76 94	99.78 15	99.84 20	99.74 18	99.56 150	99.81 20

PGM-MVS		99.32 108	98.99 134	99.71 73	99.86 66	99.31 129	99.59 89	99.86 66	97.51 204	99.75 86	98.23 191	99.94 33	99.53 77	99.29 93	99.08 90	99.65 116	99.54 91

PM-MVS		99.49 75	99.43 60	99.57 106	99.76 132	99.34 119	99.53 102	99.77 139	98.93 114	99.75 86	99.46 114	99.83 81	99.11 139	99.72 37	99.29 64	99.49 159	99.46 114

IterMVS		99.08 140	98.90 143	99.29 154	99.87 50	99.53 66	99.52 105	99.77 139	98.94 112	99.75 86	99.91 35	97.52 186	98.72 165	98.86 175	98.14 186	98.09 212	99.43 122
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
CANet		99.36 99	99.39 70	99.34 150	99.80 108	99.35 117	99.41 130	99.47 200	99.20 76	99.74 92	99.54 101	99.68 103	98.05 186	99.23 106	98.97 110	99.57 147	99.73 42

CANet_DTU		99.03 149	99.18 105	98.87 198	99.58 188	99.03 168	99.18 165	99.41 206	98.65 142	99.74 92	99.55 100	99.71 100	96.13 221	99.19 120	98.92 118	99.17 193	99.18 160

TSAR-MVS + GP.		99.33 106	99.17 109	99.51 118	99.71 151	99.00 171	98.84 205	99.71 161	98.23 178	99.74 92	99.53 107	99.90 56	99.35 103	99.38 73	98.85 130	99.72 104	99.31 147

EPNet_dtu		98.09 191	98.25 179	97.91 224	99.58 188	98.02 225	98.19 228	99.67 171	97.94 193	99.74 92	99.07 153	98.71 169	93.40 232	97.50 216	97.09 208	96.89 220	99.44 118
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
CHOSEN 1792x2688		99.65 38	99.55 39	99.77 53	99.93 25	99.60 48	99.79 32	99.92 43	99.73 9	99.74 92	99.93 19	99.98 5	99.80 13	98.83 182	99.01 102	99.45 163	99.76 33

CP-MVS		99.41 89	99.20 99	99.65 86	99.80 108	99.23 148	99.44 125	99.75 155	98.60 151	99.74 92	98.66 175	99.93 39	99.48 93	99.33 84	99.16 77	99.73 100	99.48 110

pmmvs5		99.58 55	99.47 52	99.70 76	99.84 87	99.50 80	99.58 93	99.80 124	98.98 108	99.73 98	99.92 26	99.81 84	99.49 89	99.28 97	99.05 95	99.77 87	99.73 42

v17		99.62 41	99.48 48	99.79 46	99.80 108	99.60 48	99.73 57	99.94 32	99.46 47	99.73 98	99.88 48	99.52 126	99.67 44	99.16 127	98.96 112	99.84 60	99.75 36

pmmvs4		99.34 104	99.15 112	99.57 106	99.77 127	98.90 178	99.51 108	99.77 139	99.07 98	99.73 98	99.72 76	99.84 78	99.07 141	98.85 177	98.39 173	99.55 154	99.27 153

ACMMPR		99.51 70	99.32 80	99.72 72	99.87 50	99.33 124	99.61 84	99.85 78	99.19 79	99.73 98	98.73 171	99.95 23	99.61 63	99.35 80	99.14 81	99.66 114	99.58 80

DELS-MVS		99.42 87	99.53 43	99.29 154	99.52 192	99.43 92	99.42 127	99.28 216	99.16 86	99.72 102	99.82 64	99.97 10	98.17 176	99.56 51	99.16 77	99.65 116	99.59 71
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		99.41 89	99.21 98	99.64 91	99.78 121	99.16 155	99.51 108	99.85 78	99.20 76	99.72 102	99.43 116	99.81 84	99.25 120	98.87 171	98.71 149	99.71 106	99.30 149

ACMMP		99.36 99	99.06 125	99.71 73	99.86 66	99.36 113	99.63 78	99.85 78	98.33 173	99.72 102	97.73 204	99.94 33	99.53 77	99.37 75	99.13 84	99.65 116	99.56 85
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
3Dnovator+	98.92 7	99.31 111	99.03 129	99.63 92	99.77 127	98.90 178	99.52 105	99.81 118	99.37 58	99.72 102	98.03 199	99.73 98	99.32 111	98.99 153	98.81 140	99.67 112	99.36 141

MSDG		99.32 108	99.09 122	99.58 102	99.75 136	98.74 191	99.36 140	99.54 189	99.14 90	99.72 102	99.24 135	99.89 59	99.51 81	99.30 89	98.76 144	99.62 133	98.54 194

IterMVS-LS		99.16 132	98.82 152	99.57 106	99.87 50	99.71 28	99.58 93	99.92 43	99.24 70	99.71 107	99.73 73	95.79 191	98.91 157	98.82 183	98.66 155	99.43 168	99.77 29
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
ACMM	98.37 12	99.47 79	99.23 92	99.74 69	99.86 66	99.19 153	99.68 71	99.86 66	99.16 86	99.71 107	98.52 181	99.95 23	99.62 62	99.35 80	99.02 100	99.74 96	99.42 126
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
v16		99.61 43	99.47 52	99.78 49	99.79 116	99.60 48	99.72 62	99.94 32	99.45 49	99.70 109	99.85 57	99.54 124	99.67 44	99.15 128	98.96 112	99.83 70	99.76 33

N_pmnet		98.64 177	98.23 181	99.11 183	99.78 121	99.25 141	99.75 46	99.39 210	99.65 16	99.70 109	99.78 70	99.89 59	98.81 162	97.60 214	94.28 217	97.24 218	97.15 218

XVS					99.86 66	99.30 130	99.72 62			99.69 111		99.93 39				99.60 137

v18		99.59 51	99.44 59	99.76 57	99.78 121	99.57 55	99.70 69	99.93 36	99.43 50	99.69 111	99.85 57	99.51 128	99.65 55	99.08 145	98.87 126	99.82 74	99.74 39

X-MVStestdata					99.86 66	99.30 130	99.72 62			99.69 111		99.93 39				99.60 137

X-MVS		99.30 113	98.99 134	99.66 84	99.85 79	99.30 130	99.49 117	99.82 106	98.32 174	99.69 111	97.31 213	99.93 39	99.50 85	99.37 75	99.16 77	99.60 137	99.53 96

tpm		96.56 220	94.68 221	98.74 204	99.12 225	97.90 227	98.79 210	99.93 36	96.79 224	99.69 111	99.19 140	81.48 232	97.56 195	95.46 223	93.97 218	97.37 217	97.99 207

CSCG		99.61 43	99.52 44	99.71 73	99.89 39	99.62 43	99.52 105	99.76 147	99.61 24	99.69 111	99.73 73	99.96 14	99.57 70	99.27 100	98.62 159	99.81 78	99.85 15

PMVS	94.32 17	99.27 118	99.55 39	98.94 193	99.60 181	99.43 92	99.39 132	99.54 189	98.99 105	99.69 111	99.60 94	99.81 84	95.68 223	99.88 12	99.83 3	99.73 100	99.31 147
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
MP-MVS		99.35 102	99.09 122	99.65 86	99.84 87	99.22 149	99.59 89	99.78 133	98.13 182	99.67 118	98.44 185	99.93 39	99.43 100	99.31 88	99.09 89	99.60 137	99.49 107

MVS-HIRNet		98.45 184	98.25 179	98.69 205	99.12 225	97.81 230	98.55 220	99.85 78	98.58 153	99.67 118	99.61 91	99.86 70	97.46 197	97.95 212	96.37 214	97.49 216	97.56 215

CPTT-MVS		99.21 123	98.89 144	99.58 102	99.72 148	99.12 164	99.30 153	99.76 147	98.62 147	99.66 120	97.51 207	99.89 59	99.48 93	99.01 148	98.64 157	99.58 146	99.40 134

MVS_111021_HR		99.30 113	99.14 113	99.48 125	99.58 188	99.25 141	99.27 157	99.61 177	98.74 133	99.66 120	99.02 157	99.84 78	99.33 108	99.20 115	98.76 144	99.44 165	99.18 160

PatchmatchNet		96.81 215	95.41 216	98.43 213	99.43 208	98.30 214	99.23 161	99.93 36	98.19 179	99.64 122	98.81 168	93.50 201	97.43 198	92.89 228	90.78 225	94.94 231	95.41 224
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
LS3D		99.39 94	99.28 85	99.52 116	99.77 127	99.39 99	99.55 98	99.82 106	98.93 114	99.64 122	98.52 181	99.67 105	98.58 168	99.74 35	99.63 30	99.75 92	99.06 175

UGNet		99.40 92	99.61 27	99.16 176	99.88 43	99.64 42	99.61 84	99.77 139	99.31 64	99.63 124	99.33 127	99.93 39	96.46 215	99.63 47	99.53 43	99.63 127	99.89 5
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
MTAPA										99.62 125		99.95 23

Fast-Effi-MVS+		99.39 94	99.18 105	99.63 92	99.86 66	99.28 135	99.45 124	99.91 48	98.47 160	99.61 126	99.50 112	99.57 119	99.17 125	99.24 103	98.66 155	99.78 83	99.59 71

abl_6				99.21 170	99.49 200	98.62 199	98.90 200	99.44 204	97.08 215	99.61 126	97.19 215	99.73 98	98.35 174			99.45 163	98.84 186

tpmp4_e23		95.42 228	92.99 230	98.27 215	99.32 218	97.77 231	98.74 212	99.79 127	97.11 214	99.61 126	97.47 209	80.64 233	96.36 219	92.92 226	88.79 231	95.80 226	96.19 221

MVS_111021_LR		99.25 120	99.13 115	99.39 137	99.50 199	99.14 158	99.23 161	99.50 197	98.67 140	99.61 126	99.12 146	99.81 84	99.16 129	99.28 97	98.67 154	99.35 178	99.21 158

Effi-MVS+		99.20 126	98.93 139	99.50 120	99.79 116	99.26 138	98.82 208	99.96 15	98.37 172	99.60 130	99.12 146	98.36 175	99.05 144	98.93 160	98.82 135	99.78 83	99.68 52

TAPA-MVS	98.54 10	99.30 113	99.24 91	99.36 149	99.44 206	98.77 189	99.00 189	99.41 206	99.23 71	99.60 130	99.50 112	99.86 70	99.15 133	99.29 93	98.95 116	99.56 150	99.08 172
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
zzz-MVS		99.51 70	99.36 76	99.68 81	99.88 43	99.38 104	99.53 102	99.84 90	99.11 93	99.59 132	98.93 162	99.95 23	99.58 69	99.44 67	99.21 71	99.65 116	99.52 101

MCST-MVS		99.17 129	98.82 152	99.57 106	99.75 136	98.70 195	99.25 160	99.69 165	98.62 147	99.59 132	98.54 179	99.79 91	99.53 77	98.48 199	98.15 185	99.64 125	99.43 122

PLC	97.83 16	98.88 162	98.52 168	99.30 153	99.45 204	98.60 201	98.65 215	99.49 198	98.66 141	99.59 132	96.33 218	99.59 116	99.17 125	98.87 171	98.53 164	99.46 161	99.05 177
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
PCF-MVS	97.86 15	98.95 156	98.53 166	99.44 130	99.70 155	98.80 186	98.96 192	99.69 165	98.65 142	99.59 132	99.33 127	99.94 33	99.12 138	98.01 211	97.11 207	99.59 143	97.83 210
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
HFP-MVS		99.46 83	99.30 82	99.65 86	99.82 104	99.25 141	99.50 112	99.82 106	99.23 71	99.58 136	98.86 164	99.94 33	99.56 71	99.14 131	99.12 86	99.63 127	99.56 85

OMC-MVS		99.11 137	98.95 138	99.29 154	99.37 213	98.57 202	99.19 164	99.20 218	98.87 119	99.58 136	99.13 144	99.88 64	99.00 147	99.19 120	98.46 168	99.43 168	98.57 192

ADS-MVSNet		97.29 212	96.17 214	98.59 207	99.59 185	98.70 195	99.32 146	99.86 66	98.47 160	99.56 138	99.08 151	98.16 179	97.34 199	92.92 226	91.17 223	95.91 225	94.72 227

HyFIR lowres test		99.50 72	99.26 87	99.80 39	99.95 12	99.62 43	99.76 40	99.97 1	99.67 13	99.56 138	99.94 15	98.40 174	99.78 15	98.84 181	98.59 162	99.76 89	99.72 46

CNLPA		98.82 166	98.52 168	99.18 175	99.21 223	98.50 206	98.73 213	99.34 213	98.73 135	99.56 138	97.55 206	99.42 141	99.06 143	98.93 160	98.10 189	99.21 190	98.38 198

OPM-MVS		99.39 94	99.22 94	99.59 100	99.76 132	98.82 184	99.51 108	99.79 127	99.17 82	99.53 141	99.31 131	99.95 23	99.35 103	99.22 109	98.79 143	99.60 137	99.27 153

MTMP										99.53 141		99.92 48

FC-MVSNet-train		99.70 30	99.67 18	99.74 69	99.94 22	99.71 28	99.82 25	99.91 48	99.14 90	99.53 141	99.70 80	99.88 64	99.33 108	99.88 12	99.61 33	99.94 17	99.77 29

testgi		99.43 86	99.47 52	99.38 140	99.90 34	99.67 39	99.30 153	99.73 156	98.64 146	99.53 141	99.52 108	99.90 56	98.08 182	99.65 45	99.40 58	99.75 92	99.55 90

NCCC		98.88 162	98.42 174	99.42 132	99.62 172	98.81 185	99.10 179	99.54 189	98.76 130	99.53 141	95.97 220	99.80 89	99.16 129	98.49 197	98.06 192	99.55 154	99.05 177

HQP-MVS		98.70 175	98.19 182	99.28 159	99.61 177	98.52 204	98.71 214	99.35 211	97.97 191	99.53 141	97.38 211	99.85 76	99.14 135	97.53 215	96.85 212	99.36 176	99.26 156

UA-Net		99.64 40	99.62 25	99.66 84	99.97 2	99.82 7	99.14 174	99.96 15	98.95 110	99.52 147	99.38 124	99.86 70	99.55 72	99.72 37	99.66 26	99.80 81	99.94 1

HPM-MVS++		99.23 121	98.98 136	99.53 112	99.75 136	99.02 170	99.44 125	99.77 139	98.65 142	99.52 147	98.72 172	99.92 48	99.33 108	98.77 189	98.40 172	99.40 172	99.36 141

testus		98.74 169	98.33 176	99.23 163	99.71 151	99.03 168	98.17 229	99.60 179	97.18 211	99.52 147	98.07 197	98.45 172	99.21 123	98.30 202	98.06 192	99.14 196	99.21 158

DWT-MVSNet_training		94.92 229	92.14 231	98.15 219	99.37 213	98.43 209	98.99 190	98.51 226	96.76 225	99.52 147	97.35 212	77.20 234	97.08 205	89.76 233	90.38 227	95.43 228	95.13 226

diffmvs		98.83 165	98.51 171	99.19 173	99.62 172	98.98 174	99.18 165	99.82 106	99.15 89	99.51 151	99.66 83	95.37 196	98.07 183	98.49 197	98.22 181	98.96 200	99.73 42

EPMVS		96.76 217	95.30 218	98.46 212	99.42 209	98.47 207	99.32 146	99.91 48	98.42 168	99.51 151	99.07 153	92.81 208	97.12 203	92.39 229	91.71 221	95.51 227	94.20 229

DeepMVS_CX						96.39 234	97.15 236	88.89 234	97.94 193	99.51 151	95.71 221	97.88 181	98.19 175	98.92 162		97.73 215	97.75 213

MS-PatchMatch		98.94 157	98.71 157	99.21 170	99.52 192	98.22 220	98.97 191	99.53 194	98.76 130	99.50 154	98.59 177	99.56 120	98.68 166	98.63 193	98.45 170	99.05 198	98.73 188

RPMNet		97.70 203	96.54 206	99.06 187	99.57 191	98.23 217	98.95 195	99.97 1	96.89 221	99.49 155	99.13 144	89.63 215	97.09 204	96.68 222	97.02 209	99.26 185	98.19 205

ACMMP_Plus		99.47 79	99.33 79	99.63 92	99.85 79	99.28 135	99.56 96	99.83 98	98.75 132	99.48 156	99.03 156	99.95 23	99.47 96	99.48 56	99.19 72	99.57 147	99.59 71

APD-MVS		99.17 129	98.92 140	99.46 126	99.78 121	99.24 146	99.34 144	99.78 133	97.79 197	99.48 156	98.25 190	99.88 64	98.77 163	99.18 123	98.92 118	99.63 127	99.18 160
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
new_pmnet		98.91 160	98.89 144	98.94 193	99.51 197	98.27 216	99.15 172	98.66 225	99.17 82	99.48 156	99.79 69	99.80 89	98.49 171	99.23 106	98.20 182	98.34 210	97.74 214

RPSCF		99.48 76	99.45 56	99.52 116	99.73 147	99.33 124	99.13 176	99.77 139	99.33 62	99.47 159	99.39 123	99.92 48	99.36 102	99.63 47	99.13 84	99.63 127	99.41 127

DI_MVS_plusplus_trai		98.74 169	98.08 186	99.51 118	99.79 116	99.29 134	99.61 84	99.60 179	99.20 76	99.46 160	99.09 150	92.93 204	98.97 154	98.27 205	98.35 175	99.65 116	99.45 115

tpm cat1		95.52 227	93.49 227	97.88 225	99.28 221	97.87 229	98.65 215	99.77 139	97.27 210	99.46 160	98.04 198	90.99 212	95.46 224	88.57 235	88.14 233	94.64 234	93.54 232

IB-MVS	98.10 14	97.76 201	97.40 194	98.18 217	99.62 172	99.11 165	98.24 226	98.35 229	96.56 226	99.44 162	91.28 236	98.96 164	93.84 229	98.09 207	98.62 159	99.56 150	99.18 160
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
AdaColmap		98.93 158	98.53 166	99.39 137	99.52 192	98.65 198	99.11 178	99.59 184	98.08 186	99.44 162	97.46 210	99.45 137	99.24 121	98.92 162	98.44 171	99.44 165	98.73 188

CR-MVSNet		97.91 193	96.80 201	99.22 168	99.60 181	98.23 217	98.91 198	99.97 1	96.89 221	99.43 164	99.10 149	89.24 216	98.15 177	98.04 208	97.78 197	99.26 185	98.30 201

Patchmtry						98.19 222	98.91 198	99.97 1		99.43 164

PatchT		98.11 189	97.12 197	99.26 161	99.65 169	98.34 213	99.57 95	99.97 1	97.48 206	99.43 164	99.04 155	90.84 213	98.15 177	98.04 208	97.78 197	98.82 204	98.30 201

CostFormer		95.61 225	93.35 229	98.24 216	99.48 201	98.03 224	98.65 215	99.83 98	96.93 219	99.42 167	98.83 166	83.65 231	97.08 205	90.39 232	89.54 229	94.94 231	96.11 222

MVSTER		97.55 209	96.75 202	98.48 211	99.46 203	99.54 64	98.24 226	99.77 139	97.56 203	99.41 168	99.31 131	84.86 230	94.66 228	98.86 175	97.75 199	99.34 179	99.38 136

CDS-MVSNet		99.15 134	99.10 120	99.21 170	99.59 185	99.22 149	99.48 119	99.47 200	98.89 117	99.41 168	99.84 60	98.11 180	97.76 191	99.26 102	99.01 102	99.57 147	99.38 136
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
thresconf0.02		98.10 190	96.83 200	99.58 102	99.71 151	99.28 135	99.40 131	99.72 158	98.65 142	99.39 170	98.23 191	86.73 227	99.43 100	97.73 213	98.17 184	99.86 53	99.05 177

FMVSNet1		99.50 72	99.57 37	99.42 132	99.67 165	99.65 41	99.60 88	99.91 48	99.40 55	99.39 170	99.83 62	99.27 155	98.14 179	99.68 39	99.50 45	99.81 78	99.68 52

MSLP-MVS++		98.92 159	98.73 156	99.14 177	99.44 206	99.00 171	98.36 223	99.35 211	98.82 127	99.38 172	96.06 219	99.79 91	99.07 141	98.88 170	99.05 95	99.27 184	99.53 96

tfpn_n400		99.08 140	98.56 163	99.70 76	99.85 79	99.56 60	99.63 78	99.86 66	99.21 74	99.37 173	98.95 159	94.24 197	99.55 72	99.20 115	99.29 64	99.93 19	99.44 118

tfpnconf		99.08 140	98.56 163	99.70 76	99.85 79	99.56 60	99.63 78	99.86 66	99.21 74	99.37 173	98.95 159	94.24 197	99.55 72	99.20 115	99.29 64	99.93 19	99.44 118

tfpnview11		99.04 148	98.49 172	99.68 81	99.84 87	99.58 53	99.56 96	99.86 66	98.86 120	99.37 173	98.95 159	94.24 197	99.54 76	98.87 171	99.54 42	99.91 25	99.39 135

MVS_Test		99.09 139	98.92 140	99.29 154	99.61 177	99.07 166	99.04 182	99.81 118	98.58 153	99.37 173	99.74 71	98.87 166	98.41 173	98.61 194	98.01 194	99.50 158	99.57 84

CDPH-MVS		99.05 146	98.63 159	99.54 111	99.75 136	98.78 187	99.59 89	99.68 169	97.79 197	99.37 173	98.20 194	99.86 70	99.14 135	98.58 195	98.01 194	99.68 110	99.16 166

CNVR-MVS		99.08 140	98.83 149	99.37 143	99.61 177	98.74 191	99.15 172	99.54 189	98.59 152	99.37 173	98.15 195	99.88 64	99.08 140	98.91 165	98.46 168	99.48 160	99.06 175

SMA-MVS		99.47 79	99.45 56	99.50 120	99.83 98	99.34 119	99.14 174	99.60 179	99.09 94	99.36 179	99.60 94	99.96 14	99.46 97	99.41 70	99.16 77	99.59 143	99.61 68

TSAR-MVS + MP.		99.56 66	99.54 41	99.58 102	99.69 157	99.14 158	99.73 57	99.45 202	99.50 41	99.35 180	99.60 94	99.93 39	99.50 85	99.56 51	99.37 59	99.77 87	99.64 64

SD-MVS		99.35 102	99.26 87	99.46 126	99.66 166	99.15 157	98.92 197	99.67 171	99.55 39	99.35 180	98.83 166	99.91 54	99.35 103	99.19 120	98.53 164	99.78 83	99.68 52

GA-MVS		98.59 180	98.15 183	99.09 184	99.59 185	99.13 161	98.84 205	99.52 195	98.61 150	99.35 180	99.67 82	93.03 203	97.73 193	98.90 169	98.26 178	99.51 157	99.48 110

PatchMatch-RL		98.80 168	98.52 168	99.12 180	99.38 212	98.70 195	98.56 218	99.55 188	97.81 196	99.34 183	97.57 205	99.31 152	98.67 167	99.27 100	98.62 159	99.22 189	98.35 200

MAR-MVS		98.54 181	98.15 183	98.98 189	99.37 213	98.09 223	98.56 218	99.65 176	96.11 230	99.27 184	97.16 216	99.50 130	98.03 188	98.87 171	98.23 179	99.01 199	99.13 170
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
CMPMVS	76.62 19	98.64 177	98.60 160	98.68 206	99.33 216	97.07 232	98.11 233	98.50 227	97.69 200	99.26 185	98.35 189	99.66 106	97.62 194	99.43 68	99.02 100	99.24 187	99.01 181
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
HSP-MVS		99.27 118	99.07 124	99.50 120	99.76 132	99.54 64	99.73 57	99.72 158	98.94 112	99.23 186	98.96 158	99.96 14	98.91 157	98.72 191	97.71 201	99.63 127	99.66 57

MDTV_nov1_ep13		97.41 211	96.26 213	98.76 203	99.47 202	98.43 209	99.26 159	99.82 106	98.06 188	99.23 186	99.22 138	92.86 207	98.05 186	95.33 224	93.66 219	96.73 221	96.26 220

FPMVS		98.48 183	98.83 149	98.07 222	99.09 227	97.98 226	99.07 181	98.04 233	98.99 105	99.22 188	98.85 165	99.43 140	93.79 230	99.66 44	99.11 87	99.24 187	97.76 212

TSAR-MVS + ACMM		99.31 111	99.26 87	99.37 143	99.66 166	98.97 175	99.20 163	99.56 186	99.33 62	99.19 189	99.54 101	99.91 54	99.32 111	99.12 133	98.34 176	99.29 182	99.65 61

test2356		96.34 223	94.05 225	99.00 188	99.39 211	98.28 215	98.15 230	99.51 196	96.23 228	99.16 190	97.95 200	73.39 236	98.75 164	97.07 220	96.86 211	99.06 197	98.57 192

tpmrst		96.18 224	94.47 222	98.18 217	99.52 192	97.89 228	98.96 192	99.79 127	98.07 187	99.16 190	99.30 134	92.69 209	96.69 212	90.76 231	88.85 230	94.96 230	93.69 231

Vis-MVSNet (Re-imp)		99.40 92	99.28 85	99.55 110	99.92 27	99.68 36	99.31 148	99.87 62	98.69 138	99.16 190	99.08 151	98.64 170	99.20 124	99.65 45	99.46 51	99.83 70	99.72 46

PMMVS		98.71 174	98.55 165	98.90 197	99.28 221	98.45 208	98.53 221	99.45 202	97.67 201	99.15 193	98.76 169	99.54 124	97.79 190	98.77 189	98.23 179	99.16 194	98.46 197

ESAPD		99.21 123	99.14 113	99.29 154	99.79 116	99.44 88	99.02 187	99.79 127	97.96 192	99.12 194	99.22 138	99.95 23	98.50 170	99.21 112	98.84 131	99.56 150	99.34 144

Fast-Effi-MVS+-dtu		98.82 166	98.80 154	98.84 201	99.51 197	98.90 178	98.96 192	99.91 48	98.29 175	99.11 195	98.47 183	99.63 109	96.03 222	99.21 112	98.12 187	99.52 156	99.01 181

tfpn_ndepth		98.67 176	98.03 188	99.42 132	99.65 169	99.50 80	99.29 155	99.78 133	98.17 181	99.04 196	98.36 188	93.29 202	98.88 160	98.46 200	99.26 67	99.88 37	99.14 169

Effi-MVS+-dtu		99.01 150	99.05 126	98.98 189	99.60 181	99.13 161	99.03 186	99.61 177	98.52 159	99.01 197	98.53 180	99.83 81	96.95 208	99.48 56	98.59 162	99.66 114	99.25 157

canonicalmvs		99.00 151	98.68 158	99.37 143	99.68 164	99.42 95	98.94 196	99.89 56	99.00 104	98.99 198	98.43 186	95.69 192	98.96 155	99.18 123	99.18 73	99.74 96	99.88 9

train_agg		98.89 161	98.48 173	99.38 140	99.69 157	98.76 190	99.31 148	99.60 179	97.71 199	98.98 199	97.89 201	99.89 59	99.29 116	98.32 201	97.59 204	99.42 171	99.16 166

TAMVS		99.05 146	99.02 132	99.08 185	99.69 157	99.22 149	99.33 145	99.32 215	99.16 86	98.97 200	99.87 50	97.36 187	97.76 191	99.21 112	99.00 107	99.44 165	99.33 145

DeepPCF-MVS	98.38 11	99.16 132	99.20 99	99.12 180	99.20 224	98.71 194	98.85 204	99.06 221	99.17 82	98.96 201	99.61 91	99.86 70	99.29 116	99.17 125	98.72 148	99.36 176	99.15 168

EPP-MVSNet		99.34 104	99.10 120	99.62 97	99.94 22	99.74 25	99.66 73	99.80 124	99.07 98	98.93 202	99.61 91	96.13 190	99.49 89	99.67 42	99.63 30	99.92 23	99.86 13

conf0.05thres1000		98.36 187	97.28 196	99.63 92	99.92 27	99.74 25	99.66 73	99.88 60	98.68 139	98.92 203	97.30 214	86.02 229	99.49 89	99.77 31	99.73 19	99.93 19	99.69 51

EMVS		96.47 221	95.38 217	97.74 229	99.42 209	95.37 236	98.07 234	98.27 232	97.85 195	98.90 204	97.48 208	98.73 168	97.20 201	97.21 219	90.39 226	94.59 235	90.65 234

FMVSNet2		99.07 144	99.19 102	98.93 195	99.02 229	99.53 66	99.31 148	99.84 90	98.86 120	98.88 205	99.64 87	98.44 173	96.92 209	99.35 80	99.00 107	99.61 134	99.53 96

dps		95.59 226	93.46 228	98.08 220	99.33 216	98.22 220	98.87 202	99.70 163	96.17 229	98.87 206	97.75 203	86.85 226	96.60 213	91.24 230	89.62 228	95.10 229	94.34 228

EPNet		98.06 192	98.11 185	98.00 223	99.60 181	98.99 173	98.38 222	99.68 169	98.18 180	98.85 207	97.89 201	95.60 194	92.72 233	98.30 202	98.10 189	98.76 205	99.72 46
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
TSAR-MVS + COLMAP		98.74 169	98.58 162	98.93 195	99.29 220	98.23 217	99.04 182	99.24 217	98.79 129	98.80 208	99.37 125	99.71 100	98.06 184	98.02 210	97.46 206	99.16 194	98.48 196

MIMVSNet		99.00 151	99.03 129	98.97 191	99.32 218	99.32 128	99.39 132	99.91 48	98.41 169	98.76 209	99.24 135	99.17 158	97.13 202	99.30 89	98.80 141	99.29 182	99.01 181

tfpn1000		98.73 172	98.07 187	99.50 120	99.84 87	99.61 46	99.48 119	99.84 90	98.71 137	98.74 210	98.71 173	91.70 211	99.17 125	98.81 184	99.55 40	99.90 28	99.43 122

E-PMN		96.72 218	95.78 215	97.81 226	99.45 204	95.46 235	98.14 232	98.33 231	97.99 190	98.73 211	98.09 196	98.97 163	97.54 196	97.45 217	91.09 224	94.70 233	91.40 233

thres100view900		97.69 204	96.37 211	99.23 163	99.74 143	99.21 152	98.81 209	99.43 205	96.10 231	98.70 212	92.99 229	89.10 217	98.88 160	98.58 195	99.31 63	99.82 74	99.27 153

tfpn200view9		97.85 198	96.54 206	99.38 140	99.74 143	99.52 72	99.17 167	99.76 147	96.10 231	98.70 212	92.99 229	89.10 217	99.00 147	99.11 134	99.56 34	99.88 37	99.41 127

CHOSEN 280x420		98.99 153	98.91 142	99.07 186	99.77 127	99.26 138	99.55 98	99.92 43	98.62 147	98.67 214	99.62 90	97.20 188	98.44 172	99.50 54	99.18 73	98.08 213	98.99 184

thres200		97.87 196	96.56 203	99.39 137	99.76 132	99.52 72	99.13 176	99.76 147	96.88 223	98.66 215	92.87 232	88.77 221	99.16 129	99.11 134	99.42 55	99.88 37	99.33 145

tfpn111		98.25 188	97.29 195	99.37 143	99.74 143	99.52 72	99.17 167	99.76 147	96.10 231	98.65 216	98.23 191	89.10 217	99.00 147	99.11 134	99.56 34	99.88 37	99.41 127

conf200view11		97.85 198	96.54 206	99.37 143	99.74 143	99.52 72	99.17 167	99.76 147	96.10 231	98.65 216	92.99 229	89.10 217	99.00 147	99.11 134	99.56 34	99.88 37	99.41 127

thres400		97.82 200	96.47 210	99.40 136	99.81 107	99.44 88	99.29 155	99.69 165	97.15 212	98.57 218	92.82 233	87.96 222	99.16 129	98.96 158	99.55 40	99.86 53	99.41 127

conf0.01		96.70 219	94.44 224	99.34 150	99.71 151	99.46 85	99.17 167	99.73 156	96.10 231	98.53 219	91.96 234	75.75 235	99.00 147	98.85 177	99.56 34	99.87 47	99.38 136

conf0.002		96.39 222	93.87 226	99.33 152	99.70 155	99.45 86	99.17 167	99.71 161	96.10 231	98.51 220	91.88 235	72.65 238	99.00 147	98.80 185	98.82 135	99.87 47	99.38 136

view600		97.88 195	96.55 205	99.44 130	99.84 87	99.52 72	99.38 136	99.76 147	97.36 208	98.50 221	93.29 227	87.31 224	99.26 119	99.13 132	99.76 15	99.88 37	99.48 110

GBi-Net		98.96 154	99.05 126	98.85 199	99.02 229	99.53 66	99.31 148	99.78 133	98.13 182	98.48 222	99.43 116	97.58 183	96.92 209	99.68 39	99.50 45	99.61 134	99.53 96

test1		98.96 154	99.05 126	98.85 199	99.02 229	99.53 66	99.31 148	99.78 133	98.13 182	98.48 222	99.43 116	97.58 183	96.92 209	99.68 39	99.50 45	99.61 134	99.53 96

FMVSNet3		98.63 179	98.75 155	98.49 210	98.10 236	99.44 88	99.02 187	99.78 133	98.13 182	98.48 222	99.43 116	97.58 183	96.16 220	98.85 177	98.39 173	99.40 172	99.41 127

FMVSNet5		97.69 204	96.98 198	98.53 209	98.53 234	99.36 113	98.90 200	99.54 189	96.38 227	98.44 225	95.38 222	90.08 214	97.05 207	99.46 61	99.06 92	98.73 206	99.12 171

PMMVS2		99.23 121	99.22 94	99.24 162	99.80 108	99.14 158	99.50 112	99.82 106	99.12 92	98.41 226	99.91 35	99.98 5	98.51 169	99.48 56	98.76 144	99.38 174	98.14 206

view800		97.89 194	96.56 203	99.45 128	99.86 66	99.57 55	99.42 127	99.80 124	97.50 205	98.40 227	93.78 226	86.63 228	99.31 113	99.24 103	99.68 24	99.89 33	99.54 91

thres600view7		97.86 197	96.53 209	99.41 135	99.84 87	99.52 72	99.36 140	99.76 147	97.32 209	98.38 228	93.24 228	87.25 225	99.23 122	99.11 134	99.75 17	99.88 37	99.48 110

tfpn		96.77 216	94.47 222	99.45 128	99.88 43	99.62 43	99.46 123	99.83 98	97.61 202	98.27 229	94.22 225	71.45 239	99.34 107	99.32 86	99.46 51	99.90 28	99.58 80

CVMVSNet		99.06 145	98.88 147	99.28 159	99.52 192	99.53 66	99.42 127	99.69 165	98.74 133	98.27 229	99.89 42	95.48 195	99.44 98	99.46 61	99.33 60	99.32 181	99.75 36

1111		96.83 213	95.02 219	98.95 192	99.90 34	99.57 55	99.62 82	99.97 1	98.58 153	98.06 231	99.87 50	69.04 240	96.43 217	99.36 78	99.14 81	99.73 100	99.54 91

.test1245		79.44 231	75.07 233	84.53 233	99.90 34	99.57 55	99.62 82	99.97 1	98.58 153	98.06 231	99.87 50	69.04 240	96.43 217	99.36 78	24.74 234	13.21 238	34.30 235

MVE	91.08 18	97.68 206	97.65 189	97.71 230	98.46 235	91.62 239	97.92 235	98.86 223	98.73 135	97.99 233	98.64 176	99.96 14	99.17 125	99.59 50	97.75 199	93.87 236	97.27 216
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
testpf		93.65 230	91.79 232	95.82 231	98.71 233	93.25 237	96.38 237	99.67 171	95.38 237	97.83 234	94.48 224	87.69 223	89.61 235	88.96 234	88.79 231	92.71 237	93.97 230

test-LLR		97.74 202	97.46 192	98.08 220	99.62 172	98.37 211	98.26 224	99.41 206	97.03 216	97.38 235	99.54 101	92.89 205	95.12 226	98.78 187	97.68 202	98.65 208	97.90 208

TESTMET0.1,1		97.62 208	97.46 192	97.81 226	99.07 228	98.37 211	98.26 224	98.35 229	97.03 216	97.38 235	99.54 101	92.89 205	95.12 226	98.78 187	97.68 202	98.65 208	97.90 208

IS_MVSNet		99.15 134	99.12 117	99.19 173	99.92 27	99.73 27	99.55 98	99.86 66	98.45 165	96.91 237	98.74 170	98.33 176	99.02 146	99.54 53	99.47 49	99.88 37	99.61 68

test-mter		97.65 207	97.57 191	97.75 228	98.90 232	98.56 203	98.15 230	98.45 228	96.92 220	96.84 238	99.52 108	92.53 210	95.24 225	99.04 146	98.12 187	98.90 203	98.29 203

test0.0.03 1		98.41 185	98.41 175	98.40 214	99.62 172	99.16 155	98.87 202	99.41 206	97.15 212	96.60 239	99.31 131	97.00 189	96.55 214	98.91 165	98.51 166	99.37 175	98.82 187

tmp_tt				88.14 232	96.68 237	91.91 238	93.70 238	61.38 235	99.61 24	90.51 240	99.40 122	99.71 100	90.32 234	99.22 109	99.44 54	96.25 224

GG-mvs-BLEND		70.44 232	96.91 199	39.57 234	3.32 240	96.51 233	91.01 239	4.05 238	97.03 216	33.20 241	94.67 223	97.75 182	7.59 238	98.28 204	96.85 212	98.24 211	97.26 217

testmvs		22.33 233	29.66 234	13.79 235	8.97 238	10.35 240	15.53 242	8.09 237	32.51 238	19.87 242	45.18 237	30.56 243	17.05 237	29.96 236	24.74 234	13.21 238	34.30 235

test123		21.52 234	28.47 235	13.42 236	7.29 239	10.12 241	15.70 241	8.31 236	31.54 239	19.34 243	36.33 238	37.40 242	17.14 236	27.45 237	23.17 236	12.73 240	33.30 237

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

Patchmatch-RL test							65.75 240

mPP-MVS					99.84 87							99.92 48

NP-MVS									97.37 207

High-resolution multi-view benchmark