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.

USDC		92.17 132	92.17 137	92.18 119	92.93 161	92.22 165	93.66 151	87.41 174	93.49 46	97.99 1	94.10 122	96.68 99	86.46 137	92.04 163	89.18 186	94.61 164	87.47 179

Gipuma		95.86 37	96.17 39	95.50 28	95.92 60	94.59 109	94.77 127	92.50 50	97.82 5	97.90 2	95.56 90	97.88 66	94.71 14	98.02 24	94.81 90	97.23 63	94.48 76
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
LTVRE_ROB	95.06 1	97.73 1	98.39 1	96.95 1	96.33 48	96.94 33	98.30 22	94.90 14	98.61 1	97.73 3	97.97 31	98.57 27	95.74 7	99.24 1	98.70 4	98.72 7	98.70 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
TinyColmap		93.17 106	93.33 122	93.00 107	93.84 122	92.76 151	94.75 129	88.90 146	93.97 41	97.48 4	95.28 100	95.29 135	88.37 122	95.31 92	91.58 155	94.65 159	89.10 163

v52		96.35 16	97.40 13	95.12 40	93.83 123	95.54 69	97.82 40	88.95 144	96.27 13	97.22 5	99.11 2	99.40 4	95.80 5	98.16 18	96.37 50	97.10 67	96.96 23

V4		96.35 16	97.40 13	95.12 40	93.83 123	95.54 69	97.82 40	88.95 144	96.27 13	97.21 6	99.10 3	99.40 4	95.79 6	98.17 17	96.37 50	97.10 67	96.96 23

v7n		96.49 14	97.20 17	95.65 23	95.57 71	96.04 54	97.93 36	92.49 51	96.40 11	97.13 7	98.99 4	99.41 3	93.79 28	97.84 33	96.15 56	97.00 73	95.60 54

SixPastTwentyTwo		97.36 4	97.73 9	96.92 2	97.36 13	96.15 52	98.29 23	94.43 22	96.50 10	96.96 8	98.74 7	98.74 19	96.04 3	99.03 5	97.74 16	98.44 22	97.22 12

v1240		93.89 77	93.72 103	94.09 59	93.98 112	94.31 113	97.12 49	89.37 133	90.74 119	96.92 9	98.05 29	97.89 63	92.15 55	91.53 172	91.60 154	94.99 144	91.93 134

v1921920		93.90 76	93.82 98	94.00 65	93.74 130	94.31 113	97.12 49	89.33 135	91.13 108	96.77 10	97.90 33	98.06 54	91.95 56	91.93 168	91.54 156	95.10 138	91.85 135

v748		96.05 28	97.00 19	94.95 46	94.41 95	94.77 102	96.72 67	91.03 91	96.12 15	96.71 11	98.74 7	99.59 1	93.55 31	97.97 28	95.96 60	97.28 60	95.84 49

v1192		93.98 71	93.94 93	94.01 64	93.91 119	94.63 105	97.00 54	89.75 118	91.01 111	96.50 12	97.93 32	98.26 41	91.74 62	92.06 161	92.05 143	95.18 135	91.66 140

v144192		93.89 77	93.85 96	93.94 70	93.50 139	94.33 112	97.12 49	89.49 129	90.89 114	96.49 13	97.78 41	98.27 40	91.89 58	92.17 160	91.70 151	95.19 134	91.78 138

ambc			94.61 77		98.09 5	95.14 87	91.71 192		94.18 37	96.46 14	96.26 78	96.30 107	91.26 76	94.70 105	92.00 146	93.45 180	93.67 90

TDRefinement		97.59 2	98.32 2	96.73 4	95.90 61	98.10 2	99.08 2	93.92 31	98.24 3	96.44 15	98.12 26	97.86 68	96.06 2	99.24 1	98.93 1	99.00 2	97.77 5

anonymousdsp		95.45 45	96.70 25	93.99 66	88.43 218	92.05 169	99.18 1	85.42 198	94.29 35	96.10 16	98.63 12	99.08 10	96.11 1	97.77 36	97.41 29	98.70 8	97.69 6

COLMAP_ROB	93.74 2	97.09 8	97.98 4	96.05 18	95.97 58	97.78 9	98.56 12	91.72 73	97.53 6	96.01 17	98.14 25	98.76 18	95.28 8	98.76 11	98.23 10	98.77 5	96.67 35
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
PHI-MVS		94.65 58	94.84 69	94.44 52	94.95 82	96.55 39	96.46 81	91.10 89	88.96 141	96.00 18	94.55 114	95.32 134	90.67 97	96.97 53	96.69 44	97.44 54	94.84 65

v1144		93.83 81	93.87 94	93.78 77	93.72 131	94.57 110	96.85 61	89.98 111	91.31 103	95.90 19	97.89 34	98.40 35	91.13 82	92.01 164	92.01 145	95.10 138	90.94 144

v11		94.32 66	94.62 76	93.97 67	93.95 115	95.31 80	96.83 63	91.30 84	91.95 80	95.51 20	98.32 20	98.61 25	91.44 70	92.83 142	92.23 138	94.77 153	93.08 105

MTMP										95.43 21		97.25 79

ACMH	90.17 8	96.61 11	97.69 10	95.35 31	95.29 75	96.94 33	98.43 16	92.05 65	98.04 4	95.38 22	98.07 28	99.25 6	93.23 36	98.35 15	97.16 36	97.72 48	96.00 45
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
PMVS	87.16 16	95.88 36	96.47 30	95.19 38	97.00 26	96.02 55	96.70 68	91.57 77	94.43 32	95.33 23	97.16 63	95.37 131	92.39 50	98.89 10	98.72 3	98.17 32	94.71 69
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
PS-CasMVS		97.22 5	97.84 6	96.50 5	97.08 23	97.92 6	98.17 29	97.02 2	94.71 26	95.32 24	98.52 14	98.97 11	92.91 43	99.04 4	98.47 5	98.49 18	97.24 11

v2v482		93.42 97	93.49 114	93.32 89	93.44 143	94.05 123	96.36 95	89.76 117	91.41 100	95.24 25	97.63 51	98.34 38	90.44 106	91.65 170	91.76 150	94.69 157	89.62 158

CP-MVSNet		96.97 10	97.42 12	96.44 7	97.06 24	97.82 8	98.12 31	96.98 3	93.50 45	95.21 26	97.98 30	98.44 31	92.83 46	98.93 8	98.37 8	98.46 21	96.91 27

v13		94.54 62	94.93 65	94.09 59	93.81 125	95.44 73	96.99 56	91.67 74	92.43 67	95.20 27	98.33 18	98.73 20	91.87 59	93.67 130	92.26 136	95.00 143	93.63 92

DTE-MVSNet		97.16 6	97.75 8	96.47 6	97.40 12	97.95 5	98.20 28	96.89 4	95.30 19	95.15 28	98.66 10	98.80 17	92.77 47	98.97 7	98.27 9	98.44 22	96.28 40

v12		94.44 63	94.79 71	94.02 63	93.75 129	95.37 79	96.92 57	91.61 75	92.21 73	95.10 29	98.27 22	98.69 22	91.73 63	93.49 132	92.15 141	94.97 147	93.37 97

v1		93.48 94	93.57 109	93.37 85	93.48 140	94.18 120	96.41 88	89.61 123	91.46 95	95.03 30	97.82 38	98.43 32	90.95 91	92.00 165	91.37 165	94.75 154	89.70 155

v7		93.65 85	93.73 102	93.57 82	93.38 144	94.60 108	96.83 63	89.92 114	89.69 134	95.02 31	97.89 34	98.24 42	91.27 74	92.38 152	92.18 139	94.99 144	91.12 142

v10		93.96 72	94.12 91	93.77 78	93.37 145	95.45 72	96.83 63	91.13 88	89.70 133	95.02 31	97.88 36	98.23 43	91.27 74	92.39 151	92.18 139	94.99 144	93.00 107

divwei89l23v2f112		93.47 95	93.56 110	93.37 85	93.48 140	94.17 121	96.42 86	89.62 121	91.46 95	95.00 33	97.81 39	98.42 33	90.94 92	92.00 165	91.38 163	94.75 154	89.70 155

V9		94.33 65	94.66 74	93.94 70	93.69 133	95.31 80	96.84 62	91.53 78	92.04 79	95.00 33	98.22 23	98.64 23	91.62 65	93.29 135	92.05 143	94.93 148	93.10 103

v1141		93.47 95	93.56 110	93.36 87	93.48 140	94.17 121	96.42 86	89.62 121	91.44 98	94.99 35	97.81 39	98.42 33	90.94 92	92.00 165	91.38 163	94.74 156	89.69 157

PEN-MVS		97.16 6	97.87 5	96.33 12	97.20 20	97.97 4	98.25 25	96.86 5	95.09 24	94.93 36	98.66 10	99.16 7	92.27 54	98.98 6	98.39 7	98.49 18	96.83 30

DeepC-MVS	92.47 4	96.44 15	96.75 22	96.08 17	97.57 7	97.19 29	97.96 35	94.28 23	95.29 20	94.92 37	98.31 21	96.92 89	93.69 29	96.81 61	96.50 47	98.06 37	96.27 41
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
V14		94.21 69	94.52 79	93.85 73	93.62 135	95.25 84	96.76 66	91.42 79	91.83 83	94.91 38	98.15 24	98.57 27	91.49 69	93.06 140	91.93 147	94.90 149	92.82 110

MTAPA										94.88 39		96.88 91

EG-PatchMatch MVS		94.81 55	95.53 53	93.97 67	95.89 63	94.62 106	95.55 116	88.18 157	92.77 61	94.88 39	97.04 66	98.61 25	93.31 33	96.89 57	95.19 80	95.99 114	93.56 94

v15		94.09 70	94.37 84	93.77 78	93.56 137	95.18 85	96.68 72	91.34 83	91.64 88	94.83 41	98.09 27	98.51 30	91.37 72	92.84 141	91.80 149	94.85 150	92.53 122

zzz-MVS		96.18 22	96.01 43	96.38 8	98.30 2	96.18 51	98.51 14	94.48 21	94.56 28	94.81 42	91.73 149	96.96 87	94.30 22	98.09 20	97.83 15	97.91 43	96.73 32

LS3D		95.83 39	96.35 33	95.22 37	96.47 45	97.49 15	97.99 32	92.35 54	94.92 25	94.58 43	94.88 110	95.11 140	91.52 68	98.48 13	98.05 12	98.42 24	95.49 55

Effi-MVS+		92.93 113	92.16 138	93.83 74	94.29 98	93.53 142	95.04 122	92.98 46	85.27 180	94.46 44	90.24 166	95.34 133	89.99 109	93.72 127	94.23 107	96.22 105	92.79 112

WR-MVS		97.53 3	98.20 3	96.76 3	96.93 27	98.17 1	98.60 11	96.67 6	96.39 12	94.46 44	99.14 1	98.92 12	94.57 18	99.06 3	98.80 2	99.32 1	96.92 26

MVS_0304		93.92 74	93.81 100	94.05 62	96.06 56	96.00 56	96.43 83	92.76 47	85.99 175	94.43 46	94.04 124	97.08 83	88.12 127	94.65 107	94.20 108	96.47 84	94.71 69

MP-MVS		96.13 24	95.93 46	96.37 9	98.19 4	97.31 24	98.49 15	94.53 20	91.39 101	94.38 47	94.32 119	96.43 105	94.59 17	97.75 37	97.44 26	98.04 38	96.88 28
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
HPM-MVS++		95.21 50	94.89 67	95.59 24	97.79 6	95.39 78	97.68 44	94.05 29	91.91 82	94.35 48	93.38 131	95.07 141	92.94 42	96.01 76	95.88 64	96.73 76	96.61 36

3Dnovator+	92.82 3	95.22 49	95.16 61	95.29 35	96.17 52	96.55 39	97.64 45	94.02 30	94.16 38	94.29 49	92.09 145	93.71 157	91.90 57	96.68 64	96.51 46	97.70 50	96.40 38

ACMMPR		96.54 13	96.71 24	96.35 11	97.55 9	97.63 11	98.62 10	94.54 17	94.45 30	94.19 50	95.04 106	97.35 77	94.92 13	97.85 31	97.50 23	98.26 28	97.17 14

OMC-MVS		94.74 56	95.46 56	93.91 72	94.62 89	96.26 49	96.64 74	89.36 134	94.20 36	94.15 51	94.02 125	97.73 69	91.34 73	96.15 74	95.04 84	97.37 57	94.80 66

CP-MVS		96.21 21	96.16 41	96.27 13	97.56 8	97.13 32	98.43 16	94.70 16	92.62 63	94.13 52	92.71 139	98.03 57	94.54 19	98.00 26	97.60 20	98.23 29	97.05 20

DeepPCF-MVS	90.68 7	94.56 60	94.92 66	94.15 58	94.11 106	95.71 65	97.03 52	90.65 99	93.39 50	94.08 53	95.29 98	94.15 151	93.21 37	95.22 94	94.92 88	95.82 121	95.75 52

v8		93.60 87	93.82 98	93.34 88	93.13 152	95.06 89	96.39 89	90.75 97	89.90 129	94.03 54	97.70 47	98.21 46	91.08 86	92.36 153	91.47 161	94.63 160	92.07 130

ACMMP_NAP		95.86 37	96.18 38	95.47 29	97.11 22	97.26 26	98.37 21	93.48 42	93.49 46	93.99 55	95.61 87	94.11 152	92.49 48	97.87 30	97.44 26	97.40 55	97.52 8

v1neww		93.27 101	93.40 119	93.12 100	93.13 152	94.20 117	96.39 89	89.56 124	89.87 131	93.95 56	97.71 45	98.21 46	91.09 84	92.36 153	91.49 157	94.62 162	89.96 151

v7new		93.27 101	93.40 119	93.12 100	93.13 152	94.20 117	96.39 89	89.56 124	89.87 131	93.95 56	97.71 45	98.21 46	91.09 84	92.36 153	91.49 157	94.62 162	89.96 151

v6		93.27 101	93.41 117	93.12 100	93.13 152	94.20 117	96.39 89	89.55 126	89.89 130	93.93 58	97.72 43	98.22 45	91.10 83	92.36 153	91.49 157	94.63 160	89.95 153

TAPA-MVS	88.94 13	93.78 84	94.31 87	93.18 99	94.14 103	95.99 57	95.74 106	86.98 181	93.43 48	93.88 59	90.16 167	96.88 91	91.05 87	94.33 112	93.95 109	97.28 60	95.40 56
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
ACMM	90.06 9	96.31 18	96.42 31	96.19 15	97.21 19	97.16 31	98.71 5	93.79 35	94.35 34	93.81 60	92.80 138	98.23 43	95.11 9	98.07 22	97.45 25	98.51 17	96.86 29
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
SMA-MVS		95.99 30	96.48 29	95.41 30	97.43 11	97.36 21	97.55 48	93.70 38	94.05 39	93.79 61	97.02 67	94.53 147	92.28 53	97.53 41	97.19 33	97.73 47	97.67 7

UniMVSNet_NR-MVSNet		95.34 47	95.51 54	95.14 39	95.80 66	96.55 39	96.61 75	94.79 15	90.04 126	93.78 62	97.51 54	97.25 79	91.19 78	96.68 64	96.31 54	98.65 11	94.22 78

DU-MVS		95.51 42	95.68 50	95.33 32	96.45 46	96.44 46	96.61 75	95.32 10	89.97 127	93.78 62	97.46 56	98.07 53	91.19 78	97.03 51	96.53 45	98.61 13	94.22 78

DeepC-MVS_fast	91.38 6	94.73 57	94.98 64	94.44 52	96.83 34	96.12 53	96.69 70	92.17 60	92.98 56	93.72 64	94.14 121	95.45 129	90.49 105	95.73 82	95.30 76	96.71 77	95.13 63
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
PGM-MVS		95.90 35	95.72 49	96.10 16	97.53 10	97.45 20	98.55 13	94.12 28	90.25 121	93.71 65	93.20 133	97.18 81	94.63 16	97.68 38	97.34 32	98.08 35	96.97 22

WR-MVS_H		97.06 9	97.78 7	96.23 14	96.74 36	98.04 3	98.25 25	97.32 1	94.40 33	93.71 65	98.55 13	98.89 13	92.97 40	98.91 9	98.45 6	98.38 27	97.19 13

v17		93.60 87	93.85 96	93.30 92	93.15 151	94.99 94	96.46 81	90.81 95	89.58 136	93.61 67	97.66 49	98.15 50	91.19 78	92.60 148	91.61 153	94.61 164	92.37 124

pmmvs-eth3d		92.34 128	92.33 134	92.34 115	92.67 166	90.67 183	96.37 93	89.06 138	90.98 112	93.60 68	97.13 64	97.02 85	88.29 123	90.20 183	91.42 162	94.07 174	88.89 166

CSCG		96.07 27	97.15 18	94.81 48	96.06 56	97.58 13	96.52 78	90.98 92	96.51 9	93.60 68	97.13 64	98.55 29	93.01 38	97.17 49	95.36 75	98.68 9	97.78 4

OPM-MVS		95.96 32	96.59 26	95.23 36	96.67 39	96.52 43	97.86 38	93.28 43	95.27 22	93.46 70	96.26 78	98.85 16	92.89 44	97.09 50	96.37 50	97.22 64	95.78 51

CANet		93.07 110	93.05 126	93.10 103	95.90 61	95.41 76	95.88 102	91.94 68	84.77 183	93.36 71	94.05 123	95.25 138	86.25 139	94.33 112	93.94 110	95.30 129	93.58 93

MAR-MVS		91.86 136	91.14 149	92.71 109	94.29 98	94.24 116	94.91 124	91.82 71	81.66 199	93.32 72	84.51 212	93.42 160	86.86 135	95.16 96	94.44 105	95.05 142	94.53 75
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
XVS					96.86 30	97.48 16	98.73 3			93.28 73		96.82 94				98.17 32

X-MVStestdata					96.86 30	97.48 16	98.73 3			93.28 73		96.82 94				98.17 32

X-MVS		95.33 48	95.13 63	95.57 26	97.35 14	97.48 16	98.43 16	94.28 23	92.30 72	93.28 73	86.89 198	96.82 94	91.87 59	97.85 31	97.59 21	98.19 30	96.95 25

V42		92.67 120	93.50 113	91.71 124	91.41 190	92.96 149	95.71 108	85.00 199	89.67 135	93.22 76	97.67 48	98.01 58	91.02 89	92.65 145	92.12 142	93.86 176	91.42 141

MDTV_nov1_ep13_2view		88.22 176	87.85 184	88.65 170	91.40 191	86.75 202	94.07 144	84.97 200	88.86 145	93.20 77	96.11 82	96.21 113	83.70 154	87.29 210	80.29 219	84.56 218	79.46 215

PVSNet_Blended_VisFu		93.60 87	93.41 117	93.83 74	96.31 49	95.65 67	95.71 108	90.58 102	88.08 154	93.17 78	95.29 98	92.20 166	90.72 96	94.69 106	93.41 123	96.51 83	94.54 74

gm-plane-assit		86.15 192	82.51 208	90.40 141	95.81 65	92.29 163	97.99 32	84.66 203	92.15 76	93.15 79	97.84 37	44.65 247	78.60 194	88.02 206	85.95 202	92.20 192	76.69 222

ACMMP		96.12 25	96.27 37	95.93 19	97.20 20	97.60 12	98.64 8	93.74 36	92.47 65	93.13 80	93.23 132	98.06 54	94.51 20	97.99 27	97.57 22	98.39 26	96.99 21
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
v16		93.53 92	93.80 101	93.20 97	93.10 157	94.98 95	96.43 83	90.81 95	89.39 139	93.12 81	97.63 51	98.01 58	91.19 78	92.60 148	91.65 152	94.58 166	92.36 125

v148		92.38 127	92.78 130	91.91 121	92.86 163	92.13 168	94.84 125	87.03 180	91.47 94	93.07 82	96.92 68	98.89 13	90.10 108	92.05 162	89.69 179	93.56 179	88.27 176

MVS_111021_LR		93.15 107	93.65 105	92.56 111	93.89 121	92.28 164	95.09 121	86.92 183	91.26 105	92.99 83	94.46 117	96.22 111	90.64 99	95.11 97	93.45 121	95.85 119	92.74 115

v18		93.33 100	93.59 108	93.04 106	92.94 160	94.87 99	96.31 96	90.59 101	88.96 141	92.89 84	97.51 54	97.90 62	91.01 90	92.33 157	91.48 160	94.50 167	92.05 131

TSAR-MVS + MP.		95.99 30	96.57 27	95.31 33	96.87 28	96.50 44	98.71 5	91.58 76	93.25 51	92.71 85	96.86 69	96.57 101	93.92 24	98.09 20	97.91 13	98.08 35	96.81 31

Fast-Effi-MVS+		92.93 113	92.64 132	93.27 93	93.81 125	93.88 130	95.90 101	90.61 100	83.98 188	92.71 85	92.81 137	96.22 111	90.67 97	94.90 103	93.92 111	95.92 116	92.77 113

SD-MVS		95.77 40	96.17 39	95.30 34	96.72 37	96.19 50	97.01 53	93.04 45	94.03 40	92.71 85	96.45 76	96.78 98	93.91 25	96.79 62	95.89 63	98.42 24	97.09 18

casdiffmvs1		93.02 112	93.07 125	92.96 108	94.93 83	95.42 74	96.24 97	90.96 93	91.68 87	92.69 88	93.74 128	96.88 91	87.86 128	90.19 184	89.56 182	95.09 140	94.03 83

LGP-MVS_train		96.10 26	96.29 35	95.87 20	96.72 37	97.35 23	98.43 16	93.83 34	90.81 118	92.67 89	95.05 104	98.86 15	95.01 10	98.11 19	97.37 31	98.52 16	96.50 37

UA-Net		96.56 12	96.73 23	96.36 10	98.99 1	97.90 7	97.79 42	95.64 9	92.78 60	92.54 90	96.23 81	95.02 142	94.31 21	98.43 14	98.12 11	98.89 3	98.58 2

HFP-MVS		96.18 22	96.53 28	95.77 21	97.34 16	97.26 26	98.16 30	94.54 17	94.45 30	92.52 91	95.05 104	96.95 88	93.89 26	97.28 45	97.46 24	98.19 30	97.25 9

MVS_111021_HR		93.82 82	94.26 90	93.31 90	95.01 80	93.97 128	95.73 107	89.75 118	92.06 78	92.49 92	94.01 126	96.05 117	90.61 103	95.95 77	94.78 93	96.28 99	93.04 106

thisisatest0515		93.79 83	94.41 83	93.06 105	94.14 103	92.50 158	95.56 115	88.55 151	91.61 90	92.45 93	96.84 70	95.71 122	90.62 101	94.58 108	95.07 82	97.05 70	94.58 73

CNLPA		93.14 108	93.67 104	92.53 112	94.62 89	94.73 103	95.00 123	86.57 187	92.85 59	92.43 94	90.94 154	94.67 145	90.35 107	95.41 85	93.70 115	96.23 104	93.37 97

PM-MVS		92.65 121	93.20 124	92.00 120	92.11 182	90.16 187	95.99 100	84.81 202	91.31 103	92.41 95	95.87 83	96.64 100	92.35 51	93.65 131	92.91 128	94.34 171	91.85 135

3Dnovator	91.81 5	93.36 99	94.27 89	92.29 116	92.99 159	95.03 90	95.76 105	87.79 166	93.82 43	92.38 96	92.19 144	93.37 161	88.14 126	95.26 93	94.85 89	96.69 78	95.40 56

CPTT-MVS		95.00 53	94.52 79	95.57 26	96.84 32	96.78 36	97.88 37	93.67 40	92.20 74	92.35 97	85.87 205	97.56 73	94.98 12	96.96 54	96.07 59	97.70 50	96.18 42

ACMP	89.62 11	95.96 32	96.28 36	95.59 24	96.58 42	97.23 28	98.26 24	93.22 44	92.33 71	92.31 98	94.29 120	98.73 20	94.68 15	98.04 23	97.14 37	98.47 20	96.17 43
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
PCF-MVS	87.46 14	92.44 125	91.80 141	93.19 98	94.66 87	95.80 63	96.37 93	90.19 104	87.57 159	92.23 99	89.26 175	93.97 153	89.24 113	91.32 174	90.82 170	96.46 86	93.86 87
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
PLC	87.27 15	93.08 109	92.92 127	93.26 94	94.67 86	95.03 90	94.38 133	90.10 105	91.69 85	92.14 100	87.24 194	93.91 154	91.61 66	95.05 98	94.73 99	96.67 79	92.80 111
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
UniMVSNet (Re)		95.46 43	95.86 47	95.00 45	96.09 53	96.60 38	96.68 72	94.99 12	90.36 120	92.13 101	97.64 50	98.13 51	91.38 71	96.90 56	96.74 42	98.73 6	94.63 72

NCCC		93.87 80	93.42 116	94.40 54	96.84 32	95.42 74	96.47 80	92.62 48	92.36 70	92.05 102	83.83 214	95.55 124	91.84 61	95.89 78	95.23 79	96.56 81	95.63 53

tpm		81.58 212	78.84 221	84.79 207	91.11 198	79.50 225	89.79 212	83.75 206	79.30 218	92.05 102	90.98 153	64.78 235	74.54 215	80.50 231	76.67 228	77.49 235	80.15 211

CLD-MVS		92.81 117	94.32 86	91.05 132	95.39 73	95.31 80	95.82 104	81.44 220	89.40 137	91.94 104	95.86 84	97.36 76	85.83 141	95.35 87	94.59 102	95.85 119	92.34 127
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
TranMVSNet+NR-MVSNet		95.72 41	96.42 31	94.91 47	96.21 51	96.77 37	96.90 60	94.99 12	92.62 63	91.92 105	98.51 15	98.63 24	90.82 94	97.27 46	96.83 41	98.63 12	94.31 77

SteuartSystems-ACMMP		95.96 32	96.13 42	95.76 22	97.06 24	97.36 21	98.40 20	94.24 25	91.49 92	91.91 106	94.50 115	96.89 90	94.99 11	98.01 25	97.44 26	97.97 41	97.25 9
Skip Steuart: Steuart Systems R&D Blog.
EPNet		90.17 149	89.07 167	91.45 128	97.25 18	90.62 185	94.84 125	93.54 41	80.96 201	91.85 107	86.98 197	85.88 190	77.79 202	92.30 158	92.58 132	93.41 181	94.20 80
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
MSDG		92.09 134	92.84 129	91.22 131	92.55 167	92.97 148	93.42 156	85.43 197	90.24 122	91.83 108	94.70 111	94.59 146	88.48 121	94.91 102	93.31 125	95.59 125	89.15 162

EU-MVSNet		91.63 138	92.73 131	90.35 142	88.36 219	87.89 197	96.53 77	81.51 219	92.45 66	91.82 109	96.44 77	97.05 84	93.26 35	94.10 123	88.94 190	90.61 197	92.24 128

NR-MVSNet		94.55 61	95.66 52	93.25 96	94.26 100	96.44 46	96.69 70	95.32 10	89.97 127	91.79 110	97.46 56	98.39 36	82.85 160	96.87 59	96.48 48	98.57 14	93.98 85

MCST-MVS		93.60 87	93.40 119	93.83 74	95.30 74	95.40 77	96.49 79	90.87 94	90.08 124	91.72 111	90.28 165	95.99 118	91.69 64	93.94 125	92.99 126	96.93 75	95.13 63

TSAR-MVS + COLMAP		93.06 111	93.65 105	92.36 113	94.62 89	94.28 115	95.36 119	89.46 131	92.18 75	91.64 112	95.55 91	95.27 137	88.60 120	93.24 136	92.50 133	94.46 168	92.55 121

HQP-MVS		92.87 115	92.49 133	93.31 90	95.75 67	95.01 93	95.64 111	91.06 90	88.54 148	91.62 113	88.16 184	96.25 109	89.47 112	92.26 159	91.81 148	96.34 94	95.40 56

ESAPD		96.00 29	96.80 21	95.06 43	95.87 64	97.47 19	98.25 25	93.73 37	92.38 68	91.57 114	97.55 53	97.97 60	92.98 39	97.49 42	97.61 19	97.96 42	97.16 15

HSP-MVS		95.04 52	95.45 57	94.57 51	96.87 28	97.77 10	98.71 5	93.88 33	91.21 106	91.48 115	95.36 95	98.37 37	90.73 95	94.37 111	92.98 127	95.77 122	98.08 3

casdiffmvs		91.65 137	91.03 150	92.36 113	93.69 133	94.95 98	95.60 113	91.36 81	85.32 178	91.43 116	91.77 146	95.47 128	87.29 131	88.58 200	88.39 193	94.81 152	92.75 114

MSLP-MVS++		93.91 75	94.30 88	93.45 84	95.51 72	95.83 62	93.12 167	91.93 69	91.45 97	91.40 117	87.42 193	96.12 115	93.27 34	96.57 68	96.40 49	95.49 126	96.29 39

RPSCF		95.46 43	96.95 20	93.73 80	95.72 68	95.94 58	95.58 114	88.08 161	95.31 18	91.34 118	96.26 78	98.04 56	93.63 30	98.28 16	97.67 17	98.01 39	97.13 16

EPNet_dtu		87.40 188	86.27 194	88.72 169	95.68 69	83.37 215	92.09 184	90.08 106	78.11 227	91.29 119	86.33 201	89.74 178	75.39 214	89.07 194	87.89 195	87.81 207	89.38 160
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
CNVR-MVS		94.24 68	94.47 81	93.96 69	96.56 43	95.67 66	96.43 83	91.95 67	92.08 77	91.28 120	90.51 161	95.35 132	91.20 77	96.34 73	95.50 73	96.34 94	95.88 48

AdaColmap		92.41 126	91.49 146	93.48 83	95.96 59	95.02 92	95.37 118	91.73 72	87.97 157	91.28 120	82.82 219	91.04 172	90.62 101	95.82 80	95.07 82	95.95 115	92.67 116

CDPH-MVS		93.96 72	93.86 95	94.08 61	96.31 49	95.84 61	96.92 57	91.85 70	87.21 165	91.25 122	92.83 136	96.06 116	91.05 87	95.57 83	94.81 90	97.12 65	94.72 68

QAPM		92.57 122	93.51 112	91.47 127	92.91 162	94.82 100	93.01 169	87.51 172	91.49 92	91.21 123	92.24 142	91.70 168	88.74 118	94.54 109	94.39 106	95.41 127	95.37 59

MDA-MVSNet-bldmvs		89.75 153	91.67 142	87.50 187	74.25 244	90.88 181	94.68 130	85.89 192	91.64 88	91.03 124	95.86 84	94.35 150	89.10 115	96.87 59	86.37 201	90.04 198	85.72 190

pmmvs4		89.95 152	89.32 165	90.69 137	91.60 189	89.17 191	94.37 134	87.63 169	88.07 155	91.02 125	94.50 115	90.50 176	86.13 140	86.33 213	89.40 183	93.39 182	87.29 181

TSAR-MVS + ACMM		95.17 51	95.95 44	94.26 56	96.07 55	96.46 45	95.67 110	94.21 26	93.84 42	90.99 126	97.18 62	95.24 139	93.55 31	96.60 67	95.61 71	95.06 141	96.69 34

CR-MVSNet		85.32 196	81.58 213	89.69 155	90.36 205	84.79 209	86.72 228	92.22 56	75.38 232	90.73 127	90.41 163	67.88 230	84.86 145	83.76 222	85.74 203	93.24 184	83.14 198

Patchmtry						83.74 214	86.72 228	92.22 56		90.73 127

PatchT		83.44 203	81.10 215	86.18 198	77.92 241	82.58 219	89.87 211	87.39 175	75.88 231	90.73 127	89.86 169	66.71 233	84.86 145	83.76 222	85.74 203	86.33 214	83.14 198

OpenMVS	89.22 12	91.09 142	91.42 147	90.71 136	92.79 165	93.61 139	92.74 177	85.47 196	86.10 174	90.73 127	85.71 206	93.07 164	86.69 136	94.07 124	93.34 124	95.86 117	94.02 84

CANet_DTU		88.95 166	89.51 162	88.29 180	93.12 156	91.22 178	93.61 152	83.47 211	80.07 213	90.71 131	89.19 176	93.68 158	76.27 213	91.44 173	91.17 167	92.59 190	89.83 154

IterMVS-LS		92.10 133	92.33 134	91.82 123	93.18 148	93.66 134	92.80 176	92.27 55	90.82 116	90.59 132	97.19 61	90.97 173	87.76 129	89.60 190	90.94 169	94.34 171	93.16 102
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
train_agg		93.89 77	93.46 115	94.40 54	97.35 14	93.78 132	97.63 46	92.19 59	88.12 151	90.52 133	93.57 130	95.78 121	92.31 52	94.78 104	93.46 120	96.36 89	94.70 71

ACMH+	89.90 10	96.27 19	97.52 11	94.81 48	95.19 77	97.18 30	97.97 34	92.52 49	96.72 8	90.50 134	97.31 59	99.11 8	94.10 23	98.67 12	97.90 14	98.56 15	95.79 50

APD-MVS		95.38 46	95.68 50	95.03 44	97.30 17	96.90 35	97.83 39	93.92 31	89.40 137	90.35 135	95.41 94	97.69 70	92.97 40	97.24 48	97.17 35	97.83 45	95.96 46
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
Effi-MVS+-dtu		92.32 130	91.66 143	93.09 104	95.13 79	94.73 103	94.57 132	92.14 61	81.74 198	90.33 136	88.13 185	95.91 119	89.24 113	94.23 121	93.65 119	97.12 65	93.23 100

PVSNet_BlendedMVS		90.09 150	90.12 156	90.05 147	92.40 171	92.74 153	91.74 189	85.89 192	80.54 208	90.30 137	88.54 180	95.51 125	84.69 147	92.64 146	90.25 174	95.28 131	90.61 146

PVSNet_Blended		90.09 150	90.12 156	90.05 147	92.40 171	92.74 153	91.74 189	85.89 192	80.54 208	90.30 137	88.54 180	95.51 125	84.69 147	92.64 146	90.25 174	95.28 131	90.61 146

Baseline_NR-MVSNet		94.85 54	95.35 59	94.26 56	96.45 46	93.86 131	96.70 68	94.54 17	90.07 125	90.17 139	98.77 6	97.89 63	90.64 99	97.03 51	96.16 55	97.04 72	93.67 90

v1.0		88.18 179	82.50 209	94.80 50	96.76 35	97.29 25	97.74 43	94.15 27	91.69 85	90.01 140	96.65 73	97.29 78	92.45 49	97.41 43	97.18 34	97.67 52	0.00 246

abl_6				91.88 122	93.76 128	94.98 95	95.64 111	88.97 141	86.20 173	90.00 141	86.31 202	94.50 148	87.31 130			95.60 124	92.48 123

GA-MVS		88.76 167	88.04 181	89.59 157	92.32 174	91.46 175	92.28 183	86.62 185	83.82 190	89.84 142	92.51 141	81.94 201	83.53 156	89.41 192	89.27 185	92.95 188	87.90 177

APDe-MVS		96.23 20	97.22 16	95.08 42	96.66 40	97.56 14	98.63 9	93.69 39	94.62 27	89.80 143	97.73 42	98.13 51	93.84 27	97.79 35	97.63 18	97.87 44	97.08 19

no-one		92.05 135	94.57 78	89.12 163	85.55 230	87.65 200	94.21 141	77.34 226	93.43 48	89.64 144	95.11 103	99.11 8	95.86 4	95.38 86	95.24 78	92.08 194	96.11 44

tttt0517		89.64 154	88.05 180	91.49 126	93.52 138	91.65 173	93.67 150	87.53 170	82.77 195	89.39 145	90.37 164	70.05 226	88.21 124	93.71 129	93.79 113	96.63 80	94.04 82

Vis-MVSNet		94.39 64	95.85 48	92.68 110	90.91 200	95.88 60	97.62 47	91.41 80	91.95 80	89.20 146	97.29 60	96.26 108	90.60 104	96.95 55	95.91 61	96.32 97	96.71 33
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
thisisatest0530		89.54 157	87.99 182	91.35 130	93.17 149	91.31 176	93.45 155	87.53 170	82.96 193	89.17 147	90.45 162	70.32 225	88.21 124	93.37 134	93.79 113	96.54 82	93.71 89

TSAR-MVS + GP.		94.25 67	94.81 70	93.60 81	96.52 44	95.80 63	94.37 134	92.47 52	90.89 114	88.92 148	95.34 96	94.38 149	92.85 45	96.36 72	95.62 70	96.47 84	95.28 60

PatchmatchNet		82.44 206	78.69 223	86.83 193	89.81 208	81.55 222	90.78 202	87.27 178	82.39 197	88.85 149	88.31 183	70.96 224	81.90 170	78.58 235	74.33 235	82.35 230	74.69 226
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
MDTV_nov1_ep13		82.33 207	79.66 218	85.45 201	88.83 215	83.88 213	90.09 210	81.98 216	79.07 221	88.82 150	88.70 178	73.77 220	78.41 199	80.29 232	76.08 229	84.56 218	75.83 223

CMPMVS	66.55 18	85.55 195	87.46 188	83.32 211	84.99 231	81.97 220	79.19 242	75.93 228	79.32 217	88.82 150	85.09 207	91.07 171	82.12 168	92.56 150	89.63 181	88.84 203	92.56 120
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
DI_MVS_plusplus_trai		90.68 144	90.40 154	91.00 133	92.43 170	92.61 156	94.17 143	88.98 140	88.32 150	88.76 152	93.67 129	87.58 185	86.44 138	89.74 188	90.33 172	95.24 133	90.56 148

IterMVS		88.32 172	88.25 178	88.41 175	90.83 202	91.24 177	93.07 168	81.69 217	86.77 167	88.55 153	95.61 87	86.91 189	87.01 132	87.38 208	83.77 208	89.29 200	86.06 189
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
canonicalmvs		93.38 98	94.36 85	92.24 117	93.94 117	96.41 48	94.18 142	90.47 103	93.07 55	88.47 154	88.66 179	93.78 156	88.80 117	95.74 81	95.75 67	97.57 53	97.13 16

FPMVS		90.81 143	91.60 144	89.88 150	92.52 168	88.18 193	93.31 163	83.62 208	91.59 91	88.45 155	88.96 177	89.73 179	86.96 133	96.42 71	95.69 69	94.43 169	90.65 145

DELS-MVS		92.33 129	93.61 107	90.83 135	92.84 164	95.13 88	94.76 128	87.22 179	87.78 158	88.42 156	95.78 86	95.28 136	85.71 142	94.44 110	93.91 112	96.01 113	92.97 108
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
PatchMatch-RL		89.59 155	88.80 171	90.51 139	92.20 180	88.00 196	91.72 191	86.64 184	84.75 184	88.25 157	87.10 196	90.66 175	89.85 111	93.23 137	92.28 135	94.41 170	85.60 192

pmmvs5		88.63 169	89.70 160	87.39 189	89.24 212	90.64 184	91.87 186	82.13 215	83.34 191	87.86 158	94.58 113	96.15 114	79.87 184	87.33 209	89.07 189	93.39 182	86.76 185

Fast-Effi-MVS+-dtu		89.57 156	88.42 177	90.92 134	93.35 146	91.57 174	93.01 169	95.71 8	78.94 222	87.65 159	84.68 211	93.14 163	82.00 169	90.84 178	91.01 168	93.78 178	88.77 167

MVS_Test		90.19 148	90.58 151	89.74 153	92.12 181	91.74 172	92.51 179	88.54 152	82.80 194	87.50 160	94.62 112	95.02 142	83.97 151	88.69 197	89.32 184	93.79 177	91.85 135

gg-mvs-nofinetune		88.32 172	88.81 170	87.75 186	93.07 158	89.37 190	89.06 217	95.94 7	95.29 20	87.15 161	97.38 58	76.38 217	68.05 226	91.04 176	89.10 188	93.24 184	83.10 200

tpmp4_e23		82.16 208	78.26 226	86.70 195	89.92 207	84.82 208	91.17 197	89.95 113	81.21 200	87.10 162	81.91 221	64.01 236	77.88 201	79.89 233	74.99 233	84.18 222	81.00 207

RPMNet		83.42 204	78.40 224	89.28 162	89.79 209	84.79 209	90.64 205	92.11 63	75.38 232	87.10 162	79.80 226	61.99 239	82.79 162	81.88 228	82.07 213	93.23 186	82.87 201

Anonymous20240521		93.49 93	95.15 62	91.55 125	94.05 107	95.92 59	95.15 120	91.21 85	92.76 62	87.01 164	89.71 170	97.16 82	83.90 153	97.65 39	96.87 40	97.99 40	95.95 47

CVMVSNet		88.97 165	89.73 159	88.10 182	87.33 226	85.22 206	94.68 130	78.68 222	88.94 143	86.98 165	95.55 91	85.71 191	89.87 110	91.19 175	89.69 179	91.05 195	91.78 138

MVS-HIRNet		78.28 227	75.28 238	81.79 215	80.33 237	69.38 244	76.83 243	86.59 186	70.76 239	86.66 166	89.57 172	81.04 205	77.74 203	77.81 237	71.65 236	82.62 227	66.73 238

diffmvs		90.35 146	91.54 145	88.96 165	90.90 201	92.20 166	91.93 185	88.45 156	86.34 172	86.36 167	93.13 135	96.99 86	83.54 155	90.32 182	88.69 192	92.59 190	93.09 104

tpm cat1		80.03 218	75.93 237	84.81 206	89.31 211	83.26 217	88.86 218	86.55 188	79.24 219	86.10 168	84.22 213	63.62 237	77.37 207	73.43 240	70.88 238	80.67 231	76.87 221

UGNet		92.31 131	94.70 73	89.53 158	90.99 199	95.53 71	96.19 98	92.10 64	91.35 102	85.76 169	95.31 97	95.48 127	76.84 209	95.22 94	94.79 92	95.32 128	95.19 61
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
EPP-MVSNet		93.63 86	93.95 92	93.26 94	95.15 78	96.54 42	96.18 99	91.97 66	91.74 84	85.76 169	94.95 109	84.27 194	91.60 67	97.61 40	97.38 30	98.87 4	95.18 62

DWT-MVSNet_training		79.22 222	73.99 239	85.33 202	88.57 216	84.41 211	90.56 206	80.96 221	73.90 237	85.72 171	75.62 233	50.09 246	81.30 178	76.91 239	77.02 227	84.88 216	79.97 213

diffmvs1		91.31 140	92.90 128	89.45 161	91.20 197	92.51 157	92.91 171	88.54 152	91.21 106	85.63 172	95.02 107	97.50 74	85.37 143	90.92 177	89.89 177	93.08 187	93.27 99

pmmvs6		94.58 59	97.30 15	91.40 129	94.84 85	94.61 107	93.40 157	92.43 53	98.51 2	85.61 173	98.73 9	99.53 2	84.40 149	97.88 29	97.03 38	97.72 48	94.79 67

HyFIR lowres test		88.19 178	86.56 192	90.09 145	91.24 193	92.17 167	94.30 140	88.79 148	84.06 186	85.45 174	89.52 173	85.64 192	88.64 119	85.40 220	87.28 197	92.14 193	81.87 203

TransMVSNet (Re)		93.55 91	96.32 34	90.32 143	94.38 96	94.05 123	93.30 164	89.53 127	97.15 7	85.12 175	98.83 5	97.89 63	82.21 167	96.75 63	96.14 57	97.35 58	93.46 95

IB-MVS	86.01 17	88.24 175	87.63 185	88.94 166	92.03 185	91.77 171	92.40 182	85.58 195	78.24 224	84.85 176	71.99 240	93.45 159	83.96 152	93.48 133	92.33 134	94.84 151	92.15 129
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
LP		84.09 202	84.31 200	83.85 210	79.40 239	84.34 212	90.26 207	84.02 205	87.99 156	84.66 177	91.61 150	79.13 210	80.58 181	85.90 218	81.59 214	84.16 223	79.59 214

Anonymous20231211		93.19 105	95.50 55	90.49 140	93.77 127	95.29 83	94.36 138	90.04 110	91.44 98	84.59 178	96.72 72	97.65 71	82.45 166	97.25 47	96.32 53	97.74 46	93.79 88

dps		81.42 216	77.88 231	85.56 200	87.67 222	85.17 207	88.37 221	87.46 173	74.37 235	84.55 179	86.80 199	62.18 238	80.20 182	81.13 230	77.52 225	85.10 215	77.98 220

MIMVSNet1		92.52 123	94.88 68	89.77 152	96.09 53	91.99 170	96.92 57	89.68 120	95.92 16	84.55 179	96.64 74	98.21 46	78.44 198	96.08 75	95.10 81	92.91 189	90.22 149

pmmvs3		81.69 211	83.83 202	79.19 224	78.33 240	78.57 228	89.53 214	58.71 241	78.88 223	84.34 181	88.36 182	91.96 167	77.69 204	87.48 207	82.42 212	86.54 213	79.18 217

pm-mvs1		93.27 101	95.94 45	90.16 144	94.13 105	93.66 134	92.61 178	89.91 115	95.73 17	84.28 182	98.51 15	98.29 39	82.80 161	96.44 70	95.76 66	97.25 62	93.21 101

CostFormer		82.15 209	79.54 219	85.20 204	88.92 214	85.70 205	90.87 201	86.26 191	79.19 220	83.87 183	87.89 189	69.20 228	76.62 211	77.50 238	75.28 231	84.69 217	82.02 202

conf0.05thres1000		91.24 141	91.85 140	90.53 138	94.59 92	94.56 111	94.33 139	89.52 128	93.67 44	83.77 184	91.04 152	79.10 211	83.98 150	96.66 66	95.56 72	96.98 74	92.36 125

tpmrst		78.81 225	76.18 236	81.87 214	88.56 217	77.45 231	86.74 227	81.52 218	80.08 212	83.48 185	90.84 157	66.88 231	74.54 215	73.04 241	71.02 237	76.38 237	73.95 231

tfpnnormal		92.45 124	94.77 72	89.74 153	93.95 115	93.44 144	93.25 165	88.49 154	95.27 22	83.20 186	96.51 75	96.23 110	83.17 158	95.47 84	94.52 104	96.38 88	91.97 133

view800		89.42 158	89.11 166	89.78 151	94.00 108	93.71 133	93.96 145	88.47 155	88.10 152	82.91 187	82.61 220	79.85 209	83.10 159	94.92 101	95.38 74	96.26 103	89.19 161

view600		89.09 162	88.78 172	89.46 160	93.59 136	93.33 146	93.92 148	87.76 167	87.40 160	82.79 188	81.29 222	80.71 206	82.59 165	94.28 115	95.72 68	96.12 111	88.70 168

tfpn		87.65 185	85.66 197	89.96 149	94.36 97	93.94 129	93.85 149	89.02 139	88.71 147	82.78 189	83.79 215	53.79 242	83.43 157	95.35 87	94.54 103	96.35 93	89.51 159

MVSTER		84.79 198	83.79 203	85.96 199	89.14 213	89.80 188	89.39 215	82.99 213	74.16 236	82.78 189	85.97 204	66.81 232	76.84 209	90.77 179	88.83 191	94.66 158	90.19 150

IS_MVSNet		92.76 118	93.25 123	92.19 118	94.91 84	95.56 68	95.86 103	92.12 62	88.10 152	82.71 191	93.15 134	88.30 183	88.86 116	97.29 44	96.95 39	98.66 10	93.38 96

thres600view7		89.14 161	88.83 169	89.51 159	93.71 132	93.55 140	93.93 147	88.02 162	87.30 162	82.40 192	81.18 223	80.63 207	82.69 164	94.27 116	95.90 62	96.27 101	88.94 165

thres400		88.54 170	88.15 179	88.98 164	93.17 149	92.84 150	93.56 153	86.93 182	86.45 170	82.37 193	79.96 225	81.46 204	81.83 172	93.21 138	94.76 94	96.04 112	88.39 174

CHOSEN 1792x2688		86.64 189	86.62 191	86.65 197	90.33 206	87.86 198	93.19 166	83.30 212	83.95 189	82.32 194	87.93 187	89.34 180	86.92 134	85.64 219	84.95 205	83.85 224	86.68 186

PMMVS		81.93 210	83.48 206	80.12 218	72.35 245	75.05 238	88.54 219	64.01 237	77.02 229	82.22 195	87.51 192	91.12 170	79.70 185	86.59 211	86.64 200	93.88 175	80.41 208

FMVSNet1		92.86 116	95.26 60	90.06 146	92.40 171	95.16 86	94.37 134	92.22 56	93.18 54	82.16 196	96.76 71	97.48 75	81.85 171	95.32 89	94.98 85	97.34 59	93.93 86

FC-MVSNet-train		92.75 119	95.40 58	89.66 156	95.21 76	94.82 100	97.00 54	89.40 132	91.13 108	81.71 197	97.72 43	96.43 105	77.57 205	96.89 57	96.72 43	97.05 70	94.09 81

thres200		88.29 174	87.88 183	88.76 168	92.50 169	93.55 140	92.47 181	88.02 162	84.80 181	81.44 198	79.28 227	82.20 200	81.83 172	94.27 116	93.67 116	96.27 101	87.40 180

thres100view900		86.46 191	86.00 196	86.99 192	92.28 175	91.03 179	91.09 198	84.49 204	80.90 202	80.89 199	78.25 228	82.25 196	77.57 205	90.17 185	92.84 129	95.63 123	86.57 187

tfpn200view9		87.94 181	87.51 186	88.44 174	92.28 175	93.63 138	93.35 162	88.11 160	80.90 202	80.89 199	78.25 228	82.25 196	79.65 187	94.27 116	94.76 94	96.36 89	88.48 171

tfpn111		87.59 186	86.89 189	88.41 175	92.28 175	93.64 136	93.36 158	88.12 158	80.90 202	80.71 201	73.93 237	82.25 196	79.65 187	94.27 116	94.76 94	96.36 89	88.48 171

conf200view11		87.93 182	87.51 186	88.41 175	92.28 175	93.64 136	93.36 158	88.12 158	80.90 202	80.71 201	78.25 228	82.25 196	79.65 187	94.27 116	94.76 94	96.36 89	88.48 171

CDS-MVSNet		88.41 171	89.79 158	86.79 194	94.55 93	90.82 182	92.50 180	89.85 116	83.26 192	80.52 203	91.05 151	89.93 177	69.11 223	93.17 139	92.71 131	94.21 173	87.63 178
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
conf0.01		85.72 194	83.49 205	88.32 178	92.11 182	93.35 145	93.36 158	88.02 162	80.90 202	80.51 204	74.83 235	59.86 240	79.65 187	93.80 126	94.76 94	96.29 98	86.94 184

conf0.002		84.82 197	81.84 212	88.30 179	92.05 184	93.28 147	93.36 158	88.00 165	80.90 202	80.48 205	73.43 239	52.48 245	79.65 187	93.72 127	92.82 130	96.28 99	86.22 188

MS-PatchMatch		87.72 184	88.62 176	86.66 196	90.81 203	88.18 193	90.92 200	82.25 214	85.86 176	80.40 206	90.14 168	89.29 181	84.93 144	89.39 193	89.12 187	90.67 196	88.34 175

FMVSNet2		90.28 147	92.04 139	88.23 181	91.22 194	94.05 123	92.88 173	90.69 98	86.53 169	79.89 207	94.38 118	92.73 165	78.54 195	91.64 171	92.26 136	96.17 108	92.67 116

testpf		72.68 237	66.81 241	79.53 219	86.52 228	73.89 239	83.56 236	88.74 149	58.70 245	79.68 208	71.31 241	53.64 243	62.23 230	68.68 242	66.64 241	76.46 236	74.82 225

tfpn_n400		89.03 163	89.39 163	88.61 171	93.98 112	92.33 161	91.83 187	88.97 141	92.97 57	78.90 209	84.93 208	78.24 213	81.77 174	95.00 99	93.67 116	96.22 105	88.59 169

tfpnconf		89.03 163	89.39 163	88.61 171	93.98 112	92.33 161	91.83 187	88.97 141	92.97 57	78.90 209	84.93 208	78.24 213	81.77 174	95.00 99	93.67 116	96.22 105	88.59 169

tfpnview11		88.74 168	88.95 168	88.50 173	93.91 119	92.43 160	91.70 193	88.90 146	90.93 113	78.90 209	84.93 208	78.24 213	81.71 177	94.32 114	94.60 101	95.86 117	87.23 182

test-mter		78.71 226	78.35 225	79.12 225	84.03 233	76.58 232	88.51 220	59.06 240	71.06 238	78.87 212	83.73 216	71.83 221	76.44 212	83.41 225	80.61 217	87.79 208	81.24 204

GBi-Net		89.35 159	90.58 151	87.91 184	91.22 194	94.05 123	92.88 173	90.05 107	79.40 214	78.60 213	90.58 158	87.05 186	78.54 195	95.32 89	94.98 85	96.17 108	92.67 116

test1		89.35 159	90.58 151	87.91 184	91.22 194	94.05 123	92.88 173	90.05 107	79.40 214	78.60 213	90.58 158	87.05 186	78.54 195	95.32 89	94.98 85	96.17 108	92.67 116

FMVSNet3		87.90 183	88.63 175	87.04 191	89.78 210	93.46 143	91.62 194	90.05 107	79.40 214	78.60 213	90.58 158	87.05 186	77.07 208	88.03 205	89.86 178	95.12 137	92.04 132

test-LLR		80.62 217	77.20 234	84.62 209	93.99 110	75.11 236	87.04 224	87.32 176	70.11 240	78.59 216	83.17 217	71.60 222	73.88 218	82.32 226	79.20 222	86.91 211	78.87 218

TESTMET0.1,1		77.47 231	77.20 234	77.78 228	81.94 235	75.11 236	87.04 224	58.33 242	70.11 240	78.59 216	83.17 217	71.60 222	73.88 218	82.32 226	79.20 222	86.91 211	78.87 218

CHOSEN 280x420		79.24 221	78.26 226	80.38 217	79.60 238	68.80 245	89.32 216	75.38 229	77.25 228	78.02 218	75.57 234	76.17 218	81.19 179	88.61 198	81.39 215	78.79 233	80.03 212

thresconf0.02		84.34 201	82.02 211	87.06 190	92.23 179	90.93 180	91.05 199	86.43 189	88.83 146	77.65 219	73.93 237	55.81 241	79.68 186	90.62 180	90.28 173	95.30 129	83.73 196

ADS-MVSNet		79.11 223	79.38 220	78.80 226	81.90 236	75.59 234	84.36 235	83.69 207	87.31 161	76.76 220	87.58 191	76.90 216	68.55 225	78.70 234	75.56 230	77.53 234	74.07 230

Anonymous20231206		87.45 187	89.66 161	84.87 205	94.00 108	87.73 199	91.36 196	86.41 190	88.89 144	75.03 221	92.59 140	96.82 94	72.48 220	89.72 189	88.06 194	89.93 199	83.81 195

Vis-MVSNet (Re-imp)		90.68 144	92.18 136	88.92 167	94.63 88	92.75 152	92.91 171	91.20 86	89.21 140	75.01 222	93.96 127	89.07 182	82.72 163	95.88 79	95.30 76	97.08 69	89.08 164

EMVS		77.65 230	77.49 233	77.83 227	87.75 221	71.02 242	81.13 240	70.54 234	86.38 171	74.52 223	89.38 174	80.19 208	78.22 200	89.48 191	67.13 240	74.83 241	58.84 242

tfpn1000		88.13 180	88.68 174	87.49 188	93.94 117	92.64 155	91.50 195	88.70 150	90.12 123	74.35 224	86.74 200	75.27 219	80.14 183	94.16 122	94.66 100	96.33 96	87.16 183

EPMVS		79.26 220	78.20 228	80.49 216	87.04 227	78.86 227	86.08 233	83.51 209	82.63 196	73.94 225	89.59 171	68.67 229	72.03 221	78.17 236	75.08 232	80.37 232	74.37 228

testus		78.20 228	81.50 214	74.36 234	85.59 229	79.36 226	86.99 226	65.76 235	76.01 230	73.00 226	77.98 231	93.35 162	51.30 242	86.33 213	82.79 211	83.50 226	74.68 227

FC-MVSNet-test		91.49 139	94.43 82	88.07 183	94.97 81	90.53 186	95.42 117	91.18 87	93.24 52	72.94 227	98.37 17	93.86 155	78.78 192	97.82 34	96.13 58	95.13 136	91.05 143

E-PMN		77.81 229	77.88 231	77.73 229	88.26 220	70.48 243	80.19 241	71.20 233	86.66 168	72.89 228	88.09 186	81.74 203	78.75 193	90.02 187	68.30 239	75.10 239	59.85 241

test20.03		88.20 177	91.26 148	84.63 208	96.64 41	89.39 189	90.73 204	89.97 112	91.07 110	72.02 229	94.98 108	95.45 129	69.35 222	92.70 144	91.19 166	89.06 202	84.02 193

new-patchmatchnet		84.45 200	88.75 173	79.43 220	93.28 147	81.87 221	81.68 239	83.48 210	94.47 29	71.53 230	98.33 18	97.88 66	58.61 234	90.35 181	77.33 226	87.99 205	81.05 206

tfpn_ndepth		85.89 193	86.40 193	85.30 203	91.31 192	92.47 159	90.78 202	87.75 168	84.79 182	71.04 231	76.95 232	78.80 212	74.52 217	92.72 143	93.43 122	96.39 87	85.65 191

FMVSNet5		79.08 224	78.83 222	79.38 222	87.52 225	86.78 201	87.64 222	78.15 223	69.54 242	70.64 232	65.97 244	65.44 234	63.87 229	90.17 185	90.46 171	88.48 204	83.45 197

testmv		81.49 215	84.76 199	77.67 230	87.67 222	80.25 224	90.12 208	77.62 224	80.34 211	69.71 233	90.92 156	96.47 103	56.57 236	88.58 200	84.92 207	84.33 221	71.86 236

test1235678		81.50 214	84.78 198	77.67 230	87.67 222	80.27 223	90.12 208	77.62 224	80.36 210	69.71 233	90.93 155	96.51 102	56.57 236	88.60 199	84.93 206	84.34 220	71.87 235

MIMVSNet		84.76 199	86.75 190	82.44 212	91.71 188	85.95 204	89.74 213	89.49 129	85.28 179	69.69 235	87.93 187	90.88 174	64.85 228	88.26 203	87.74 196	89.18 201	81.24 204

testgi		86.49 190	90.31 155	82.03 213	95.63 70	88.18 193	93.47 154	84.89 201	93.23 53	69.54 236	87.16 195	97.96 61	60.66 231	91.90 169	89.90 176	87.99 205	83.84 194

test2356		72.95 236	71.24 240	74.95 233	84.89 232	75.49 235	82.67 238	75.38 229	68.02 243	68.65 237	74.40 236	52.81 244	55.61 239	81.50 229	79.80 220	82.50 228	66.70 239

TAMVS		82.96 205	86.15 195	79.24 223	90.57 204	83.12 218	87.29 223	75.12 231	84.06 186	65.81 238	92.22 143	88.27 184	69.11 223	88.72 195	87.26 199	87.56 210	79.38 216

test0.0.03 1		81.51 213	83.30 207	79.42 221	93.99 110	86.50 203	85.93 234	87.32 176	78.16 225	61.62 239	80.78 224	81.78 202	59.87 232	88.40 202	87.27 198	87.78 209	80.19 210

1111		76.85 232	78.03 229	75.46 232	94.16 101	78.29 229	86.40 230	89.12 136	87.23 163	61.26 240	95.15 101	44.14 248	51.46 240	86.04 216	81.00 216	70.40 243	74.37 228

.test1245		60.07 239	56.75 242	63.93 239	94.16 101	78.29 229	86.40 230	89.12 136	87.23 163	61.26 240	95.15 101	44.14 248	51.46 240	86.04 216	2.51 244	1.21 248	3.92 244

test12356		75.40 234	80.89 216	69.01 236	77.43 242	75.75 233	83.03 237	61.48 238	78.13 226	59.08 242	87.69 190	94.95 144	57.37 235	88.18 204	80.59 218	75.65 238	60.93 240

N_pmnet		79.33 219	84.22 201	73.62 235	91.72 187	73.72 240	86.11 232	76.36 227	92.38 68	53.38 243	95.54 93	95.62 123	59.14 233	84.23 221	74.84 234	75.03 240	73.25 232

tmp_tt				28.44 241	36.05 247	15.86 249	21.29 249	6.40 244	54.52 246	51.96 244	50.37 245	38.68 250	9.55 244	61.75 244	59.66 242	45.36 246

new_pmnet		76.65 233	83.52 204	68.63 237	82.60 234	72.08 241	76.76 244	64.17 236	84.41 185	49.73 245	91.77 146	91.53 169	56.16 238	86.59 211	83.26 210	82.37 229	75.02 224

MVE	60.41 19	73.21 235	80.84 217	64.30 238	56.34 246	57.24 247	75.28 246	72.76 232	87.14 166	41.39 246	86.31 202	85.30 193	80.66 180	86.17 215	83.36 209	59.35 244	80.38 209
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
DeepMVS_CX						47.68 248	53.20 248	19.21 243	63.24 244	26.96 247	66.50 243	69.82 227	66.91 227	64.27 243		54.91 245	72.72 233

PMMVS2		69.86 238	82.14 210	55.52 240	75.19 243	63.08 246	75.52 245	60.97 239	88.50 149	25.11 248	91.77 146	96.44 104	25.43 243	88.70 196	79.34 221	70.93 242	67.17 237

GG-mvs-BLEND		54.28 240	77.89 230	26.72 242	0.37 250	83.31 216	70.04 247	0.39 247	74.71 234	5.36 249	68.78 242	83.06 195	0.62 247	83.73 224	78.99 224	83.55 225	72.68 234

test123		2.16 242	2.82 244	1.41 243	0.62 249	1.18 250	1.53 251	0.82 246	2.78 248	2.27 250	4.18 247	1.98 252	1.64 245	2.58 246	3.01 243	1.56 247	4.00 243

testmvs		2.38 241	3.35 243	1.26 244	0.83 248	0.96 251	1.53 251	0.83 245	3.59 247	1.63 251	6.03 246	2.93 251	1.55 246	3.49 245	2.51 244	1.21 248	3.92 244

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

Anonymous202405211			94.63 75		94.51 94	94.96 97	93.94 146	91.35 82	90.82 116		95.60 89	95.85 120	81.74 176	96.47 69	95.84 65	97.39 56	92.85 109

our_test_3					91.78 186	88.87 192	94.37 134

Patchmatch-RL test							8.96 250

mPP-MVS					98.24 3							97.65 71

NP-MVS									85.48 177

High-resolution multi-view benchmark