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.

v7n		99.68 4	99.61 3	99.76 8	99.89 13	99.74 7	99.87 1	99.82 13	99.20 5	99.71 5	99.96 1	99.73 11	99.76 5	99.58 16	99.59 14	99.52 45	99.46 17

v52		99.67 5	99.59 6	99.76 8	99.91 8	99.69 10	99.85 3	99.79 15	99.12 8	99.68 11	99.95 2	99.72 13	99.77 2	99.58 16	99.61 10	99.54 40	99.50 12

V4		99.67 5	99.60 5	99.76 8	99.91 8	99.69 10	99.85 3	99.79 15	99.13 7	99.68 11	99.95 2	99.72 13	99.77 2	99.58 16	99.61 10	99.54 40	99.50 12

gg-mvs-nofinetune		96.77 183	96.52 172	97.06 207	99.66 76	97.82 197	97.54 222	99.86 8	98.69 16	98.61 131	99.94 4	89.62 201	88.37 242	97.55 169	96.67 188	98.30 185	95.35 207

v1240		98.86 71	98.41 90	99.38 51	99.59 101	99.05 93	99.65 22	99.14 159	97.68 60	99.66 14	99.93 5	98.72 130	99.45 33	97.38 180	97.72 148	98.79 154	98.35 122

new-patchmatchnet		97.26 169	96.12 182	98.58 152	99.55 110	98.63 151	99.14 102	97.04 231	98.80 15	99.19 65	99.92 6	99.19 81	98.92 86	95.51 215	87.04 231	97.66 200	93.73 219

anonymousdsp		99.64 8	99.55 8	99.74 13	99.87 16	99.56 22	99.82 6	99.73 27	98.54 18	99.71 5	99.92 6	99.84 6	99.61 12	99.70 5	99.63 5	99.69 26	99.64 2

v1921920		98.89 62	98.46 76	99.39 46	99.58 103	99.04 98	99.64 25	99.17 154	97.91 41	99.64 16	99.92 6	98.99 116	99.44 36	97.44 176	97.57 160	98.84 145	98.35 122

v1192		98.91 58	98.48 75	99.41 41	99.61 100	99.03 103	99.64 25	99.25 140	97.91 41	99.58 19	99.92 6	99.07 109	99.45 33	97.55 169	97.68 150	98.93 125	98.23 133

v11		99.19 30	98.95 31	99.47 32	99.66 76	99.54 28	99.65 22	99.73 27	98.06 30	99.38 36	99.92 6	99.40 54	99.55 19	98.29 108	98.50 96	98.88 136	98.92 68

v1144		98.94 53	98.53 70	99.42 40	99.62 97	99.03 103	99.58 36	99.36 114	97.99 34	99.49 29	99.91 11	99.20 80	99.51 24	97.61 165	97.85 133	98.95 123	98.10 146

v748		99.67 5	99.61 3	99.75 12	99.87 16	99.68 12	99.84 5	99.79 15	99.14 6	99.64 16	99.89 12	99.88 4	99.72 8	99.58 16	99.57 16	99.62 31	99.50 12

v144192		98.88 65	98.46 76	99.37 53	99.56 108	99.03 103	99.61 32	99.26 137	97.79 48	99.58 19	99.88 13	99.11 100	99.43 38	97.38 180	97.61 156	98.80 152	98.43 117

v13		99.22 28	98.99 30	99.49 30	99.68 69	99.58 20	99.67 19	99.77 20	98.10 28	99.36 37	99.88 13	99.37 57	99.54 21	98.50 87	98.51 95	98.92 128	99.03 51

v12		99.19 30	98.95 31	99.48 31	99.67 72	99.56 22	99.66 21	99.76 21	98.06 30	99.33 43	99.88 13	99.34 62	99.53 22	98.42 95	98.43 100	98.91 131	98.97 59

v7		98.91 58	98.53 70	99.36 55	99.53 120	98.99 109	99.57 37	99.36 114	97.58 69	99.32 45	99.88 13	99.23 74	99.50 26	97.77 153	97.98 121	98.91 131	98.26 131

v10		99.01 45	98.66 59	99.41 41	99.52 125	99.39 43	99.57 37	99.66 41	97.59 67	99.32 45	99.88 13	99.23 74	99.50 26	97.77 153	97.98 121	98.92 128	98.78 86

pmmvs6		99.74 2	99.75 1	99.73 14	99.92 4	99.67 14	99.76 13	99.84 10	99.59 1	99.52 26	99.87 18	99.91 1	99.43 38	99.87 1	99.81 2	99.89 5	99.52 9

FC-MVSNet-test		99.32 22	99.33 13	99.31 65	99.87 16	99.65 17	99.63 28	99.75 24	97.76 49	97.29 209	99.87 18	99.63 33	99.52 23	99.66 8	99.63 5	99.77 18	99.12 37

V14		99.13 37	98.85 44	99.45 34	99.65 82	99.52 30	99.63 28	99.74 26	97.97 35	99.30 50	99.87 18	99.27 70	99.49 28	98.23 114	98.24 106	98.88 136	98.83 76

V9		99.16 34	98.90 39	99.46 33	99.66 76	99.54 28	99.65 22	99.75 24	98.01 33	99.31 47	99.87 18	99.31 66	99.51 24	98.34 102	98.34 103	98.90 133	98.91 69

v1141		98.87 66	98.45 80	99.36 55	99.65 82	99.04 98	99.56 39	99.38 103	97.83 45	99.29 52	99.86 22	99.16 86	99.40 42	97.68 159	97.78 136	98.86 141	97.82 158

divwei89l23v2f112		98.87 66	98.45 80	99.36 55	99.65 82	99.04 98	99.56 39	99.38 103	97.83 45	99.29 52	99.86 22	99.15 90	99.40 42	97.68 159	97.78 136	98.86 141	97.82 158

v15		99.09 40	98.79 46	99.43 38	99.64 90	99.50 31	99.61 32	99.73 27	97.92 39	99.28 57	99.86 22	99.24 72	99.47 30	98.12 125	98.14 111	98.87 138	98.76 88

v1		98.87 66	98.45 80	99.36 55	99.65 82	99.04 98	99.55 42	99.38 103	97.83 45	99.30 50	99.86 22	99.17 83	99.40 42	97.68 159	97.77 143	98.86 141	97.82 158

MIMVSNet1		99.46 16	99.34 12	99.60 18	99.83 23	99.68 12	99.74 17	99.71 32	98.20 26	99.41 34	99.86 22	99.66 26	99.41 41	99.50 23	99.39 25	99.50 51	99.10 42

pm-mvs1		99.47 15	99.38 11	99.57 21	99.82 25	99.49 32	99.63 28	99.65 43	98.88 12	99.31 47	99.85 27	99.02 112	99.23 64	99.60 14	99.58 15	99.80 14	99.22 31

v2v482		98.85 74	98.40 91	99.38 51	99.65 82	98.98 110	99.55 42	99.39 96	97.92 39	99.35 40	99.85 27	99.14 92	99.39 52	97.50 171	97.78 136	98.98 120	97.60 165

Baseline_NR-MVSNet		99.18 33	98.87 41	99.54 25	99.74 53	99.56 22	99.36 71	99.62 52	96.53 132	99.29 52	99.85 27	98.64 138	99.40 42	99.03 54	99.63 5	99.83 10	98.86 75

SixPastTwentyTwo		99.70 3	99.59 6	99.82 2	99.93 2	99.80 1	99.86 2	99.87 6	98.87 13	99.79 4	99.85 27	99.33 63	99.74 7	99.85 2	99.82 1	99.74 21	99.63 4

TransMVSNet (Re)		99.45 17	99.32 15	99.61 16	99.88 15	99.60 18	99.75 14	99.63 47	99.11 9	99.28 57	99.83 31	98.35 146	99.27 61	99.70 5	99.62 9	99.84 9	99.03 51

v8		98.94 53	98.60 62	99.35 60	99.54 113	99.39 43	99.55 42	99.67 40	97.48 74	99.13 77	99.81 32	99.10 101	99.39 52	97.86 142	97.89 127	98.81 147	98.66 98

v6		98.84 75	98.46 76	99.30 66	99.54 113	98.98 110	99.54 46	99.37 111	97.49 73	99.11 81	99.81 32	99.13 95	99.40 42	97.86 142	97.89 127	98.81 147	98.04 149

V42		98.81 84	98.49 74	99.18 79	99.52 125	98.92 125	99.50 54	99.29 133	97.43 80	98.97 99	99.81 32	99.00 115	99.30 58	97.93 135	98.01 117	98.51 178	98.34 126

v1neww		98.84 75	98.45 80	99.29 69	99.54 113	98.98 110	99.54 46	99.37 111	97.48 74	99.10 82	99.80 35	99.12 96	99.40 42	97.85 145	97.89 127	98.81 147	98.04 149

v7new		98.84 75	98.45 80	99.29 69	99.54 113	98.98 110	99.54 46	99.37 111	97.48 74	99.10 82	99.80 35	99.12 96	99.40 42	97.85 145	97.89 127	98.81 147	98.04 149

v17		98.96 51	98.63 60	99.35 60	99.54 113	99.41 40	99.55 42	99.70 34	97.40 82	99.10 82	99.79 37	99.10 101	99.40 42	97.96 132	97.99 119	98.80 152	98.77 87

LTVRE_ROB	98.82 1	99.76 1	99.75 1	99.77 7	99.87 16	99.71 8	99.77 11	99.76 21	99.52 2	99.80 2	99.79 37	99.91 1	99.56 17	99.83 3	99.75 3	99.86 8	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
WR-MVS		99.61 9	99.44 10	99.82 2	99.92 4	99.80 1	99.80 7	99.89 1	98.54 18	99.66 14	99.78 39	99.16 86	99.68 10	99.70 5	99.63 5	99.94 1	99.49 15

v18		98.89 62	98.54 68	99.30 66	99.50 128	99.37 46	99.51 51	99.68 37	97.25 95	99.00 97	99.76 40	99.04 110	99.36 54	97.81 149	97.86 132	98.77 157	98.68 97

v16		98.95 52	98.62 61	99.34 62	99.53 120	99.41 40	99.54 46	99.70 34	97.34 87	99.07 88	99.76 40	99.10 101	99.40 42	97.96 132	98.00 118	98.79 154	98.76 88

DeepC-MVS	97.88 4	99.33 21	99.15 24	99.53 28	99.73 58	99.05 93	99.49 55	99.40 94	98.42 21	99.55 23	99.71 42	99.89 3	99.49 28	99.14 40	98.81 67	99.54 40	99.02 54
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
gm-plane-assit		94.62 215	91.39 225	98.39 163	99.90 12	99.47 35	99.40 65	99.65 43	97.44 78	99.56 22	99.68 43	59.40 250	94.23 218	96.17 205	94.77 216	97.61 201	92.79 224

EU-MVSNet		98.68 90	98.94 35	98.37 165	99.14 186	98.74 143	99.64 25	98.20 213	98.21 25	99.17 68	99.66 44	99.18 82	99.08 79	99.11 42	98.86 60	95.00 222	98.83 76

no-one		99.01 45	98.94 35	99.09 91	98.97 200	98.55 158	99.37 69	99.04 170	97.59 67	99.36 37	99.66 44	99.75 8	99.57 15	98.47 88	99.27 35	98.21 190	99.30 25

FMVSNet1		98.90 60	99.10 26	98.67 142	99.54 113	99.48 33	99.22 91	99.66 41	98.39 24	97.50 196	99.66 44	99.04 110	96.58 182	99.05 49	99.03 52	99.52 45	99.08 44

TDRefinement		99.54 10	99.50 9	99.60 18	99.70 65	99.35 47	99.77 11	99.58 56	99.40 4	99.28 57	99.66 44	99.41 51	99.55 19	99.74 4	99.65 4	99.70 23	99.25 26

pmmvs-eth3d		98.68 90	98.14 109	99.29 69	99.49 131	98.45 167	99.45 61	99.38 103	97.21 97	99.50 28	99.65 48	99.21 78	99.16 72	97.11 188	97.56 161	98.79 154	97.82 158

v148		98.77 87	98.45 80	99.15 82	99.68 69	98.94 123	99.49 55	99.31 131	97.95 37	98.91 109	99.65 48	99.62 35	99.18 67	97.99 131	97.64 154	98.33 184	97.38 174

test20.03		98.84 75	98.74 50	98.95 111	99.77 39	99.33 50	99.21 93	99.46 88	97.29 89	98.88 114	99.65 48	99.10 101	97.07 176	99.11 42	98.76 73	99.32 83	97.98 154

Vis-MVSNet		99.25 25	99.32 15	99.17 80	99.65 82	99.55 26	99.63 28	99.33 124	98.16 27	99.29 52	99.65 48	99.77 7	97.56 158	99.44 29	99.14 42	99.58 37	99.51 11
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
PMMVS2		96.29 195	97.05 159	95.40 233	98.32 230	96.16 222	98.18 194	97.46 225	97.20 99	84.51 247	99.60 52	98.68 134	96.37 187	98.59 82	97.38 167	97.58 202	91.76 229

PS-CasMVS		99.50 13	99.23 20	99.82 2	99.92 4	99.75 6	99.78 10	99.89 1	97.30 88	99.71 5	99.60 52	99.23 74	99.71 9	99.65 9	99.55 17	99.90 2	99.56 7

PEN-MVS		99.54 10	99.30 17	99.83 1	99.92 4	99.76 4	99.80 7	99.88 3	97.60 66	99.71 5	99.59 54	99.52 43	99.75 6	99.64 11	99.51 18	99.90 2	99.46 17

MDTV_nov1_ep13_2view		97.12 174	96.19 181	98.22 176	99.13 188	98.05 187	99.24 88	99.47 85	97.61 65	99.15 74	99.59 54	99.01 113	98.40 119	94.87 223	90.14 226	93.91 227	94.04 218

CHOSEN 1792x2688		98.31 120	98.02 118	98.66 144	99.55 110	98.57 157	99.38 68	99.25 140	98.42 21	98.48 147	99.58 56	99.85 5	98.31 124	95.75 211	95.71 205	96.96 211	98.27 130

DELS-MVS		98.63 96	98.70 54	98.55 155	99.24 174	99.04 98	98.96 120	98.52 200	96.83 110	98.38 152	99.58 56	99.68 21	97.06 177	98.74 77	98.44 99	99.10 101	98.59 103
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
WR-MVS_H		99.48 14	99.23 20	99.76 8	99.91 8	99.76 4	99.75 14	99.88 3	97.27 91	99.58 19	99.56 58	99.24 72	99.56 17	99.60 14	99.60 13	99.88 7	99.58 6

PVSNet_Blended_VisFu		98.98 48	98.79 46	99.21 74	99.76 45	99.34 48	99.35 73	99.35 119	97.12 102	99.46 31	99.56 58	98.89 120	98.08 139	99.05 49	98.58 89	99.27 89	98.98 58

IterMVS		97.40 167	96.67 166	98.25 170	99.45 137	98.66 149	98.87 133	98.73 188	96.40 137	98.94 106	99.56 58	95.26 189	97.58 157	95.38 216	94.70 217	95.90 220	96.72 191
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
DTE-MVSNet		99.52 12	99.27 18	99.82 2	99.93 2	99.77 3	99.79 9	99.87 6	97.89 44	99.70 10	99.55 61	99.21 78	99.77 2	99.65 9	99.43 22	99.90 2	99.36 21

1111		94.22 222	92.26 220	96.51 224	99.71 63	98.75 141	99.03 111	99.83 11	95.01 172	93.39 242	99.54 62	60.23 248	89.58 238	97.90 138	97.62 155	97.50 204	96.75 190

.test1245		74.10 239	68.09 242	81.11 240	99.71 63	98.75 141	99.03 111	99.83 11	95.01 172	93.39 242	99.54 62	60.23 248	89.58 238	97.90 138	10.38 244	5.14 248	14.81 243

TranMVSNet+NR-MVSNet		99.23 26	98.91 38	99.61 16	99.81 29	99.45 36	99.47 57	99.68 37	97.28 90	99.39 35	99.54 62	99.08 107	99.45 33	99.09 45	98.84 64	99.83 10	99.04 49

Anonymous20231206		98.50 108	98.03 117	99.05 95	99.50 128	99.01 107	99.15 101	99.26 137	96.38 138	99.12 79	99.50 65	99.12 96	98.60 105	97.68 159	97.24 174	98.66 164	97.30 176

FC-MVSNet-train		99.13 37	99.05 27	99.21 74	99.87 16	99.57 21	99.67 19	99.60 55	96.75 118	98.28 160	99.48 66	99.52 43	98.10 136	99.47 26	99.37 27	99.76 20	99.21 32

CVMVSNet		97.38 168	97.39 144	97.37 200	98.58 218	97.72 203	98.70 145	97.42 226	97.21 97	95.95 226	99.46 67	93.31 196	97.38 167	97.60 166	97.78 136	96.18 216	98.66 98

pmmvs5		98.37 117	97.81 128	99.03 98	99.46 135	98.97 117	99.03 111	98.96 177	95.85 156	99.05 91	99.45 68	98.66 137	98.79 94	96.02 208	97.52 162	98.87 138	98.21 136

COLMAP_ROB	98.29 2	99.37 20	99.25 19	99.51 29	99.74 53	99.12 84	99.56 39	99.39 96	98.96 11	99.17 68	99.44 69	99.63 33	99.58 14	99.48 25	99.27 35	99.60 36	98.81 81
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
CSCG		99.23 26	99.15 24	99.32 64	99.83 23	99.45 36	98.97 119	99.21 145	98.83 14	99.04 94	99.43 70	99.64 31	99.26 62	98.85 69	98.20 109	99.62 31	99.62 5

N_pmnet		96.68 185	95.70 194	97.84 191	99.42 143	98.00 190	99.35 73	98.21 211	98.40 23	98.13 169	99.42 71	99.30 67	97.44 166	94.00 230	88.79 228	94.47 226	91.96 227

ACMH	97.81 6	99.44 18	99.33 13	99.56 22	99.81 29	99.42 39	99.73 18	99.58 56	99.02 10	99.10 82	99.41 72	99.69 18	99.60 13	99.45 27	99.26 37	99.55 39	99.05 48
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
PMVS	92.51 17	98.66 92	98.86 42	98.43 160	99.26 169	98.98 110	98.60 162	98.59 197	97.73 56	99.45 32	99.38 73	98.54 141	95.24 203	99.62 13	99.61 10	99.42 63	98.17 142
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
IterMVS-LS		98.23 125	97.66 134	98.90 114	99.63 95	99.38 45	99.07 108	99.48 84	97.75 52	98.81 119	99.37 74	94.57 192	97.88 146	96.54 199	97.04 179	98.53 175	98.97 59
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
CP-MVSNet		99.39 19	99.04 28	99.80 6	99.91 8	99.70 9	99.75 14	99.88 3	96.82 111	99.68 11	99.32 75	98.86 122	99.68 10	99.57 20	99.47 20	99.89 5	99.52 9

thisisatest0515		99.16 34	98.94 35	99.41 41	99.75 48	99.43 38	99.36 71	99.63 47	97.68 60	99.35 40	99.31 76	98.90 119	99.09 78	98.95 59	99.20 38	99.27 89	99.11 38

EG-PatchMatch MVS		99.01 45	98.77 49	99.28 73	99.64 90	98.90 130	98.81 139	99.27 136	96.55 130	99.71 5	99.31 76	99.66 26	99.17 70	99.28 37	99.11 45	99.10 101	98.57 106

UniMVSNet (Re)		99.08 41	98.69 56	99.54 25	99.75 48	99.33 50	99.29 80	99.64 46	96.75 118	99.48 30	99.30 78	98.69 132	99.26 62	98.94 61	98.76 73	99.78 17	99.02 54

ESAPD		98.84 75	98.69 56	99.00 104	99.05 196	99.26 58	99.19 96	99.35 119	95.85 156	98.74 125	99.27 79	99.66 26	98.30 125	98.90 67	98.93 57	99.37 73	99.00 56

TAMVS		96.95 179	96.94 163	96.97 213	99.07 195	97.67 206	97.98 201	97.12 230	95.04 171	95.41 232	99.27 79	95.57 188	94.09 219	97.32 182	97.11 177	98.16 192	96.59 193

testmv		97.48 165	96.83 165	98.24 174	99.37 147	97.79 199	98.59 163	99.07 167	92.40 217	97.59 191	99.24 81	98.11 153	97.66 155	97.64 163	97.11 177	97.17 207	95.54 206

test1235678		97.49 163	96.84 164	98.24 174	99.37 147	97.79 199	98.59 163	99.07 167	92.41 216	97.59 191	99.24 81	98.15 152	97.66 155	97.64 163	97.12 176	97.17 207	95.55 205

CDS-MVSNet		97.75 146	97.68 133	97.83 192	99.08 193	98.20 182	98.68 147	98.61 196	95.63 160	97.80 183	99.24 81	96.93 177	94.09 219	97.96 132	97.82 134	98.71 162	97.99 152
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
HyFIR lowres test		98.08 134	97.16 156	99.14 85	99.72 60	98.91 127	99.41 63	99.58 56	97.93 38	98.82 118	99.24 81	95.81 187	98.73 98	95.16 221	95.13 213	98.60 170	97.94 155

new_pmnet		96.59 186	96.40 175	96.81 216	98.24 233	95.46 235	97.71 215	94.75 239	96.92 105	96.80 220	99.23 85	97.81 163	96.69 180	96.58 198	95.16 212	96.69 212	93.64 220

ACMH+	97.53 7	99.29 24	99.20 23	99.40 45	99.81 29	99.22 66	99.59 35	99.50 79	98.64 17	98.29 159	99.21 86	99.69 18	99.57 15	99.53 22	99.33 30	99.66 28	98.81 81

APDe-MVS		99.15 36	98.95 31	99.39 46	99.77 39	99.28 57	99.52 50	99.54 68	97.22 96	99.06 89	99.20 87	99.64 31	99.05 82	99.14 40	99.02 55	99.39 71	99.17 35

FMVSNet2		97.94 138	98.08 114	97.77 194	98.71 209	99.21 67	98.62 158	99.47 85	96.62 122	96.37 222	99.20 87	97.70 164	94.39 214	97.39 178	97.75 144	99.08 107	98.70 94

USDC		98.26 123	97.57 139	99.06 92	99.42 143	97.98 193	98.83 135	98.85 181	97.57 70	99.59 18	99.15 89	98.59 139	98.99 84	97.42 177	96.08 203	98.69 163	96.23 198

test12356		95.71 205	95.55 195	95.89 231	98.27 232	96.48 217	96.90 234	97.35 228	92.13 221	95.64 228	99.13 90	97.97 159	92.34 231	96.94 190	96.55 192	94.87 224	89.61 236

PM-MVS		98.57 102	98.24 104	98.95 111	99.26 169	98.59 154	99.03 111	98.74 187	96.84 108	99.44 33	99.13 90	98.31 148	98.75 97	98.03 129	98.21 107	98.48 179	98.58 104

TSAR-MVS + ACMM		98.64 95	98.58 65	98.72 135	99.17 182	98.63 151	98.69 146	99.10 166	97.69 59	98.30 158	99.12 92	99.38 56	98.70 99	98.45 89	97.51 163	98.35 182	99.25 26

CLD-MVS		98.48 110	98.15 108	98.86 121	99.53 120	98.35 172	98.55 167	97.83 223	96.02 152	98.97 99	99.08 93	99.75 8	99.03 83	98.10 127	97.33 170	99.28 88	98.44 116
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
SMA-MVS		98.87 66	98.73 51	99.04 97	99.72 60	99.05 93	98.64 155	99.17 154	96.31 141	98.80 120	99.07 94	99.70 17	98.67 101	98.93 64	98.82 65	99.23 95	99.23 30

TSAR-MVS + GP.		98.54 106	98.29 101	98.82 126	99.28 167	98.59 154	97.73 212	99.24 142	95.93 154	98.59 133	99.07 94	99.17 83	98.86 90	98.44 90	98.10 113	99.26 91	98.72 92

UniMVSNet_NR-MVSNet		98.97 49	98.46 76	99.56 22	99.76 45	99.34 48	99.29 80	99.61 53	96.55 130	99.55 23	99.05 96	97.96 160	99.36 54	98.84 70	98.50 96	99.81 13	98.97 59

DU-MVS		99.04 43	98.59 63	99.56 22	99.74 53	99.23 63	99.29 80	99.63 47	96.58 126	99.55 23	99.05 96	98.68 134	99.36 54	99.03 54	98.60 87	99.77 18	98.97 59

NR-MVSNet		99.10 39	98.68 58	99.58 20	99.89 13	99.23 63	99.35 73	99.63 47	96.58 126	99.36 37	99.05 96	98.67 136	99.46 31	99.63 12	98.73 77	99.80 14	98.88 74

TSAR-MVS + MP.		99.02 44	98.95 31	99.11 88	99.23 175	98.79 136	99.51 51	98.73 188	97.50 72	98.56 135	99.03 99	99.59 39	99.16 72	99.29 35	99.17 40	99.50 51	99.24 29

pmmvs4		97.87 143	97.02 160	98.86 121	99.20 177	97.68 205	98.89 131	99.03 171	96.57 128	99.12 79	99.03 99	97.26 173	98.42 118	95.16 221	96.34 194	98.53 175	97.10 186

MDA-MVSNet-bldmvs		97.75 146	97.26 148	98.33 166	99.35 154	98.45 167	99.32 78	97.21 229	97.90 43	99.05 91	99.01 101	96.86 178	99.08 79	99.36 31	92.97 223	95.97 219	96.25 197

UA-Net		99.30 23	99.22 22	99.39 46	99.94 1	99.66 16	98.91 127	99.86 8	97.74 54	98.74 125	99.00 102	99.60 38	99.17 70	99.50 23	99.39 25	99.70 23	99.64 2

ambc			97.89 126		99.45 137	97.88 195	97.78 209		97.27 91	99.80 2	98.99 103	98.48 143	98.55 110	97.80 150	96.68 187	98.54 174	98.10 146

RPSCF		98.84 75	98.81 45	98.89 116	99.37 147	98.95 119	98.51 169	98.85 181	97.73 56	98.33 156	98.97 104	99.14 92	98.95 85	99.18 39	98.68 81	99.31 84	98.99 57

DeepPCF-MVS	96.68 10	98.20 128	98.26 102	98.12 180	97.03 244	98.11 185	98.44 175	97.70 224	96.77 115	98.52 140	98.91 105	99.17 83	98.58 107	98.41 96	98.02 116	98.46 180	98.46 113

TSAR-MVS + COLMAP		97.62 155	97.31 146	97.98 186	98.47 224	97.39 209	98.29 187	98.25 209	96.68 120	97.54 195	98.87 106	98.04 157	97.08 175	96.78 193	96.26 195	98.26 187	97.12 185

CANet		98.47 111	98.30 99	98.67 142	99.65 82	98.87 132	98.82 138	99.01 173	96.14 148	99.29 52	98.86 107	99.01 113	96.54 183	98.36 100	98.08 114	98.72 161	98.80 85

MVS_Test		97.69 151	97.15 157	98.33 166	99.27 168	98.43 169	98.25 191	99.29 133	95.00 174	97.39 202	98.86 107	98.00 158	97.14 174	95.38 216	96.22 196	98.62 168	98.15 144

QAPM		98.62 97	98.40 91	98.89 116	99.57 107	98.80 135	98.63 156	99.35 119	96.82 111	98.60 132	98.85 109	99.08 107	98.09 138	98.31 106	98.21 107	99.08 107	98.72 92

UGNet		98.52 107	99.00 29	97.96 188	99.58 103	99.26 58	99.27 84	99.40 94	98.07 29	98.28 160	98.76 110	99.71 16	92.24 232	98.94 61	98.85 62	99.00 119	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
OPM-MVS		98.84 75	98.59 63	99.12 86	99.52 125	98.50 164	99.13 103	99.22 143	97.76 49	98.76 122	98.70 111	99.61 36	98.90 87	98.67 79	98.37 102	99.19 97	98.57 106

MVS-HIRNet		94.86 212	93.83 210	96.07 227	97.07 243	94.00 242	94.31 244	99.17 154	91.23 237	98.17 165	98.69 112	97.43 169	95.66 198	94.05 229	91.92 224	92.04 238	89.46 237

EPP-MVSNet		98.61 98	98.19 107	99.11 88	99.86 21	99.60 18	99.44 62	99.53 72	97.37 86	96.85 218	98.69 112	93.75 193	99.18 67	99.22 38	99.35 29	99.82 12	99.32 23

TinyColmap		98.27 122	97.62 138	99.03 98	99.29 164	97.79 199	98.92 125	98.95 178	97.48 74	99.52 26	98.65 114	97.86 162	98.90 87	98.34 102	97.27 172	98.64 167	95.97 201

casdiffmvs1		98.43 114	97.95 123	98.98 108	99.49 131	99.08 87	98.80 140	99.56 59	97.38 84	99.14 76	98.62 115	98.51 142	97.85 149	96.20 203	96.80 185	99.04 115	99.08 44

TAPA-MVS	96.65 12	98.23 125	97.96 122	98.55 155	98.81 206	98.16 183	98.40 177	97.94 220	96.68 120	98.49 145	98.61 116	98.89 120	98.57 108	97.45 174	97.59 158	99.09 106	98.35 122
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
MSDG		98.20 128	97.88 127	98.56 154	99.33 155	97.74 202	98.27 190	98.10 214	97.20 99	98.06 172	98.59 117	99.16 86	98.76 96	98.39 97	97.71 149	98.86 141	96.38 195

Anonymous20231211		98.89 62	98.79 46	98.99 107	99.82 25	99.41 40	99.18 98	99.31 131	96.92 105	98.54 138	98.58 118	98.84 125	97.46 160	99.45 27	99.29 33	99.65 29	99.08 44

diffmvs1		98.09 133	97.95 123	98.25 170	99.23 175	98.55 158	98.39 179	99.18 153	97.44 78	97.04 215	98.58 118	98.96 117	97.32 171	96.66 197	96.63 189	98.34 183	98.83 76

OMC-MVS		98.35 118	98.10 112	98.64 148	98.85 204	97.99 191	98.56 166	98.21 211	97.26 93	98.87 117	98.54 120	99.27 70	98.43 117	98.34 102	97.66 151	98.92 128	97.65 164

v1.0		90.99 237	84.28 240	98.83 123	99.56 108	99.21 67	98.66 154	99.47 85	95.22 167	98.35 154	98.48 121	99.67 25	97.84 150	98.80 74	98.57 91	99.10 101	0.00 246

MVS_111021_LR		98.39 116	98.11 111	98.71 137	99.08 193	98.54 162	98.23 193	98.56 199	96.57 128	99.13 77	98.41 122	98.86 122	98.65 103	98.23 114	97.87 131	98.65 166	98.28 128

tfpnnormal		99.19 30	98.90 39	99.54 25	99.81 29	99.55 26	99.60 34	99.54 68	98.53 20	99.23 61	98.40 123	98.23 149	99.40 42	99.29 35	99.36 28	99.63 30	98.95 65

CHOSEN 280x420		96.80 182	96.30 178	97.39 199	99.09 191	96.52 216	98.76 143	99.29 133	93.88 203	97.65 190	98.34 124	93.66 194	96.29 192	98.28 111	97.73 147	93.27 231	95.70 203

PHI-MVS		98.57 102	98.20 106	99.00 104	99.48 134	98.91 127	98.68 147	99.17 154	94.97 175	99.27 60	98.33 125	99.33 63	98.05 140	98.82 72	98.62 86	99.34 79	98.38 120

CANet_DTU		97.65 154	97.50 142	97.82 193	99.19 180	98.08 186	98.41 176	98.67 192	94.40 191	99.16 71	98.32 126	98.69 132	93.96 221	97.87 141	97.61 156	97.51 203	97.56 168

MVS_0304		98.57 102	98.36 94	98.82 126	99.72 60	98.94 123	98.92 125	99.14 159	96.76 116	99.33 43	98.30 127	99.73 11	96.74 179	98.05 128	97.79 135	99.08 107	98.97 59

Vis-MVSNet (Re-imp)		98.46 113	98.23 105	98.73 133	99.81 29	99.29 56	98.79 141	99.50 79	96.20 145	96.03 223	98.29 128	96.98 176	98.54 112	99.11 42	99.08 46	99.70 23	98.62 100

pmmvs3		96.30 194	95.87 190	96.80 217	97.66 239	96.48 217	97.93 202	93.80 240	93.40 209	98.54 138	98.27 129	97.50 168	97.37 169	97.49 172	93.11 222	95.52 221	94.85 212

ACMMP_Plus		98.94 53	98.72 52	99.21 74	99.67 72	99.08 87	99.26 85	99.39 96	96.84 108	98.88 114	98.22 130	99.68 21	98.82 92	99.06 48	98.90 59	99.25 92	99.25 26

DeepC-MVS_fast	97.38 8	98.65 93	98.34 96	99.02 101	99.33 155	98.29 173	98.99 116	98.71 190	97.40 82	99.31 47	98.20 131	99.40 54	98.54 112	98.33 105	98.18 110	99.23 95	98.58 104
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
HSP-MVS		98.50 108	98.05 116	99.03 98	99.67 72	99.33 50	99.51 51	99.26 137	95.28 166	98.51 141	98.19 132	99.74 10	98.29 126	97.69 158	96.70 186	98.96 121	99.41 20

MVE	82.47 18	93.12 227	94.09 204	91.99 239	90.79 246	82.50 248	93.93 245	96.30 233	96.06 151	88.81 246	98.19 132	96.38 182	97.56 158	97.24 186	95.18 211	84.58 245	93.07 221
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
GBi-Net		97.69 151	97.75 131	97.62 195	98.71 209	99.21 67	98.62 158	99.33 124	94.09 198	95.60 229	98.17 134	95.97 184	94.39 214	99.05 49	99.03 52	99.08 107	98.70 94

test1		97.69 151	97.75 131	97.62 195	98.71 209	99.21 67	98.62 158	99.33 124	94.09 198	95.60 229	98.17 134	95.97 184	94.39 214	99.05 49	99.03 52	99.08 107	98.70 94

FMVSNet3		96.85 180	96.67 166	97.06 207	97.56 240	99.01 107	97.99 200	99.33 124	94.09 198	95.60 229	98.17 134	95.97 184	93.26 226	94.76 225	96.22 196	98.59 171	98.46 113

casdiffmvs		97.89 141	97.17 154	98.73 133	99.41 145	98.79 136	98.49 170	99.52 75	95.60 161	98.88 114	98.09 137	97.63 167	97.33 170	95.28 218	96.20 198	98.77 157	98.60 102

HFP-MVS		98.97 49	98.70 54	99.29 69	99.67 72	98.98 110	99.13 103	99.53 72	97.76 49	98.90 110	98.07 138	99.50 48	99.14 76	98.64 81	98.78 71	99.37 73	99.18 34

MVS_111021_HR		98.58 101	98.26 102	98.96 110	99.32 158	98.81 134	98.48 171	98.99 175	96.81 113	99.16 71	98.07 138	99.23 74	98.89 89	98.43 93	98.27 105	98.90 133	98.24 132

Fast-Effi-MVS+		98.42 115	97.79 129	99.15 82	99.69 68	98.66 149	98.94 122	99.68 37	94.49 185	99.05 91	98.06 140	98.86 122	98.48 115	98.18 117	97.78 136	99.05 114	98.54 110

SD-MVS		98.73 88	98.54 68	98.95 111	99.14 186	98.76 139	98.46 173	99.14 159	97.71 58	98.56 135	98.06 140	99.61 36	98.85 91	98.56 83	97.74 145	99.54 40	99.32 23

GA-MVS		96.84 181	95.86 191	97.98 186	99.16 184	98.29 173	97.91 204	98.64 195	95.14 169	97.71 189	98.04 142	88.90 203	96.50 184	96.41 200	96.61 191	97.97 197	97.60 165

MIMVSNet		97.24 170	97.15 157	97.36 201	99.03 197	98.52 163	98.55 167	99.73 27	94.94 177	94.94 237	97.98 143	97.37 171	93.66 223	97.60 166	97.34 169	98.23 189	96.29 196

PCF-MVS	95.58 16	97.60 156	96.67 166	98.69 140	99.44 140	98.23 180	98.37 181	98.81 184	93.01 214	98.22 163	97.97 144	99.59 39	98.20 134	95.72 213	95.08 214	99.08 107	97.09 188
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
diffmvs		97.72 150	97.44 143	98.04 184	99.15 185	98.43 169	97.93 202	99.21 145	96.18 146	97.46 197	97.96 145	98.71 131	96.41 186	96.34 201	95.84 204	98.10 193	98.62 100

Anonymous202405211			98.44 86		99.79 34	99.32 54	99.05 109	99.34 123	96.59 125		97.95 146	97.68 165	97.16 173	99.36 31	99.28 34	99.61 35	98.90 71

tpmrst		92.45 230	89.48 233	95.92 230	98.43 227	95.03 239	97.14 230	97.92 221	94.16 197	97.56 194	97.86 147	81.63 233	93.56 224	85.89 242	82.86 238	90.91 243	88.95 240

LP		95.33 210	93.45 214	97.54 198	98.68 213	97.40 208	98.73 144	98.41 205	96.33 140	98.92 108	97.84 148	88.30 204	95.92 195	92.98 231	89.38 227	94.56 225	91.90 228

DI_MVS_plusplus_trai		97.57 161	96.55 171	98.77 130	99.55 110	98.76 139	99.22 91	99.00 174	97.08 103	97.95 180	97.78 149	91.35 200	98.02 141	96.20 203	96.81 184	98.87 138	97.87 157

ACMMPR		99.05 42	98.72 52	99.44 35	99.79 34	99.12 84	99.35 73	99.56 59	97.74 54	99.21 62	97.72 150	99.55 41	99.29 59	98.90 67	98.81 67	99.41 66	99.19 33

MS-PatchMatch		97.60 156	97.22 152	98.04 184	98.67 214	97.18 211	97.91 204	98.28 208	95.82 158	98.34 155	97.66 151	98.38 145	97.77 151	97.10 189	97.25 173	97.27 206	97.18 184

PVSNet_BlendedMVS		97.93 139	97.66 134	98.25 170	99.30 161	98.67 147	98.31 185	97.95 218	94.30 195	98.75 123	97.63 152	98.76 127	96.30 190	98.29 108	97.78 136	98.93 125	98.18 140

PVSNet_Blended		97.93 139	97.66 134	98.25 170	99.30 161	98.67 147	98.31 185	97.95 218	94.30 195	98.75 123	97.63 152	98.76 127	96.30 190	98.29 108	97.78 136	98.93 125	98.18 140

FPMVS		96.97 178	97.20 153	96.70 219	97.75 237	96.11 225	97.72 213	95.47 235	97.13 101	98.02 174	97.57 154	96.67 179	92.97 227	99.00 57	98.34 103	98.28 186	95.58 204

IS_MVSNet		98.20 128	98.00 119	98.44 159	99.82 25	99.48 33	99.25 87	99.56 59	95.58 162	93.93 240	97.56 155	96.52 180	98.27 128	99.08 47	99.20 38	99.80 14	98.56 109

SteuartSystems-ACMMP		98.94 53	98.52 72	99.43 38	99.79 34	99.13 82	99.33 77	99.55 62	96.17 147	99.04 94	97.53 156	99.65 30	99.46 31	99.04 53	98.76 73	99.44 58	99.35 22
Skip Steuart: Steuart Systems R&D Blog.
LGP-MVS_train		98.84 75	98.33 97	99.44 35	99.78 37	98.98 110	99.39 66	99.55 62	95.41 164	98.90 110	97.51 157	99.68 21	99.44 36	99.03 54	98.81 67	99.57 38	98.91 69

LS3D		98.79 85	98.52 72	99.12 86	99.64 90	99.09 86	99.24 88	99.46 88	97.75 52	98.93 107	97.47 158	98.23 149	97.98 142	99.36 31	99.30 32	99.46 55	98.42 118

thisisatest0530		97.20 172	95.95 187	98.66 144	99.46 135	98.84 133	98.29 187	99.20 149	94.51 183	98.25 162	97.42 159	85.03 222	97.68 153	98.43 93	98.56 93	99.08 107	98.89 73

MDTV_nov1_ep13		94.47 219	92.15 221	97.17 204	98.54 222	96.42 219	98.10 196	98.89 179	94.49 185	98.02 174	97.41 160	86.49 211	95.56 199	90.85 234	87.95 229	93.91 227	91.45 231

testgi		98.18 131	98.44 86	97.89 189	99.78 37	99.23 63	98.78 142	99.21 145	97.26 93	97.41 199	97.39 161	99.36 61	92.85 228	98.82 72	98.66 84	99.31 84	98.35 122

tttt0517		97.18 173	95.92 189	98.65 147	99.49 131	98.92 125	98.29 187	99.20 149	94.37 193	98.17 165	97.37 162	84.72 224	97.68 153	98.55 84	98.56 93	99.10 101	98.95 65

MVSTER		95.38 208	93.99 209	97.01 211	98.83 205	98.95 119	96.62 236	99.14 159	92.17 220	97.44 198	97.29 163	77.88 236	91.63 236	97.45 174	96.18 200	98.41 181	97.99 152

APD-MVS		98.47 111	97.97 121	99.05 95	99.64 90	98.91 127	98.94 122	99.45 92	94.40 191	98.77 121	97.26 164	99.41 51	98.21 133	98.67 79	98.57 91	99.31 84	98.57 106
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
HPM-MVS++		98.56 105	98.08 114	99.11 88	99.53 120	98.61 153	99.02 115	99.32 129	96.29 143	99.06 89	97.23 165	99.50 48	98.77 95	98.15 121	97.90 125	98.96 121	98.90 71

EPNet_dtu		96.31 193	95.96 186	96.72 218	99.18 181	95.39 236	97.03 233	99.13 163	93.02 213	99.35 40	97.23 165	97.07 175	90.70 237	95.74 212	95.08 214	94.94 223	98.16 143
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
tpm		93.89 224	91.21 226	97.03 209	98.36 228	96.07 226	97.53 225	99.65 43	92.24 218	98.64 130	97.23 165	74.67 240	94.64 212	92.68 232	90.73 225	93.37 230	94.82 213

EPMVS		93.67 226	90.82 229	96.99 212	98.62 217	96.39 220	98.40 177	99.11 164	95.54 163	97.87 182	97.14 168	81.27 234	94.97 207	88.54 239	86.80 232	92.95 233	90.06 235

PMMVS		96.47 189	95.81 192	97.23 203	97.38 242	95.96 229	97.31 227	96.91 232	93.21 211	97.93 181	97.14 168	97.64 166	95.70 197	95.24 219	96.18 200	98.17 191	95.33 208

CP-MVS		98.86 71	98.43 89	99.36 55	99.68 69	98.97 117	99.19 96	99.46 88	96.60 124	99.20 63	97.11 170	99.51 46	99.15 74	98.92 65	98.82 65	99.45 56	99.08 44

ACMP	96.54 13	98.87 66	98.40 91	99.41 41	99.74 53	98.88 131	99.29 80	99.50 79	96.85 107	98.96 101	97.05 171	99.66 26	99.43 38	98.98 58	98.60 87	99.52 45	98.81 81
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
test-LLR		94.79 213	93.71 211	96.06 228	99.20 177	96.16 222	96.31 237	98.50 201	89.98 240	94.08 238	97.01 172	86.43 212	92.20 233	96.76 195	95.31 209	96.05 217	94.31 215

TESTMET0.1,1		94.44 220	93.71 211	95.30 235	97.84 236	96.16 222	96.31 237	95.32 237	89.98 240	94.08 238	97.01 172	86.43 212	92.20 233	96.76 195	95.31 209	96.05 217	94.31 215

test-mter		94.62 215	94.02 208	95.32 234	97.72 238	96.75 214	96.23 239	95.67 234	89.83 243	93.23 244	96.99 174	85.94 218	92.66 230	97.32 182	96.11 202	96.44 213	95.22 209

ADS-MVSNet		94.41 221	92.13 222	97.07 206	98.86 203	96.60 215	98.38 180	98.47 204	96.13 150	98.02 174	96.98 175	87.50 210	95.87 196	89.89 235	87.58 230	92.79 235	90.27 233

MP-MVS		98.78 86	98.30 99	99.34 62	99.75 48	98.95 119	99.26 85	99.46 88	95.78 159	99.17 68	96.98 175	99.72 13	99.06 81	98.84 70	98.74 76	99.33 80	99.11 38
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
canonicalmvs		98.34 119	97.92 125	98.83 123	99.45 137	99.21 67	98.37 181	99.53 72	97.06 104	97.74 188	96.95 177	95.05 190	98.36 121	98.77 76	98.85 62	99.51 50	99.53 8

CR-MVSNet		95.38 208	93.01 216	98.16 179	98.63 216	95.85 231	97.64 218	99.78 18	91.27 234	98.50 142	96.84 178	82.16 230	96.34 188	94.40 226	95.50 207	98.05 195	95.04 210

test0.0.03 1		95.81 203	95.77 193	95.85 232	99.20 177	98.15 184	97.49 226	98.50 201	92.24 218	92.74 245	96.82 179	92.70 197	88.60 241	97.31 184	97.01 182	98.57 173	96.19 199

PGM-MVS		98.69 89	98.09 113	99.39 46	99.76 45	99.07 89	99.30 79	99.51 76	94.76 180	99.18 67	96.70 180	99.51 46	99.20 65	98.79 75	98.71 80	99.39 71	99.11 38

PatchT		95.49 206	93.29 215	98.06 183	98.65 215	96.20 221	98.91 127	99.73 27	92.00 229	98.50 142	96.67 181	83.25 228	96.34 188	94.40 226	95.50 207	96.21 215	95.04 210

zzz-MVS		98.94 53	98.57 66	99.37 53	99.77 39	99.15 80	99.24 88	99.55 62	97.38 84	99.16 71	96.64 182	99.69 18	99.15 74	99.09 45	98.92 58	99.37 73	99.11 38

Effi-MVS+		98.11 132	97.29 147	99.06 92	99.62 97	98.55 158	98.16 195	99.80 14	94.64 181	99.15 74	96.59 183	97.43 169	98.44 116	97.46 173	97.90 125	99.17 98	98.45 115

E-PMN		92.28 233	90.12 230	94.79 236	98.56 220	90.90 244	95.16 242	93.68 241	95.36 165	95.10 236	96.56 184	89.05 202	95.24 203	95.21 220	81.84 241	90.98 241	81.94 241

3Dnovator	98.16 3	98.65 93	98.35 95	99.00 104	99.59 101	98.70 145	98.90 130	99.36 114	97.97 35	99.09 86	96.55 185	99.09 105	97.97 143	98.70 78	98.65 85	99.12 100	98.81 81

Gipuma		99.22 28	98.86 42	99.64 15	99.70 65	99.24 61	99.17 99	99.63 47	99.52 2	99.89 1	96.54 186	99.14 92	99.93 1	99.42 30	99.15 41	99.52 45	99.04 49
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
ACMM	96.66 11	98.90 60	98.44 86	99.44 35	99.74 53	98.95 119	99.47 57	99.55 62	97.66 62	99.09 86	96.43 187	99.41 51	99.35 57	98.95 59	98.67 82	99.45 56	99.03 51
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
train_agg		97.99 135	97.26 148	98.83 123	99.43 142	98.22 181	98.91 127	99.07 167	94.43 189	97.96 179	96.42 188	99.30 67	98.81 93	97.39 178	96.62 190	98.82 146	98.47 112

PatchmatchNet		93.88 225	91.08 228	97.14 205	98.75 208	96.01 228	98.25 191	99.39 96	94.95 176	98.96 101	96.32 189	85.35 221	95.50 200	88.89 237	85.89 235	91.99 239	90.15 234
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
CostFormer		92.75 228	89.49 232	96.55 222	98.78 207	95.83 233	97.55 221	98.59 197	91.83 231	97.34 205	96.31 190	78.53 235	94.50 213	86.14 240	84.92 236	92.54 236	92.84 223

ACMMP		98.82 83	98.33 97	99.39 46	99.77 39	99.14 81	99.37 69	99.54 68	96.47 136	99.03 96	96.26 191	99.52 43	99.28 60	98.92 65	98.80 70	99.37 73	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
CNLPA		97.75 146	97.26 148	98.32 168	98.58 218	97.86 196	97.80 208	98.09 215	96.49 133	98.49 145	96.15 192	98.08 154	98.35 122	98.00 130	97.03 180	98.61 169	97.21 183

MCST-MVS		98.25 124	97.57 139	99.06 92	99.53 120	98.24 179	98.63 156	99.17 154	95.88 155	98.58 134	96.11 193	99.09 105	99.18 67	97.58 168	97.31 171	99.25 92	98.75 90

CDPH-MVS		97.99 135	97.23 151	98.87 118	99.58 103	98.29 173	98.83 135	99.20 149	93.76 204	98.11 170	96.11 193	99.16 86	98.23 132	97.80 150	97.22 175	99.29 87	98.28 128

3Dnovator+	97.85 5	98.61 98	98.14 109	99.15 82	99.62 97	98.37 171	99.10 107	99.51 76	98.04 32	98.98 98	96.07 195	98.75 129	98.55 110	98.51 86	98.40 101	99.17 98	98.82 79

Effi-MVS+-dtu		97.78 145	97.37 145	98.26 169	99.25 172	98.50 164	97.89 206	99.19 152	94.51 183	98.16 167	95.93 196	98.80 126	95.97 194	98.27 113	97.38 167	99.10 101	98.23 133

Anonymous20240521		98.86 71	98.57 66	99.19 78	99.86 21	99.67 14	99.39 66	99.71 32	97.53 71	98.69 128	95.85 197	98.48 143	97.75 152	99.57 20	99.41 24	99.72 22	99.48 16

CNVR-MVS		98.22 127	97.76 130	98.76 131	99.33 155	98.26 177	98.48 171	98.88 180	96.22 144	98.47 149	95.79 198	99.33 63	98.35 122	98.37 98	97.99 119	99.03 117	98.38 120

PatchMatch-RL		97.24 170	96.45 174	98.17 177	98.70 212	97.57 207	97.31 227	98.48 203	94.42 190	98.39 151	95.74 199	96.35 183	97.88 146	97.75 155	97.48 165	98.24 188	95.87 202

EPNet		96.44 190	96.08 183	96.86 214	99.32 158	97.15 212	97.69 216	99.32 129	93.67 205	98.11 170	95.64 200	93.44 195	89.07 240	96.86 192	96.83 183	97.67 199	98.97 59
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
EMVS		91.84 234	89.39 234	94.70 237	98.44 226	90.84 245	95.27 241	93.53 242	95.18 168	95.26 234	95.62 201	87.59 209	94.77 210	94.87 223	80.72 242	90.95 242	80.88 242

tfpn1000		97.10 176	95.97 185	98.41 161	99.64 90	99.30 55	98.89 131	99.49 83	96.49 133	95.97 225	95.31 202	85.62 220	96.92 178	97.86 142	99.13 44	99.53 44	98.11 145

MAR-MVS		97.12 174	96.28 179	98.11 181	98.94 201	97.22 210	97.65 217	99.38 103	90.93 238	98.15 168	95.17 203	97.13 174	96.48 185	97.71 157	97.40 166	98.06 194	98.40 119
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
dps		92.35 232	88.78 236	96.52 223	98.21 234	95.94 230	97.78 209	98.38 206	89.88 242	96.81 219	95.07 204	75.31 238	94.70 211	88.62 238	86.21 234	93.21 232	90.41 232

OpenMVS	97.26 9	97.88 142	97.17 154	98.70 138	99.50 128	98.55 158	98.34 184	99.11 164	93.92 202	98.90 110	95.04 205	98.23 149	97.38 167	98.11 126	98.12 112	98.95 123	98.23 133

Fast-Effi-MVS+-dtu		96.99 177	96.46 173	97.61 197	98.98 199	97.89 194	97.54 222	99.76 21	93.43 208	96.55 221	94.93 206	98.06 155	94.32 217	96.93 191	96.50 193	98.53 175	97.47 169

tpm cat1		91.52 235	87.70 238	95.97 229	98.33 229	94.98 240	97.06 232	98.03 217	92.11 222	98.03 173	94.77 207	77.19 237	92.71 229	83.56 243	82.24 240	91.67 240	89.04 239

HQP-MVS		97.58 160	96.65 169	98.66 144	99.30 161	97.99 191	97.88 207	98.65 193	94.58 182	98.66 129	94.65 208	99.15 90	98.59 106	96.10 206	95.59 206	98.90 133	98.50 111

tfpn_n400		97.59 158	96.36 176	99.01 102	99.66 76	99.19 72	99.21 93	99.55 62	97.62 63	97.77 184	94.60 209	87.78 206	98.27 128	98.44 90	98.72 78	99.62 31	98.21 136

tfpnconf		97.59 158	96.36 176	99.01 102	99.66 76	99.19 72	99.21 93	99.55 62	97.62 63	97.77 184	94.60 209	87.78 206	98.27 128	98.44 90	98.72 78	99.62 31	98.21 136

tfpnview11		97.49 163	96.22 180	98.97 109	99.63 95	99.24 61	99.12 105	99.54 68	96.76 116	97.77 184	94.60 209	87.78 206	98.25 131	97.93 135	99.14 42	99.52 45	98.08 148

abl_6				98.38 164	99.03 197	98.04 188	98.08 198	98.65 193	93.23 210	98.56 135	94.58 212	98.57 140	97.17 172			98.81 147	97.42 172

CMPMVS	74.71 19	96.17 198	96.06 184	96.30 226	97.41 241	94.52 241	94.83 243	95.46 236	91.57 232	97.26 210	94.45 213	98.33 147	94.98 206	98.28 111	97.59 158	97.86 198	97.68 163
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
conf0.05thres1000		97.44 166	95.93 188	99.20 77	99.82 25	99.56 22	99.41 63	99.61 53	97.42 81	98.01 177	94.34 214	82.73 229	98.68 100	99.33 34	99.42 23	99.67 27	98.74 91

X-MVS		98.59 100	97.99 120	99.30 66	99.75 48	99.07 89	99.17 99	99.50 79	96.62 122	98.95 103	93.95 215	99.37 57	99.11 77	98.94 61	98.86 60	99.35 78	99.09 43

RPMNet		94.72 214	92.01 223	97.88 190	98.56 220	95.85 231	97.78 209	99.70 34	91.27 234	98.33 156	93.69 216	81.88 231	94.91 208	92.60 233	94.34 219	98.01 196	94.46 214

testus		96.13 201	95.13 197	97.28 202	99.13 188	97.00 213	96.84 235	97.89 222	90.48 239	97.40 200	93.60 217	96.47 181	95.39 201	96.21 202	96.19 199	97.05 209	95.99 200

MSLP-MVS++		97.99 135	97.64 137	98.40 162	98.91 202	98.47 166	97.12 231	98.78 185	96.49 133	98.48 147	93.57 218	99.12 96	98.51 114	98.31 106	98.58 89	98.58 172	98.95 65

PLC	95.63 15	97.73 149	97.01 161	98.57 153	99.10 190	97.80 198	97.72 213	98.77 186	96.34 139	98.38 152	93.46 219	98.06 155	98.66 102	97.90 138	97.65 153	98.77 157	97.90 156
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
CPTT-MVS		98.28 121	97.51 141	99.16 81	99.54 113	98.78 138	98.96 120	99.36 114	96.30 142	98.89 113	93.10 220	99.30 67	99.20 65	98.35 101	97.96 124	99.03 117	98.82 79

tpmp4_e23		92.43 231	88.82 235	96.64 221	98.46 225	95.17 238	97.61 220	98.85 181	92.42 215	98.18 164	93.03 221	74.92 239	93.80 222	88.91 236	84.60 237	92.95 233	92.66 225

tfpn_ndepth		96.69 184	95.49 196	98.09 182	99.17 182	99.13 82	98.61 161	99.38 103	94.90 178	95.85 227	92.85 222	88.19 205	96.07 193	97.28 185	98.67 82	99.49 53	97.44 170

test2356		92.46 229	88.72 237	96.82 215	98.48 223	95.34 237	96.22 240	98.09 215	87.46 245	96.01 224	92.82 223	64.42 244	95.10 205	94.08 228	94.05 220	97.02 210	92.87 222

AdaColmap		97.57 161	96.57 170	98.74 132	99.25 172	98.01 189	98.36 183	98.98 176	94.44 188	98.47 149	92.44 224	97.91 161	98.62 104	98.19 116	97.74 145	98.73 160	97.28 177

DeepMVS_CX						87.86 247	92.27 247	61.98 243	93.64 206	93.62 241	91.17 225	91.67 199	94.90 209	95.99 209		92.48 237	94.18 217

NCCC		97.84 144	96.96 162	98.87 118	99.39 146	98.27 176	98.46 173	99.02 172	96.78 114	98.73 127	91.12 226	98.91 118	98.57 108	97.83 148	97.49 164	99.04 115	98.33 127

tfpn111		96.48 187	94.67 199	98.59 150	99.37 147	99.18 74	98.68 147	99.39 96	92.02 223	97.21 211	90.63 227	86.34 214	97.45 161	98.15 121	99.08 46	99.43 60	97.28 177

thresconf0.02		95.49 206	92.74 218	98.70 138	99.32 158	98.70 145	98.87 133	99.21 145	95.95 153	97.57 193	90.63 227	73.55 241	97.86 148	96.09 207	97.03 180	99.40 68	97.22 182

DWT-MVSNet_training		91.07 236	86.55 239	96.35 225	98.28 231	95.82 234	98.00 199	95.03 238	91.24 236	97.99 178	90.35 229	63.43 245	95.25 202	86.06 241	86.62 233	93.55 229	92.30 226

tfpn		94.97 211	91.60 224	98.90 114	99.73 58	99.33 50	99.11 106	99.51 76	95.05 170	97.19 214	89.03 230	62.62 247	98.37 120	98.53 85	98.97 56	99.48 54	97.70 162

view800		96.48 187	94.42 200	98.87 118	99.70 65	99.26 58	99.05 109	99.45 92	94.77 179	97.32 206	88.21 231	83.40 227	98.28 127	98.37 98	99.33 30	99.44 58	97.58 167

view600		96.39 191	94.30 201	98.82 126	99.65 82	99.16 79	98.98 117	99.36 114	94.46 187	97.39 202	87.28 232	84.16 225	98.16 135	98.16 118	99.48 19	99.40 68	97.42 172

GG-mvs-BLEND		65.66 240	92.62 219	34.20 242	1.45 250	93.75 243	85.40 248	1.64 247	91.37 233	17.21 249	87.25 233	94.78 191	3.25 247	95.64 214	93.80 221	96.27 214	91.74 230

thres600view7		96.35 192	94.27 202	98.79 129	99.66 76	99.18 74	98.94 122	99.38 103	94.37 193	97.21 211	87.19 234	84.10 226	98.10 136	98.16 118	99.47 20	99.42 63	97.43 171

FMVSNet5		94.57 217	92.77 217	96.67 220	97.88 235	98.72 144	97.54 222	98.70 191	88.64 244	95.11 235	86.90 235	81.77 232	93.27 225	97.92 137	98.07 115	97.50 204	97.34 175

testpf		87.81 238	83.90 241	92.37 238	96.76 245	88.65 246	93.04 246	98.24 210	85.20 246	95.28 233	86.82 236	72.43 242	82.35 243	82.62 244	82.30 239	88.55 244	89.29 238

conf200view11		96.16 200	94.08 205	98.59 150	99.37 147	99.18 74	98.68 147	99.39 96	92.02 223	97.21 211	86.53 237	86.34 214	97.45 161	98.15 121	99.08 46	99.43 60	97.28 177

thres100view900		95.74 204	93.66 213	98.17 177	99.37 147	98.59 154	98.10 196	98.33 207	92.02 223	97.30 207	86.53 237	86.34 214	96.69 180	96.77 194	98.47 98	99.24 94	96.89 189

tfpn200view9		96.17 198	94.08 205	98.60 149	99.37 147	99.18 74	98.68 147	99.39 96	92.02 223	97.30 207	86.53 237	86.34 214	97.45 161	98.15 121	99.08 46	99.43 60	97.28 177

thres400		96.22 197	94.08 205	98.72 135	99.58 103	99.05 93	98.83 135	99.22 143	94.01 201	97.40 200	86.34 240	84.91 223	97.93 144	97.85 145	99.08 46	99.37 73	97.28 177

thres200		96.23 196	94.13 203	98.69 140	99.44 140	99.18 74	98.58 165	99.38 103	93.52 207	97.35 204	86.33 241	85.83 219	97.93 144	98.16 118	98.78 71	99.42 63	97.10 186

tmp_tt				65.28 241	82.24 247	71.50 249	70.81 249	23.21 244	96.14 148	81.70 248	85.98 242	92.44 198	49.84 244	95.81 210	94.36 218	83.86 246

conf0.01		94.53 218	91.09 227	98.53 157	99.29 164	99.05 93	98.68 147	99.35 119	92.02 223	97.04 215	84.45 243	68.52 243	97.45 161	97.79 152	99.08 46	99.41 66	96.70 192

conf0.002		93.97 223	90.06 231	98.52 158	99.26 169	99.02 106	98.68 147	99.33 124	92.02 223	97.01 217	83.82 244	63.41 246	97.45 161	97.73 156	97.98 121	99.40 68	96.47 194

IB-MVS	95.85 14	95.87 202	94.88 198	97.02 210	99.09 191	98.25 178	97.16 229	97.38 227	91.97 230	97.77 184	83.61 245	97.29 172	92.03 235	97.16 187	97.66 151	98.66 164	98.20 139
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
testmvs		9.73 241	13.38 243	5.48 244	3.62 248	4.12 250	6.40 251	3.19 246	14.92 247	7.68 251	22.10 246	13.89 252	6.83 245	13.47 245	10.38 244	5.14 248	14.81 243

test123		9.37 242	12.26 244	6.00 243	3.32 249	4.06 251	6.39 252	3.41 245	13.20 248	10.48 250	16.43 247	16.22 251	6.76 246	11.37 246	10.40 243	5.62 247	14.10 245

sosnet-low-res		0.00 243	0.00 245	0.00 245	0.00 251	0.00 252	0.00 253	0.00 248	0.00 249	0.00 252	0.00 248	0.00 253	0.00 248	0.00 247	0.00 246	0.00 250	0.00 246

sosnet		0.00 243	0.00 245	0.00 245	0.00 251	0.00 252	0.00 253	0.00 248	0.00 249	0.00 252	0.00 248	0.00 253	0.00 248	0.00 247	0.00 246	0.00 250	0.00 246

our_test_3					99.29 164	97.72 203	98.98 117

MTAPA										99.19 65		99.68 21

MTMP										99.20 63		99.54 42

Patchmatch-RL test							32.47 250

XVS					99.77 39	99.07 89	99.46 59			98.95 103		99.37 57				99.33 80

X-MVStestdata					99.77 39	99.07 89	99.46 59			98.95 103		99.37 57				99.33 80

mPP-MVS					99.75 48							99.49 50

NP-MVS									93.07 212

Patchmtry						96.05 227	97.64 218	99.78 18		98.50 142

High-resolution multi-view benchmark