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.

pmmvs6		99.88 7	99.87 1	99.89 12	99.97 1	99.76 18	99.89 8	99.96 14	99.82 2	99.90 14	99.92 26	99.95 22	99.68 39	99.93 4	99.88 2	99.95 9	99.86 12

UA-Net		99.64 39	99.62 24	99.66 84	99.97 1	99.82 6	99.14 178	99.96 14	98.95 110	99.52 150	99.38 126	99.86 70	99.55 71	99.72 36	99.66 25	99.80 83	99.94 1

v7n		99.89 2	99.86 4	99.93 1	99.97 1	99.83 2	99.93 1	99.96 14	99.77 5	99.89 18	99.99 1	99.86 70	99.84 5	99.89 9	99.81 10	99.97 1	99.88 9

v52		99.89 2	99.85 6	99.92 4	99.97 1	99.80 11	99.92 2	99.97 1	99.78 3	99.90 14	99.96 5	99.85 76	99.82 7	99.88 12	99.82 6	99.96 3	99.89 5

V4		99.89 2	99.85 6	99.92 4	99.97 1	99.80 11	99.92 2	99.97 1	99.78 3	99.90 14	99.96 5	99.84 78	99.82 7	99.88 12	99.82 6	99.96 3	99.89 5

SixPastTwentyTwo		99.89 2	99.85 6	99.93 1	99.97 1	99.88 1	99.92 2	99.97 1	99.66 14	99.94 3	99.94 15	99.74 95	99.81 9	99.97 2	99.89 1	99.96 3	99.89 5

LTVRE_ROB	99.39 1	99.90 1	99.87 1	99.93 1	99.97 1	99.82 6	99.91 5	99.92 43	99.75 7	99.93 4	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
FC-MVSNet-test		99.84 9	99.80 9	99.89 12	99.96 8	99.83 2	99.84 17	99.95 28	99.37 57	99.77 75	99.95 10	99.96 13	99.85 3	99.93 4	99.83 3	99.95 9	99.72 47

v748		99.89 2	99.87 1	99.92 4	99.96 8	99.80 11	99.91 5	99.95 28	99.77 5	99.92 8	99.96 5	99.93 39	99.81 9	99.92 7	99.82 6	99.96 3	99.90 2

NR-MVSNet		99.52 68	99.29 84	99.80 38	99.96 8	99.38 107	99.55 99	99.81 120	98.86 121	99.87 28	99.51 111	98.81 171	99.72 34	99.86 17	99.04 101	99.89 32	99.54 95

gm-plane-assit		96.82 217	94.84 224	99.13 181	99.95 11	99.78 16	99.69 70	99.92 43	99.19 78	99.84 44	99.92 26	72.93 242	96.44 221	98.21 211	97.01 215	98.92 206	96.87 223

anonymousdsp		99.87 8	99.86 4	99.88 15	99.95 11	99.75 23	99.90 7	99.96 14	99.69 10	99.83 53	99.96 5	99.99 3	99.74 26	99.95 3	99.83 3	99.91 24	99.88 9

PS-CasMVS		99.73 20	99.59 32	99.90 11	99.95 11	99.80 11	99.85 16	99.97 1	98.95 110	99.86 33	99.73 72	99.36 146	99.81 9	99.83 20	99.67 24	99.95 9	99.83 15

PEN-MVS		99.77 13	99.65 18	99.91 8	99.95 11	99.80 11	99.86 13	99.97 1	99.08 93	99.89 18	99.69 80	99.68 103	99.84 5	99.81 24	99.64 27	99.95 9	99.81 19

TransMVSNet (Re)		99.72 24	99.59 32	99.88 15	99.95 11	99.76 18	99.88 10	99.94 31	99.58 30	99.92 8	99.90 39	98.55 175	99.65 54	99.89 9	99.76 14	99.95 9	99.70 52

DTE-MVSNet		99.75 17	99.61 26	99.92 4	99.95 11	99.81 9	99.86 13	99.96 14	99.18 80	99.92 8	99.66 83	99.45 137	99.85 3	99.80 25	99.56 33	99.96 3	99.79 25

CP-MVSNet		99.68 33	99.51 44	99.89 12	99.95 11	99.76 18	99.83 20	99.96 14	98.83 127	99.84 44	99.65 85	99.09 161	99.80 13	99.78 28	99.62 31	99.95 9	99.82 16

WR-MVS_H		99.73 20	99.61 26	99.88 15	99.95 11	99.82 6	99.83 20	99.96 14	99.01 103	99.84 44	99.71 78	99.41 143	99.74 26	99.77 30	99.70 22	99.95 9	99.82 16

WR-MVS		99.79 11	99.68 16	99.91 8	99.95 11	99.83 2	99.87 12	99.96 14	99.39 56	99.93 4	99.87 50	99.29 154	99.77 17	99.83 20	99.72 20	99.97 1	99.82 16

HyFIR lowres test		99.50 71	99.26 88	99.80 38	99.95 11	99.62 43	99.76 39	99.97 1	99.67 12	99.56 141	99.94 15	98.40 179	99.78 15	98.84 184	98.59 166	99.76 91	99.72 47

FC-MVSNet-train		99.70 29	99.67 17	99.74 68	99.94 21	99.71 28	99.82 24	99.91 48	99.14 89	99.53 144	99.70 79	99.88 64	99.33 107	99.88 12	99.61 32	99.94 16	99.77 30

pm-mvs1		99.77 13	99.69 15	99.86 18	99.94 21	99.68 36	99.84 17	99.93 35	99.59 28	99.87 28	99.92 26	99.21 157	99.65 54	99.88 12	99.77 13	99.93 18	99.78 28

EPP-MVSNet		99.34 106	99.10 123	99.62 97	99.94 21	99.74 25	99.66 73	99.80 126	99.07 96	98.93 206	99.61 91	96.13 196	99.49 88	99.67 42	99.63 29	99.92 22	99.86 12

tfpnnormal		99.74 18	99.63 21	99.86 18	99.93 24	99.75 23	99.80 29	99.89 56	99.31 63	99.88 23	99.43 118	99.66 106	99.77 17	99.80 25	99.71 21	99.92 22	99.76 34

CHOSEN 1792x2688		99.65 37	99.55 38	99.77 52	99.93 24	99.60 48	99.79 31	99.92 43	99.73 8	99.74 91	99.93 19	99.98 4	99.80 13	98.83 185	99.01 105	99.45 168	99.76 34

Anonymous20240521		99.43 85	99.23 93	99.67 82	99.92 26	99.76 18	99.64 76	99.93 35	99.06 98	99.68 117	97.77 207	98.97 166	98.97 153	99.72 36	99.54 41	99.88 36	99.81 19

conf0.05thres1000		98.36 190	97.28 200	99.63 92	99.92 26	99.74 25	99.66 73	99.88 60	98.68 141	98.92 207	97.30 219	86.02 234	99.49 88	99.77 30	99.73 18	99.93 18	99.69 53

Vis-MVSNet		99.76 15	99.78 10	99.75 62	99.92 26	99.77 17	99.83 20	99.85 80	99.43 49	99.85 40	99.84 60	100.00 1	99.13 136	99.83 20	99.66 25	99.90 27	99.90 2
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
IS_MVSNet		99.15 136	99.12 120	99.19 177	99.92 26	99.73 27	99.55 99	99.86 66	98.45 170	96.91 241	98.74 173	98.33 182	99.02 145	99.54 55	99.47 50	99.88 36	99.61 72

Vis-MVSNet (Re-imp)		99.40 94	99.28 86	99.55 111	99.92 26	99.68 36	99.31 152	99.87 62	98.69 140	99.16 194	99.08 154	98.64 174	99.20 123	99.65 46	99.46 52	99.83 72	99.72 47

test20.03		99.68 33	99.60 30	99.76 56	99.91 31	99.70 33	99.68 71	99.87 62	99.05 100	99.88 23	99.92 26	99.88 64	99.50 84	99.77 30	99.42 57	99.75 94	99.49 111

MDA-MVSNet-bldmvs		99.11 139	99.11 122	99.12 183	99.91 31	99.38 107	99.77 36	98.72 229	99.31 63	99.85 40	99.43 118	98.26 184	99.48 92	99.85 18	98.47 171	96.99 224	99.08 176

PVSNet_Blended_VisFu		99.66 36	99.64 19	99.67 82	99.91 31	99.71 28	99.61 85	99.79 129	99.41 52	99.91 12	99.85 57	99.61 110	99.00 146	99.67 42	99.42 57	99.81 80	99.81 19

new-patchmatchnet		98.49 185	97.60 194	99.53 113	99.90 34	99.55 63	99.77 36	99.48 204	99.67 12	99.86 33	99.98 3	99.98 4	99.50 84	96.90 226	91.52 227	98.67 212	95.62 227

testgi		99.43 85	99.47 51	99.38 143	99.90 34	99.67 39	99.30 157	99.73 159	98.64 150	99.53 144	99.52 109	99.90 56	98.08 187	99.65 46	99.40 60	99.75 94	99.55 94

1111		96.83 216	95.02 223	98.95 196	99.90 34	99.57 56	99.62 83	99.97 1	98.58 157	98.06 235	99.87 50	69.04 245	96.43 222	99.36 79	99.14 82	99.73 102	99.54 95

.test1245		79.44 235	75.07 238	84.53 237	99.90 34	99.57 56	99.62 83	99.97 1	98.58 157	98.06 235	99.87 50	69.04 245	96.43 222	99.36 79	24.74 239	13.21 243	34.30 239

DeepC-MVS	99.05 5	99.74 18	99.64 19	99.84 24	99.90 34	99.39 102	99.79 31	99.81 120	99.69 10	99.90 14	99.87 50	99.98 4	99.81 9	99.62 51	99.32 64	99.83 72	99.65 63
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
Anonymous20231211		99.47 78	99.39 69	99.57 106	99.89 39	99.60 48	99.50 113	99.69 169	98.91 116	99.62 127	99.17 144	99.35 149	98.86 162	99.63 48	99.46 52	99.84 61	99.62 71

v1192		99.60 48	99.41 62	99.82 30	99.89 39	99.43 95	99.81 26	99.84 93	99.63 19	99.85 40	99.95 10	99.35 149	99.72 34	99.01 151	98.90 123	99.82 76	99.58 84

v1240		99.58 54	99.38 75	99.82 30	99.89 39	99.49 84	99.82 24	99.83 101	99.63 19	99.86 33	99.96 5	98.92 169	99.75 21	99.15 131	98.96 115	99.76 91	99.56 89

MIMVSNet1		99.79 11	99.75 12	99.84 24	99.89 39	99.83 2	99.84 17	99.89 56	99.31 63	99.93 4	99.92 26	99.97 9	99.68 39	99.89 9	99.64 27	99.82 76	99.66 59

CSCG		99.61 42	99.52 43	99.71 72	99.89 39	99.62 43	99.52 106	99.76 149	99.61 23	99.69 110	99.73 72	99.96 13	99.57 69	99.27 102	98.62 163	99.81 80	99.85 14

tfpn		96.77 219	94.47 226	99.45 130	99.88 44	99.62 43	99.46 126	99.83 101	97.61 207	98.27 233	94.22 230	71.45 244	99.34 106	99.32 87	99.46 52	99.90 27	99.58 84

zzz-MVS		99.51 69	99.36 76	99.68 80	99.88 44	99.38 107	99.53 103	99.84 93	99.11 92	99.59 135	98.93 165	99.95 22	99.58 68	99.44 69	99.21 73	99.65 119	99.52 105

v1921920		99.59 50	99.40 65	99.82 30	99.88 44	99.45 89	99.81 26	99.83 101	99.65 15	99.86 33	99.95 10	99.29 154	99.75 21	98.98 157	98.86 132	99.78 85	99.59 74

v1144		99.61 42	99.43 58	99.82 30	99.88 44	99.41 99	99.76 39	99.86 66	99.64 17	99.84 44	99.95 10	99.49 133	99.74 26	99.00 153	98.93 120	99.84 61	99.58 84

SteuartSystems-ACMMP		99.47 78	99.22 96	99.76 56	99.88 44	99.36 116	99.65 75	99.84 93	98.47 165	99.80 65	98.68 177	99.96 13	99.68 39	99.37 76	99.06 95	99.72 106	99.66 59
Skip Steuart: Steuart Systems R&D Blog.
UGNet		99.40 94	99.61 26	99.16 179	99.88 44	99.64 42	99.61 85	99.77 141	99.31 63	99.63 126	99.33 129	99.93 39	96.46 220	99.63 48	99.53 43	99.63 131	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
ACMH+	98.94 6	99.69 31	99.59 32	99.81 34	99.88 44	99.41 99	99.75 45	99.86 66	99.43 49	99.80 65	99.54 101	99.97 9	99.73 30	99.82 23	99.52 44	99.85 57	99.43 126

Anonymous202405211			99.14 116		99.87 51	99.55 63	99.50 113	99.70 166	98.55 161		98.61 180	98.46 176	98.76 167	99.66 44	99.50 45	99.85 57	99.63 67

pmmvs-eth3d		99.61 42	99.48 47	99.75 62	99.87 51	99.30 134	99.75 45	99.89 56	99.23 70	99.85 40	99.88 48	99.97 9	99.49 88	99.46 63	99.01 105	99.68 113	99.52 105

v144192		99.58 54	99.39 69	99.80 38	99.87 51	99.44 91	99.77 36	99.84 93	99.64 17	99.86 33	99.93 19	99.35 149	99.72 34	98.92 165	98.82 138	99.74 98	99.66 59

v148		99.58 54	99.43 58	99.76 56	99.87 51	99.40 101	99.76 39	99.85 80	99.48 44	99.83 53	99.82 64	99.83 81	99.51 80	99.20 118	98.82 138	99.75 94	99.45 119

v1141		99.58 54	99.39 69	99.80 38	99.87 51	99.39 102	99.74 53	99.85 80	99.58 30	99.84 44	99.92 26	99.49 133	99.68 39	98.98 157	98.83 135	99.84 61	99.52 105

divwei89l23v2f112		99.58 54	99.39 69	99.80 38	99.87 51	99.39 102	99.74 53	99.85 80	99.57 33	99.84 44	99.92 26	99.48 135	99.67 43	98.98 157	98.83 135	99.84 61	99.52 105

v13		99.73 20	99.63 21	99.85 21	99.87 51	99.71 28	99.80 29	99.96 14	99.62 22	99.83 53	99.93 19	99.66 106	99.75 21	99.41 72	99.26 69	99.89 32	99.80 24

v11		99.72 24	99.62 24	99.85 21	99.87 51	99.71 28	99.81 26	99.96 14	99.63 19	99.83 53	99.97 4	99.58 117	99.75 21	99.33 85	99.33 62	99.87 47	99.79 25

v2v482		99.56 65	99.35 78	99.81 34	99.87 51	99.35 120	99.75 45	99.85 80	99.56 35	99.87 28	99.95 10	99.44 139	99.66 51	98.91 168	98.76 147	99.86 53	99.45 119

v1		99.58 54	99.39 69	99.80 38	99.87 51	99.39 102	99.74 53	99.85 80	99.58 30	99.84 44	99.92 26	99.51 128	99.67 43	98.98 157	98.82 138	99.84 61	99.52 105

ACMMPR		99.51 69	99.32 81	99.72 71	99.87 51	99.33 127	99.61 85	99.85 80	99.19 78	99.73 97	98.73 174	99.95 22	99.61 62	99.35 81	99.14 82	99.66 117	99.58 84

TranMVSNet+NR-MVSNet		99.59 50	99.42 61	99.80 38	99.87 51	99.55 63	99.64 76	99.86 66	99.05 100	99.88 23	99.72 75	99.33 152	99.64 57	99.47 61	99.14 82	99.91 24	99.67 57

LGP-MVS_train		99.46 82	99.18 108	99.78 48	99.87 51	99.25 145	99.71 68	99.87 62	98.02 194	99.79 68	98.90 166	99.96 13	99.66 51	99.49 57	99.17 77	99.79 84	99.49 111

IterMVS-LS		99.16 134	98.82 157	99.57 106	99.87 51	99.71 28	99.58 94	99.92 43	99.24 69	99.71 106	99.73 72	95.79 197	98.91 157	98.82 186	98.66 158	99.43 173	99.77 30
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
IterMVS		99.08 143	98.90 146	99.29 158	99.87 51	99.53 68	99.52 106	99.77 141	98.94 112	99.75 85	99.91 35	97.52 192	98.72 169	98.86 178	98.14 190	98.09 217	99.43 126
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
ACMH	99.11 4	99.72 24	99.63 21	99.84 24	99.87 51	99.59 52	99.83 20	99.88 60	99.46 46	99.87 28	99.66 83	99.95 22	99.76 19	99.73 35	99.47 50	99.84 61	99.52 105
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
Gipuma		99.55 67	99.23 93	99.91 8	99.87 51	99.52 74	99.86 13	99.93 35	99.87 1	99.96 2	96.72 222	99.55 121	99.97 1	99.77 30	99.46 52	99.87 47	99.74 40
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
view800		97.89 197	96.56 207	99.45 130	99.86 68	99.57 56	99.42 131	99.80 126	97.50 210	98.40 231	93.78 231	86.63 233	99.31 112	99.24 105	99.68 23	99.89 32	99.54 95

Fast-Effi-MVS+		99.39 96	99.18 108	99.63 92	99.86 68	99.28 139	99.45 128	99.91 48	98.47 165	99.61 129	99.50 113	99.57 119	99.17 124	99.24 105	98.66 158	99.78 85	99.59 74

XVS					99.86 68	99.30 134	99.72 62			99.69 110		99.93 39				99.60 141

X-MVStestdata					99.86 68	99.30 134	99.72 62			99.69 110		99.93 39				99.60 141

v15		99.67 35	99.54 40	99.83 29	99.86 68	99.67 39	99.76 39	99.95 28	99.59 28	99.83 53	99.93 19	99.55 121	99.71 38	99.23 108	99.05 98	99.87 47	99.75 37

v12		99.72 24	99.61 26	99.85 21	99.86 68	99.70 33	99.79 31	99.96 14	99.61 23	99.83 53	99.93 19	99.61 110	99.74 26	99.38 74	99.22 71	99.89 32	99.79 25

V14		99.69 31	99.56 37	99.84 24	99.86 68	99.68 36	99.78 34	99.96 14	99.60 27	99.83 53	99.93 19	99.58 117	99.72 34	99.28 99	99.11 90	99.88 36	99.77 30

V9		99.71 28	99.59 32	99.84 24	99.86 68	99.69 35	99.78 34	99.96 14	99.61 23	99.84 44	99.93 19	99.61 110	99.73 30	99.34 84	99.17 77	99.88 36	99.78 28

PGM-MVS		99.32 111	98.99 137	99.71 72	99.86 68	99.31 132	99.59 90	99.86 66	97.51 209	99.75 85	98.23 195	99.94 33	99.53 76	99.29 94	99.08 93	99.65 119	99.54 95

UniMVSNet_NR-MVSNet		99.41 90	99.12 120	99.76 56	99.86 68	99.48 85	99.50 113	99.81 120	98.84 125	99.89 18	99.45 116	98.32 183	99.59 65	99.22 111	98.89 124	99.90 27	99.63 67

UniMVSNet (Re)		99.50 71	99.29 84	99.75 62	99.86 68	99.47 87	99.51 109	99.82 109	98.90 117	99.89 18	99.64 86	99.00 164	99.55 71	99.32 87	99.08 93	99.90 27	99.59 74

ACMMP		99.36 101	99.06 128	99.71 72	99.86 68	99.36 116	99.63 79	99.85 80	98.33 178	99.72 101	97.73 209	99.94 33	99.53 76	99.37 76	99.13 86	99.65 119	99.56 89
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
ACMM	98.37 12	99.47 78	99.23 93	99.74 68	99.86 68	99.19 158	99.68 71	99.86 66	99.16 85	99.71 106	98.52 185	99.95 22	99.62 61	99.35 81	99.02 103	99.74 98	99.42 130
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
tfpn_n400		99.08 143	98.56 168	99.70 75	99.85 81	99.56 61	99.63 79	99.86 66	99.21 73	99.37 177	98.95 162	94.24 202	99.55 71	99.20 118	99.29 66	99.93 18	99.44 122

tfpnconf		99.08 143	98.56 168	99.70 75	99.85 81	99.56 61	99.63 79	99.86 66	99.21 73	99.37 177	98.95 162	94.24 202	99.55 71	99.20 118	99.29 66	99.93 18	99.44 122

MVS_0304		99.36 101	99.35 78	99.37 146	99.85 81	99.36 116	99.39 136	99.56 191	99.36 59	99.75 85	99.23 139	99.90 56	97.97 193	99.00 153	98.83 135	99.69 112	99.77 30

ACMMP_Plus		99.47 78	99.33 80	99.63 92	99.85 81	99.28 139	99.56 97	99.83 101	98.75 133	99.48 158	99.03 159	99.95 22	99.47 95	99.48 58	99.19 74	99.57 151	99.59 74

X-MVS		99.30 116	98.99 137	99.66 84	99.85 81	99.30 134	99.49 120	99.82 109	98.32 179	99.69 110	97.31 218	99.93 39	99.50 84	99.37 76	99.16 79	99.60 141	99.53 100

APDe-MVS		99.60 48	99.48 47	99.73 70	99.85 81	99.51 81	99.75 45	99.85 80	99.17 81	99.81 63	99.56 98	99.94 33	99.44 96	99.42 71	99.22 71	99.67 115	99.54 95

DU-MVS		99.48 75	99.26 88	99.75 62	99.85 81	99.38 107	99.50 113	99.81 120	98.86 121	99.89 18	99.51 111	98.98 165	99.59 65	99.46 63	98.97 113	99.87 47	99.63 67

Baseline_NR-MVSNet		99.62 40	99.48 47	99.78 48	99.85 81	99.76 18	99.59 90	99.82 109	98.84 125	99.88 23	99.91 35	99.04 163	99.61 62	99.46 63	99.78 12	99.94 16	99.60 73

tfpnview11		99.04 151	98.49 176	99.68 80	99.84 89	99.58 54	99.56 97	99.86 66	98.86 121	99.37 177	98.95 162	94.24 202	99.54 75	98.87 174	99.54 41	99.91 24	99.39 140

tfpn1000		98.73 175	98.07 191	99.50 121	99.84 89	99.61 46	99.48 122	99.84 93	98.71 139	98.74 214	98.71 176	91.70 216	99.17 124	98.81 187	99.55 39	99.90 27	99.43 126

view600		97.88 198	96.55 209	99.44 133	99.84 89	99.52 74	99.38 140	99.76 149	97.36 213	98.50 225	93.29 232	87.31 229	99.26 118	99.13 135	99.76 14	99.88 36	99.48 114

pmmvs5		99.58 54	99.47 51	99.70 75	99.84 89	99.50 82	99.58 94	99.80 126	98.98 108	99.73 97	99.92 26	99.81 84	99.49 88	99.28 99	99.05 98	99.77 89	99.73 43

V42		99.57 61	99.41 62	99.75 62	99.84 89	99.37 113	99.73 57	99.83 101	99.41 52	99.75 85	99.89 42	99.42 141	99.60 64	99.15 131	98.96 115	99.76 91	99.65 63

thres600view7		97.86 200	96.53 213	99.41 138	99.84 89	99.52 74	99.36 144	99.76 149	97.32 214	98.38 232	93.24 233	87.25 230	99.23 121	99.11 137	99.75 16	99.88 36	99.48 114

MP-MVS		99.35 104	99.09 125	99.65 86	99.84 89	99.22 153	99.59 90	99.78 135	98.13 187	99.67 119	98.44 189	99.93 39	99.43 98	99.31 89	99.09 92	99.60 141	99.49 111
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
testmv		99.39 96	99.19 105	99.62 97	99.84 89	99.38 107	99.37 142	99.86 66	98.47 165	99.79 68	99.82 64	99.39 145	99.63 59	99.30 90	98.70 154	99.21 195	99.28 155

test1235678		99.39 96	99.20 101	99.62 97	99.84 89	99.38 107	99.38 140	99.86 66	98.47 165	99.79 68	99.82 64	99.41 143	99.63 59	99.30 90	98.71 152	99.21 195	99.28 155

mPP-MVS					99.84 89							99.92 48

COLMAP_ROB	99.18 2	99.70 29	99.60 30	99.81 34	99.84 89	99.37 113	99.76 39	99.84 93	99.54 39	99.82 60	99.64 86	99.95 22	99.75 21	99.79 27	99.56 33	99.83 72	99.37 145
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
gg-mvs-nofinetune		98.40 189	98.26 182	98.57 212	99.83 100	98.86 189	98.77 216	99.97 1	99.57 33	99.99 1	99.99 1	93.81 205	93.50 236	98.91 168	98.20 186	99.33 185	98.52 199

v7		99.61 42	99.46 54	99.79 45	99.83 100	99.37 113	99.75 45	99.84 93	99.56 35	99.76 78	99.94 15	99.60 114	99.73 30	99.11 137	99.01 105	99.85 57	99.63 67

v10		99.65 37	99.51 44	99.81 34	99.83 100	99.61 46	99.75 45	99.94 31	99.56 35	99.76 78	99.94 15	99.60 114	99.73 30	99.11 137	99.01 105	99.85 57	99.74 40

TDRefinement		99.81 10	99.76 11	99.86 18	99.83 100	99.53 68	99.89 8	99.91 48	99.73 8	99.88 23	99.83 62	99.96 13	99.76 19	99.91 8	99.81 10	99.86 53	99.59 74

ACMP	98.32 13	99.44 84	99.18 108	99.75 62	99.83 100	99.18 159	99.64 76	99.83 101	98.81 129	99.79 68	98.42 191	99.96 13	99.64 57	99.46 63	98.98 112	99.74 98	99.44 122
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
SMA-MVS		99.43 85	99.41 62	99.45 130	99.82 105	99.31 132	99.02 190	99.59 188	99.06 98	99.34 186	99.53 107	99.96 13	99.38 100	99.29 94	99.13 86	99.53 160	99.59 74

HFP-MVS		99.46 82	99.30 83	99.65 86	99.82 105	99.25 145	99.50 113	99.82 109	99.23 70	99.58 139	98.86 167	99.94 33	99.56 70	99.14 134	99.12 89	99.63 131	99.56 89

EU-MVSNet		99.76 15	99.74 13	99.78 48	99.82 105	99.81 9	99.88 10	99.87 62	99.31 63	99.75 85	99.91 35	99.76 94	99.78 15	99.84 19	99.74 17	99.56 154	99.81 19

EG-PatchMatch MVS		99.59 50	99.49 46	99.70 75	99.82 105	99.26 142	99.39 136	99.83 101	98.99 105	99.93 4	99.54 101	99.92 48	99.51 80	99.78 28	99.50 45	99.73 102	99.41 131

thres400		97.82 203	96.47 214	99.40 139	99.81 109	99.44 91	99.29 159	99.69 169	97.15 217	98.57 222	92.82 238	87.96 227	99.16 128	98.96 161	99.55 39	99.86 53	99.41 131

CANet		99.36 101	99.39 69	99.34 154	99.80 110	99.35 120	99.41 134	99.47 205	99.20 75	99.74 91	99.54 101	99.68 103	98.05 190	99.23 108	98.97 113	99.57 151	99.73 43

v1neww		99.57 61	99.40 65	99.77 52	99.80 110	99.34 123	99.72 62	99.82 109	99.49 41	99.76 78	99.89 42	99.50 130	99.67 43	99.10 145	98.89 124	99.84 61	99.59 74

v7new		99.57 61	99.40 65	99.77 52	99.80 110	99.34 123	99.72 62	99.82 109	99.49 41	99.76 78	99.89 42	99.50 130	99.67 43	99.10 145	98.89 124	99.84 61	99.59 74

v17		99.62 40	99.48 47	99.79 45	99.80 110	99.60 48	99.73 57	99.94 31	99.46 46	99.73 97	99.88 48	99.52 126	99.67 43	99.16 130	98.96 115	99.84 61	99.75 37

v8		99.61 42	99.45 55	99.79 45	99.80 110	99.59 52	99.73 57	99.93 35	99.48 44	99.77 75	99.90 39	99.48 135	99.67 43	99.11 137	98.89 124	99.84 61	99.73 43

v6		99.57 61	99.40 65	99.77 52	99.80 110	99.34 123	99.72 62	99.82 109	99.49 41	99.76 78	99.89 42	99.52 126	99.67 43	99.10 145	98.89 124	99.84 61	99.59 74

PMMVS2		99.23 124	99.22 96	99.24 166	99.80 110	99.14 163	99.50 113	99.82 109	99.12 91	98.41 230	99.91 35	99.98 4	98.51 173	99.48 58	98.76 147	99.38 179	98.14 210

CP-MVS		99.41 90	99.20 101	99.65 86	99.80 110	99.23 152	99.44 129	99.75 157	98.60 155	99.74 91	98.66 178	99.93 39	99.48 92	99.33 85	99.16 79	99.73 102	99.48 114

v1.0		91.57 234	84.95 237	99.29 158	99.79 118	99.44 91	99.02 190	99.79 129	97.96 197	99.12 198	99.22 140	99.95 22	98.50 174	99.21 114	98.84 134	99.56 154	0.00 242

Effi-MVS+		99.20 128	98.93 142	99.50 121	99.79 118	99.26 142	98.82 213	99.96 14	98.37 177	99.60 133	99.12 149	98.36 181	99.05 143	98.93 163	98.82 138	99.78 85	99.68 54

Anonymous20231206		99.48 75	99.31 82	99.69 79	99.79 118	99.57 56	99.63 79	99.79 129	98.88 119	99.91 12	99.72 75	99.93 39	99.59 65	99.24 105	98.63 162	99.43 173	99.18 164

DI_MVS_plusplus_trai		98.74 172	98.08 190	99.51 119	99.79 118	99.29 138	99.61 85	99.60 184	99.20 75	99.46 162	99.09 153	92.93 209	98.97 153	98.27 209	98.35 179	99.65 119	99.45 119

v16		99.61 42	99.47 51	99.78 48	99.79 118	99.60 48	99.72 62	99.94 31	99.45 48	99.70 108	99.85 57	99.54 124	99.67 43	99.15 131	98.96 115	99.83 72	99.76 34

v18		99.59 50	99.44 57	99.76 56	99.78 123	99.57 56	99.70 69	99.93 35	99.43 49	99.69 110	99.85 57	99.51 128	99.65 54	99.08 148	98.87 129	99.82 76	99.74 40

test12356		99.12 138	99.03 132	99.23 167	99.78 123	98.95 182	99.10 182	99.72 161	98.26 182	99.81 63	99.87 50	99.20 158	98.06 188	99.47 61	98.80 144	98.91 207	98.67 195

APD-MVS		99.17 131	98.92 143	99.46 128	99.78 123	99.24 150	99.34 148	99.78 135	97.79 202	99.48 158	98.25 194	99.88 64	98.77 166	99.18 126	98.92 121	99.63 131	99.18 164
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
N_pmnet		98.64 180	98.23 185	99.11 186	99.78 123	99.25 145	99.75 45	99.39 215	99.65 15	99.70 108	99.78 69	99.89 59	98.81 164	97.60 219	94.28 222	97.24 223	97.15 222

QAPM		99.41 90	99.21 100	99.64 91	99.78 123	99.16 160	99.51 109	99.85 80	99.20 75	99.72 101	99.43 118	99.81 84	99.25 119	98.87 174	98.71 152	99.71 109	99.30 153

3Dnovator	99.16 3	99.42 88	99.22 96	99.65 86	99.78 123	99.13 166	99.50 113	99.85 80	99.40 54	99.80 65	98.59 181	99.79 91	99.30 114	99.20 118	99.06 95	99.71 109	99.35 148

CHOSEN 280x420		98.99 156	98.91 145	99.07 190	99.77 129	99.26 142	99.55 99	99.92 43	98.62 151	98.67 218	99.62 90	97.20 194	98.44 176	99.50 56	99.18 75	98.08 218	98.99 188

pmmvs4		99.34 106	99.15 115	99.57 106	99.77 129	98.90 184	99.51 109	99.77 141	99.07 96	99.73 97	99.72 75	99.84 78	99.07 140	98.85 180	98.39 177	99.55 158	99.27 157

no-one		99.73 20	99.70 14	99.76 56	99.77 129	99.58 54	99.76 39	99.90 55	99.08 93	99.86 33	99.90 39	99.98 4	99.66 51	99.98 1	99.73 18	99.59 147	99.67 57

3Dnovator+	98.92 7	99.31 114	99.03 132	99.63 92	99.77 129	98.90 184	99.52 106	99.81 120	99.37 57	99.72 101	98.03 203	99.73 98	99.32 110	98.99 156	98.81 143	99.67 115	99.36 146

LS3D		99.39 96	99.28 86	99.52 117	99.77 129	99.39 102	99.55 99	99.82 109	98.93 114	99.64 124	98.52 185	99.67 105	98.58 172	99.74 34	99.63 29	99.75 94	99.06 179

HSP-MVS		99.27 121	99.07 127	99.50 121	99.76 134	99.54 66	99.73 57	99.72 161	98.94 112	99.23 190	98.96 161	99.96 13	98.91 157	98.72 194	97.71 206	99.63 131	99.66 59

OPM-MVS		99.39 96	99.22 96	99.59 100	99.76 134	98.82 190	99.51 109	99.79 129	99.17 81	99.53 144	99.31 133	99.95 22	99.35 102	99.22 111	98.79 146	99.60 141	99.27 157

thres200		97.87 199	96.56 207	99.39 140	99.76 134	99.52 74	99.13 179	99.76 149	96.88 228	98.66 219	92.87 237	88.77 226	99.16 128	99.11 137	99.42 57	99.88 36	99.33 149

PM-MVS		99.49 74	99.43 58	99.57 106	99.76 134	99.34 123	99.53 103	99.77 141	98.93 114	99.75 85	99.46 115	99.83 81	99.11 138	99.72 36	99.29 66	99.49 164	99.46 118

our_test_3					99.75 138	99.11 171	99.74 53

HPM-MVS++		99.23 124	98.98 139	99.53 113	99.75 138	99.02 177	99.44 129	99.77 141	98.65 144	99.52 150	98.72 175	99.92 48	99.33 107	98.77 192	98.40 176	99.40 177	99.36 146

MCST-MVS		99.17 131	98.82 157	99.57 106	99.75 138	98.70 201	99.25 164	99.69 169	98.62 151	99.59 135	98.54 183	99.79 91	99.53 76	98.48 202	98.15 189	99.64 129	99.43 126

CDPH-MVS		99.05 149	98.63 164	99.54 112	99.75 138	98.78 193	99.59 90	99.68 174	97.79 202	99.37 177	98.20 198	99.86 70	99.14 134	98.58 199	98.01 198	99.68 113	99.16 170

casdiffmvs1		99.33 108	99.20 101	99.48 125	99.75 138	99.35 120	99.18 169	99.86 66	99.16 85	99.67 119	99.64 86	99.07 162	98.78 165	98.71 195	98.64 160	99.65 119	99.81 19

MDTV_nov1_ep13_2view		98.73 175	98.31 181	99.22 172	99.75 138	99.24 150	99.75 45	99.93 35	99.31 63	99.84 44	99.86 56	99.81 84	99.31 112	97.40 223	94.77 221	96.73 226	97.81 215

OpenMVS	98.82 8	99.17 131	98.85 153	99.53 113	99.75 138	99.06 174	99.36 144	99.82 109	98.28 181	99.76 78	98.47 187	99.61 110	98.91 157	98.80 188	98.70 154	99.60 141	99.04 184

MSDG		99.32 111	99.09 125	99.58 102	99.75 138	98.74 197	99.36 144	99.54 194	99.14 89	99.72 101	99.24 137	99.89 59	99.51 80	99.30 90	98.76 147	99.62 137	98.54 198

CLD-MVS		99.30 116	99.01 136	99.63 92	99.75 138	98.89 187	99.35 147	99.60 184	98.53 162	99.86 33	99.57 97	99.94 33	99.52 79	98.96 161	98.10 193	99.70 111	99.08 176
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
tfpn111		98.25 191	97.29 199	99.37 146	99.74 147	99.52 74	99.17 171	99.76 149	96.10 236	98.65 220	98.23 195	89.10 222	99.00 146	99.11 137	99.56 33	99.88 36	99.41 131

conf200view11		97.85 201	96.54 210	99.37 146	99.74 147	99.52 74	99.17 171	99.76 149	96.10 236	98.65 220	92.99 234	89.10 222	99.00 146	99.11 137	99.56 33	99.88 36	99.41 131

thres100view900		97.69 207	96.37 215	99.23 167	99.74 147	99.21 156	98.81 214	99.43 210	96.10 236	98.70 216	92.99 234	89.10 222	98.88 160	98.58 199	99.31 65	99.82 76	99.27 157

tfpn200view9		97.85 201	96.54 210	99.38 143	99.74 147	99.52 74	99.17 171	99.76 149	96.10 236	98.70 216	92.99 234	89.10 222	99.00 146	99.11 137	99.56 33	99.88 36	99.41 131

RPSCF		99.48 75	99.45 55	99.52 117	99.73 151	99.33 127	99.13 179	99.77 141	99.33 61	99.47 161	99.39 125	99.92 48	99.36 101	99.63 48	99.13 86	99.63 131	99.41 131

ambc			98.83 154		99.72 152	98.52 210	98.84 210		98.96 109	99.92 8	99.34 128	99.74 95	99.04 144	98.68 196	97.57 210	99.46 166	98.99 188

CPTT-MVS		99.21 126	98.89 147	99.58 102	99.72 152	99.12 169	99.30 157	99.76 149	98.62 151	99.66 122	97.51 212	99.89 59	99.48 92	99.01 151	98.64 160	99.58 149	99.40 138

USDC		99.29 120	98.98 139	99.65 86	99.72 152	98.87 188	99.47 124	99.66 180	99.35 60	99.87 28	99.58 96	99.87 69	99.51 80	98.85 180	97.93 200	99.65 119	98.38 202

conf0.01		96.70 222	94.44 228	99.34 154	99.71 155	99.46 88	99.17 171	99.73 159	96.10 236	98.53 223	91.96 239	75.75 240	99.00 146	98.85 180	99.56 33	99.87 47	99.38 141

thresconf0.02		98.10 193	96.83 204	99.58 102	99.71 155	99.28 139	99.40 135	99.72 161	98.65 144	99.39 174	98.23 195	86.73 232	99.43 98	97.73 218	98.17 188	99.86 53	99.05 181

testus		98.74 172	98.33 180	99.23 167	99.71 155	99.03 175	98.17 235	99.60 184	97.18 216	99.52 150	98.07 201	98.45 177	99.21 122	98.30 206	98.06 196	99.14 202	99.21 162

casdiffmvs		99.09 141	98.86 152	99.36 152	99.71 155	99.21 156	98.95 199	99.85 80	98.65 144	99.68 117	99.56 98	98.38 180	98.36 178	98.25 210	98.24 183	99.58 149	99.73 43

TSAR-MVS + GP.		99.33 108	99.17 112	99.51 119	99.71 155	99.00 178	98.84 210	99.71 164	98.23 183	99.74 91	99.53 107	99.90 56	99.35 102	99.38 74	98.85 133	99.72 106	99.31 151

conf0.002		96.39 225	93.87 230	99.33 156	99.70 160	99.45 89	99.17 171	99.71 164	96.10 236	98.51 224	91.88 240	72.65 243	99.00 146	98.80 188	98.82 138	99.87 47	99.38 141

PCF-MVS	97.86 15	98.95 160	98.53 171	99.44 133	99.70 160	98.80 192	98.96 196	99.69 169	98.65 144	99.59 135	99.33 129	99.94 33	99.12 137	98.01 216	97.11 212	99.59 147	97.83 214
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
TSAR-MVS + MP.		99.56 65	99.54 40	99.58 102	99.69 162	99.14 163	99.73 57	99.45 207	99.50 40	99.35 183	99.60 94	99.93 39	99.50 84	99.56 53	99.37 61	99.77 89	99.64 66

train_agg		98.89 165	98.48 177	99.38 143	99.69 162	98.76 196	99.31 152	99.60 184	97.71 204	98.98 203	97.89 205	99.89 59	99.29 115	98.32 204	97.59 209	99.42 176	99.16 170

LP		97.43 213	96.28 216	98.77 206	99.69 162	98.92 183	99.49 120	99.70 166	98.53 162	99.82 60	99.12 149	95.67 199	97.30 205	94.65 230	91.76 225	96.65 228	95.34 229

PVSNet_BlendedMVS		99.20 128	99.17 112	99.23 167	99.69 162	99.33 127	99.04 185	99.13 224	98.41 174	99.79 68	99.33 129	99.36 146	98.10 185	99.29 94	98.87 129	99.65 119	99.56 89

PVSNet_Blended		99.20 128	99.17 112	99.23 167	99.69 162	99.33 127	99.04 185	99.13 224	98.41 174	99.79 68	99.33 129	99.36 146	98.10 185	99.29 94	98.87 129	99.65 119	99.56 89

TAMVS		99.05 149	99.02 135	99.08 189	99.69 162	99.22 153	99.33 149	99.32 220	99.16 85	98.97 204	99.87 50	97.36 193	97.76 196	99.21 114	99.00 110	99.44 170	99.33 149

PHI-MVS		99.33 108	99.19 105	99.49 124	99.69 162	99.25 145	99.27 161	99.59 188	98.44 171	99.78 74	99.15 145	99.92 48	98.95 156	99.39 73	99.04 101	99.64 129	99.18 164

canonicalmvs		99.00 154	98.68 163	99.37 146	99.68 169	99.42 98	98.94 201	99.89 56	99.00 104	98.99 202	98.43 190	95.69 198	98.96 155	99.18 126	99.18 75	99.74 98	99.88 9

FMVSNet1		99.50 71	99.57 36	99.42 135	99.67 170	99.65 41	99.60 89	99.91 48	99.40 54	99.39 174	99.83 62	99.27 156	98.14 184	99.68 39	99.50 45	99.81 80	99.68 54

ESAPD		99.41 90	99.36 76	99.47 127	99.66 171	99.48 85	99.46 126	99.75 157	98.65 144	99.41 171	99.67 81	99.95 22	98.82 163	99.21 114	99.14 82	99.72 106	99.40 138

SD-MVS		99.35 104	99.26 88	99.46 128	99.66 171	99.15 162	98.92 202	99.67 176	99.55 38	99.35 183	98.83 169	99.91 54	99.35 102	99.19 123	98.53 168	99.78 85	99.68 54

TSAR-MVS + ACMM		99.31 114	99.26 88	99.37 146	99.66 171	98.97 181	99.20 167	99.56 191	99.33 61	99.19 193	99.54 101	99.91 54	99.32 110	99.12 136	98.34 180	99.29 187	99.65 63

pmmvs3		98.85 168	98.60 165	99.13 181	99.66 171	98.72 199	99.37 142	99.06 226	98.44 171	99.76 78	99.74 70	99.55 121	99.15 132	99.04 149	96.00 220	97.80 219	98.72 194

tfpn_ndepth		98.67 179	98.03 192	99.42 135	99.65 175	99.50 82	99.29 159	99.78 135	98.17 186	99.04 200	98.36 192	93.29 207	98.88 160	98.46 203	99.26 69	99.88 36	99.14 173

PatchT		98.11 192	97.12 201	99.26 165	99.65 175	98.34 219	99.57 96	99.97 1	97.48 211	99.43 166	99.04 158	90.84 218	98.15 182	98.04 213	97.78 201	98.82 209	98.30 205

DeepC-MVS_fast	98.69 9	99.32 111	99.13 118	99.53 113	99.63 177	98.78 193	99.53 103	99.33 219	99.08 93	99.77 75	99.18 143	99.89 59	99.29 115	99.00 153	98.70 154	99.65 119	99.30 153
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
test-LLR		97.74 205	97.46 196	98.08 224	99.62 178	98.37 217	98.26 230	99.41 211	97.03 221	97.38 239	99.54 101	92.89 210	95.12 231	98.78 190	97.68 207	98.65 213	97.90 212

test0.0.03 1		98.41 188	98.41 179	98.40 218	99.62 178	99.16 160	98.87 207	99.41 211	97.15 217	96.60 243	99.31 133	97.00 195	96.55 219	98.91 168	98.51 170	99.37 180	98.82 191

diffmvs		98.99 156	98.88 150	99.11 186	99.62 178	99.12 169	98.70 220	99.86 66	98.72 138	99.43 166	99.44 117	99.14 160	97.87 194	98.31 205	97.73 205	99.18 198	99.72 47

NCCC		98.88 166	98.42 178	99.42 135	99.62 178	98.81 191	99.10 182	99.54 194	98.76 131	99.53 144	95.97 225	99.80 89	99.16 128	98.49 201	98.06 196	99.55 158	99.05 181

IB-MVS	98.10 14	97.76 204	97.40 198	98.18 221	99.62 178	99.11 171	98.24 232	98.35 234	96.56 231	99.44 164	91.28 241	98.96 168	93.84 234	98.09 212	98.62 163	99.56 154	99.18 164
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
MVS_Test		99.09 141	98.92 143	99.29 158	99.61 183	99.07 173	99.04 185	99.81 120	98.58 157	99.37 177	99.74 70	98.87 170	98.41 177	98.61 198	98.01 198	99.50 163	99.57 88

CNVR-MVS		99.08 143	98.83 154	99.37 146	99.61 183	98.74 197	99.15 176	99.54 194	98.59 156	99.37 177	98.15 199	99.88 64	99.08 139	98.91 168	98.46 172	99.48 165	99.06 179

HQP-MVS		98.70 178	98.19 186	99.28 163	99.61 183	98.52 210	98.71 219	99.35 216	97.97 196	99.53 144	97.38 216	99.85 76	99.14 134	97.53 220	96.85 217	99.36 181	99.26 160

TinyColmap		99.21 126	98.89 147	99.59 100	99.61 183	98.61 206	99.47 124	99.67 176	99.02 102	99.82 60	99.15 145	99.74 95	99.35 102	99.17 128	98.33 181	99.63 131	98.22 208

Effi-MVS+-dtu		99.01 153	99.05 129	98.98 193	99.60 187	99.13 166	99.03 189	99.61 182	98.52 164	99.01 201	98.53 184	99.83 81	96.95 213	99.48 58	98.59 166	99.66 117	99.25 161

EPNet		98.06 195	98.11 189	98.00 227	99.60 187	98.99 180	98.38 228	99.68 174	98.18 185	98.85 211	97.89 205	95.60 200	92.72 238	98.30 206	98.10 193	98.76 210	99.72 47
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
CR-MVSNet		97.91 196	96.80 205	99.22 172	99.60 187	98.23 223	98.91 203	99.97 1	96.89 226	99.43 166	99.10 152	89.24 221	98.15 182	98.04 213	97.78 201	99.26 190	98.30 205

PMVS	94.32 17	99.27 121	99.55 38	98.94 197	99.60 187	99.43 95	99.39 136	99.54 194	98.99 105	99.69 110	99.60 94	99.81 84	95.68 228	99.88 12	99.83 3	99.73 102	99.31 151
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
GA-MVS		98.59 183	98.15 187	99.09 188	99.59 191	99.13 166	98.84 210	99.52 200	98.61 154	99.35 183	99.67 81	93.03 208	97.73 198	98.90 172	98.26 182	99.51 162	99.48 114

ADS-MVSNet		97.29 215	96.17 218	98.59 211	99.59 191	98.70 201	99.32 150	99.86 66	98.47 165	99.56 141	99.08 154	98.16 185	97.34 204	92.92 231	91.17 228	95.91 230	94.72 231

CDS-MVSNet		99.15 136	99.10 123	99.21 174	99.59 191	99.22 153	99.48 122	99.47 205	98.89 118	99.41 171	99.84 60	98.11 186	97.76 196	99.26 104	99.01 105	99.57 151	99.38 141
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
CANet_DTU		99.03 152	99.18 108	98.87 202	99.58 194	99.03 175	99.18 169	99.41 211	98.65 144	99.74 91	99.55 100	99.71 100	96.13 226	99.19 123	98.92 121	99.17 199	99.18 164

EPNet_dtu		98.09 194	98.25 183	97.91 228	99.58 194	98.02 231	98.19 234	99.67 176	97.94 198	99.74 91	99.07 156	98.71 173	93.40 237	97.50 221	97.09 213	96.89 225	99.44 122
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
MVS_111021_HR		99.30 116	99.14 116	99.48 125	99.58 194	99.25 145	99.27 161	99.61 182	98.74 134	99.66 122	99.02 160	99.84 78	99.33 107	99.20 118	98.76 147	99.44 170	99.18 164

RPMNet		97.70 206	96.54 210	99.06 191	99.57 197	98.23 223	98.95 199	99.97 1	96.89 226	99.49 157	99.13 147	89.63 220	97.09 209	96.68 227	97.02 214	99.26 190	98.19 209

MS-PatchMatch		98.94 161	98.71 162	99.21 174	99.52 198	98.22 226	98.97 195	99.53 199	98.76 131	99.50 156	98.59 181	99.56 120	98.68 170	98.63 197	98.45 174	99.05 204	98.73 192

CVMVSNet		99.06 148	98.88 150	99.28 163	99.52 198	99.53 68	99.42 131	99.69 169	98.74 134	98.27 233	99.89 42	95.48 201	99.44 96	99.46 63	99.33 62	99.32 186	99.75 37

tpmrst		96.18 227	94.47 226	98.18 221	99.52 198	97.89 234	98.96 196	99.79 129	98.07 192	99.16 194	99.30 136	92.69 214	96.69 217	90.76 236	88.85 235	94.96 235	93.69 235

DELS-MVS		99.42 88	99.53 42	99.29 158	99.52 198	99.43 95	99.42 131	99.28 221	99.16 85	99.72 101	99.82 64	99.97 9	98.17 181	99.56 53	99.16 79	99.65 119	99.59 74
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
AdaColmap		98.93 162	98.53 171	99.39 140	99.52 198	98.65 204	99.11 181	99.59 188	98.08 191	99.44 164	97.46 215	99.45 137	99.24 120	98.92 165	98.44 175	99.44 170	98.73 192

Fast-Effi-MVS+-dtu		98.82 169	98.80 159	98.84 205	99.51 203	98.90 184	98.96 196	99.91 48	98.29 180	99.11 199	98.47 187	99.63 109	96.03 227	99.21 114	98.12 191	99.52 161	99.01 185

new_pmnet		98.91 164	98.89 147	98.94 197	99.51 203	98.27 222	99.15 176	98.66 230	99.17 81	99.48 158	99.79 68	99.80 89	98.49 175	99.23 108	98.20 186	98.34 215	97.74 218

MVS_111021_LR		99.25 123	99.13 118	99.39 140	99.50 205	99.14 163	99.23 165	99.50 202	98.67 142	99.61 129	99.12 149	99.81 84	99.16 128	99.28 99	98.67 157	99.35 183	99.21 162

abl_6				99.21 174	99.49 206	98.62 205	98.90 205	99.44 209	97.08 220	99.61 129	97.19 220	99.73 98	98.35 179			99.45 168	98.84 190

CostFormer		95.61 228	93.35 233	98.24 220	99.48 207	98.03 230	98.65 221	99.83 101	96.93 224	99.42 170	98.83 169	83.65 236	97.08 210	90.39 237	89.54 234	94.94 236	96.11 226

MDTV_nov1_ep13		97.41 214	96.26 217	98.76 207	99.47 208	98.43 215	99.26 163	99.82 109	98.06 193	99.23 190	99.22 140	92.86 212	98.05 190	95.33 229	93.66 224	96.73 226	96.26 224

MVSTER		97.55 212	96.75 206	98.48 215	99.46 209	99.54 66	98.24 232	99.77 141	97.56 208	99.41 171	99.31 133	84.86 235	94.66 233	98.86 178	97.75 203	99.34 184	99.38 141

E-PMN		96.72 221	95.78 219	97.81 230	99.45 210	95.46 241	98.14 238	98.33 236	97.99 195	98.73 215	98.09 200	98.97 166	97.54 201	97.45 222	91.09 229	94.70 238	91.40 237

PLC	97.83 16	98.88 166	98.52 173	99.30 157	99.45 210	98.60 207	98.65 221	99.49 203	98.66 143	99.59 135	96.33 223	99.59 116	99.17 124	98.87 174	98.53 168	99.46 166	99.05 181
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
MSLP-MVS++		98.92 163	98.73 161	99.14 180	99.44 212	99.00 178	98.36 229	99.35 216	98.82 128	99.38 176	96.06 224	99.79 91	99.07 140	98.88 173	99.05 98	99.27 189	99.53 100

TAPA-MVS	98.54 10	99.30 116	99.24 92	99.36 152	99.44 212	98.77 195	99.00 193	99.41 211	99.23 70	99.60 133	99.50 113	99.86 70	99.15 132	99.29 94	98.95 119	99.56 154	99.08 176
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
PatchmatchNet		96.81 218	95.41 220	98.43 217	99.43 214	98.30 220	99.23 165	99.93 35	98.19 184	99.64 124	98.81 171	93.50 206	97.43 203	92.89 233	90.78 230	94.94 236	95.41 228
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
EMVS		96.47 224	95.38 221	97.74 233	99.42 215	95.37 242	98.07 240	98.27 237	97.85 200	98.90 208	97.48 213	98.73 172	97.20 206	97.21 224	90.39 231	94.59 240	90.65 238

EPMVS		96.76 220	95.30 222	98.46 216	99.42 215	98.47 213	99.32 150	99.91 48	98.42 173	99.51 154	99.07 156	92.81 213	97.12 208	92.39 234	91.71 226	95.51 232	94.20 233

test2356		96.34 226	94.05 229	99.00 192	99.39 217	98.28 221	98.15 236	99.51 201	96.23 233	99.16 194	97.95 204	73.39 241	98.75 168	97.07 225	96.86 216	99.06 203	98.57 196

PatchMatch-RL		98.80 171	98.52 173	99.12 183	99.38 218	98.70 201	98.56 224	99.55 193	97.81 201	99.34 186	97.57 210	99.31 153	98.67 171	99.27 102	98.62 163	99.22 194	98.35 204

DWT-MVSNet_training		94.92 232	92.14 235	98.15 223	99.37 219	98.43 215	98.99 194	98.51 231	96.76 230	99.52 150	97.35 217	77.20 239	97.08 210	89.76 238	90.38 232	95.43 233	95.13 230

OMC-MVS		99.11 139	98.95 141	99.29 158	99.37 219	98.57 208	99.19 168	99.20 223	98.87 120	99.58 139	99.13 147	99.88 64	99.00 146	99.19 123	98.46 172	99.43 173	98.57 196

MAR-MVS		98.54 184	98.15 187	98.98 193	99.37 219	98.09 229	98.56 224	99.65 181	96.11 235	99.27 188	97.16 221	99.50 130	98.03 192	98.87 174	98.23 184	99.01 205	99.13 174
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		95.59 229	93.46 232	98.08 224	99.33 222	98.22 226	98.87 207	99.70 166	96.17 234	98.87 210	97.75 208	86.85 231	96.60 218	91.24 235	89.62 233	95.10 234	94.34 232

CMPMVS	76.62 19	98.64 180	98.60 165	98.68 210	99.33 222	97.07 238	98.11 239	98.50 232	97.69 205	99.26 189	98.35 193	99.66 106	97.62 199	99.43 70	99.02 103	99.24 192	99.01 185
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
tpmp4_e23		95.42 231	92.99 234	98.27 219	99.32 224	97.77 237	98.74 217	99.79 129	97.11 219	99.61 129	97.47 214	80.64 238	96.36 224	92.92 231	88.79 236	95.80 231	96.19 225

MIMVSNet		99.00 154	99.03 132	98.97 195	99.32 224	99.32 131	99.39 136	99.91 48	98.41 174	98.76 213	99.24 137	99.17 159	97.13 207	99.30 90	98.80 144	99.29 187	99.01 185

TSAR-MVS + COLMAP		98.74 172	98.58 167	98.93 199	99.29 226	98.23 223	99.04 185	99.24 222	98.79 130	98.80 212	99.37 127	99.71 100	98.06 188	98.02 215	97.46 211	99.16 200	98.48 200

tpm cat1		95.52 230	93.49 231	97.88 229	99.28 227	97.87 235	98.65 221	99.77 141	97.27 215	99.46 162	98.04 202	90.99 217	95.46 229	88.57 240	88.14 238	94.64 239	93.54 236

PMMVS		98.71 177	98.55 170	98.90 201	99.28 227	98.45 214	98.53 227	99.45 207	97.67 206	99.15 197	98.76 172	99.54 124	97.79 195	98.77 192	98.23 184	99.16 200	98.46 201

CNLPA		98.82 169	98.52 173	99.18 178	99.21 229	98.50 212	98.73 218	99.34 218	98.73 136	99.56 141	97.55 211	99.42 141	99.06 142	98.93 163	98.10 193	99.21 195	98.38 202

DeepPCF-MVS	98.38 11	99.16 134	99.20 101	99.12 183	99.20 230	98.71 200	98.85 209	99.06 226	99.17 81	98.96 205	99.61 91	99.86 70	99.29 115	99.17 128	98.72 151	99.36 181	99.15 172

tpm		96.56 223	94.68 225	98.74 208	99.12 231	97.90 233	98.79 215	99.93 35	96.79 229	99.69 110	99.19 142	81.48 237	97.56 200	95.46 228	93.97 223	97.37 222	97.99 211

MVS-HIRNet		98.45 187	98.25 183	98.69 209	99.12 231	97.81 236	98.55 226	99.85 80	98.58 157	99.67 119	99.61 91	99.86 70	97.46 202	97.95 217	96.37 219	97.49 221	97.56 219

FPMVS		98.48 186	98.83 154	98.07 226	99.09 233	97.98 232	99.07 184	98.04 238	98.99 105	99.22 192	98.85 168	99.43 140	93.79 235	99.66 44	99.11 90	99.24 192	97.76 216

TESTMET0.1,1		97.62 211	97.46 196	97.81 230	99.07 234	98.37 217	98.26 230	98.35 234	97.03 221	97.38 239	99.54 101	92.89 210	95.12 231	98.78 190	97.68 207	98.65 213	97.90 212

GBi-Net		98.96 158	99.05 129	98.85 203	99.02 235	99.53 68	99.31 152	99.78 135	98.13 187	98.48 226	99.43 118	97.58 189	96.92 214	99.68 39	99.50 45	99.61 138	99.53 100

test1		98.96 158	99.05 129	98.85 203	99.02 235	99.53 68	99.31 152	99.78 135	98.13 187	98.48 226	99.43 118	97.58 189	96.92 214	99.68 39	99.50 45	99.61 138	99.53 100

FMVSNet2		99.07 147	99.19 105	98.93 199	99.02 235	99.53 68	99.31 152	99.84 93	98.86 121	98.88 209	99.64 86	98.44 178	96.92 214	99.35 81	99.00 110	99.61 138	99.53 100

test-mter		97.65 210	97.57 195	97.75 232	98.90 238	98.56 209	98.15 236	98.45 233	96.92 225	96.84 242	99.52 109	92.53 215	95.24 230	99.04 149	98.12 191	98.90 208	98.29 207

testpf		93.65 233	91.79 236	95.82 235	98.71 239	93.25 243	96.38 243	99.67 176	95.38 242	97.83 238	94.48 229	87.69 228	89.61 240	88.96 239	88.79 236	92.71 242	93.97 234

FMVSNet5		97.69 207	96.98 202	98.53 213	98.53 240	99.36 116	98.90 205	99.54 194	96.38 232	98.44 229	95.38 227	90.08 219	97.05 212	99.46 63	99.06 95	98.73 211	99.12 175

MVE	91.08 18	97.68 209	97.65 193	97.71 234	98.46 241	91.62 245	97.92 241	98.86 228	98.73 136	97.99 237	98.64 179	99.96 13	99.17 124	99.59 52	97.75 203	93.87 241	97.27 220
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
FMVSNet3		98.63 182	98.75 160	98.49 214	98.10 242	99.44 91	99.02 190	99.78 135	98.13 187	98.48 226	99.43 118	97.58 189	96.16 225	98.85 180	98.39 177	99.40 177	99.41 131

tmp_tt				88.14 236	96.68 243	91.91 244	93.70 244	61.38 240	99.61 23	90.51 244	99.40 124	99.71 100	90.32 239	99.22 111	99.44 56	96.25 229

testmvs		22.33 237	29.66 239	13.79 239	8.97 244	10.35 246	15.53 248	8.09 242	32.51 243	19.87 246	45.18 242	30.56 248	17.05 242	29.96 241	24.74 239	13.21 243	34.30 239

test123		21.52 238	28.47 240	13.42 240	7.29 245	10.12 247	15.70 247	8.31 241	31.54 244	19.34 247	36.33 243	37.40 247	17.14 241	27.45 242	23.17 241	12.73 245	33.30 241

GG-mvs-BLEND		70.44 236	96.91 203	39.57 238	3.32 246	96.51 239	91.01 245	4.05 243	97.03 221	33.20 245	94.67 228	97.75 188	7.59 243	98.28 208	96.85 217	98.24 216	97.26 221

sosnet-low-res		0.00 239	0.00 241	0.00 241	0.00 247	0.00 248	0.00 249	0.00 244	0.00 245	0.00 248	0.00 244	0.00 249	0.00 244	0.00 243	0.00 242	0.00 246	0.00 242

sosnet		0.00 239	0.00 241	0.00 241	0.00 247	0.00 248	0.00 249	0.00 244	0.00 245	0.00 248	0.00 244	0.00 249	0.00 244	0.00 243	0.00 242	0.00 246	0.00 242

MTAPA										99.62 127		99.95 22

MTMP										99.53 144		99.92 48

Patchmatch-RL test							65.75 246

NP-MVS									97.37 212

Patchmtry						98.19 228	98.91 203	99.97 1		99.43 166

DeepMVS_CX						96.39 240	97.15 242	88.89 239	97.94 198	99.51 154	95.71 226	97.88 187	98.19 180	98.92 165		97.73 220	97.75 217

High-resolution multi-view benchmark