High-res multi-view results

This table lists the benchmark results for the high-res multi-view scenario. The following metrics are evaluated:

Accuracy [%]: The fraction of the reconstruction which is closer to the ground truth than the evaluation threshold distance (*). Larger is better.
Completeness [%]: The fraction of the ground truth which is closer to the reconstruction than the evaluation threshold distance (*). Larger is better.
F₁ score [%]: The harmonic mean of accuracy and completeness, used to rank methods based on both metrics. Larger is better.
Time [s]: The runtime of the method. Smaller is better.

(*) For exact definitions, detailing how potentially incomplete ground truth is taken into account, see our paper.

The datasets are grouped into different categories, and result averages are computed for a category and method if results of the method are available for all datasets within the category. Note that the category "all" includes both the high-res multi-view and the low-res many-view scenarios.

Methods with suffix _ROB may participate in the Robust Vision Challenge.

Click a dataset result cell to show a visualization of the reconstruction. For training datasets, ground truth and accuracy / completeness visualizations are also available. The visualizations may not work with mobile browsers.

Method	all	high-res multi-view	indoor	outdoor	courty.	delive.	electro	facade	kicker	meadow	office	pipes	playgr.	relief	relief.	terrace	terrai.

Anonymous20231211		96.59 12	98.43 1	94.44 51	95.89 63	96.12 52	95.23 116	95.91 8	99.42 1	92.75 85	98.87 5	99.94 1	88.19 122	98.64 13	98.50 5	98.66 10	97.49 9

v748		96.05 30	97.00 20	94.95 45	94.41 93	94.77 97	96.72 66	91.03 88	96.12 16	96.71 11	98.74 8	99.59 2	93.55 31	97.97 29	95.96 58	97.28 57	95.84 49

pmmvs6		94.58 60	97.30 16	91.40 124	94.84 84	94.61 102	93.40 148	92.43 53	98.51 3	85.61 166	98.73 10	99.53 3	84.40 143	97.88 30	97.03 38	97.72 45	94.79 67

v7n		96.49 15	97.20 18	95.65 23	95.57 71	96.04 54	97.93 35	92.49 51	96.40 12	97.13 7	98.99 4	99.41 4	93.79 28	97.84 34	96.15 54	97.00 69	95.60 54

v52		96.35 17	97.40 14	95.12 40	93.83 119	95.54 68	97.82 39	88.95 139	96.27 14	97.22 5	99.11 2	99.40 5	95.80 5	98.16 19	96.37 49	97.10 64	96.96 23

V4		96.35 17	97.40 14	95.12 40	93.83 119	95.54 68	97.82 39	88.95 139	96.27 14	97.21 6	99.10 3	99.40 5	95.79 6	98.17 18	96.37 49	97.10 64	96.96 23

ACMH	90.17 8	96.61 11	97.69 11	95.35 31	95.29 75	96.94 32	98.43 16	92.05 65	98.04 5	95.38 22	98.07 29	99.25 7	93.23 36	98.35 16	97.16 36	97.72 45	96.00 46
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
PEN-MVS		97.16 6	97.87 6	96.33 12	97.20 20	97.97 4	98.25 25	96.86 5	95.09 25	94.93 36	98.66 11	99.16 8	92.27 53	98.98 6	98.39 8	98.49 19	96.83 31

no-one		92.05 132	94.57 76	89.12 155	85.55 220	87.65 190	94.21 135	77.34 217	93.43 49	89.64 141	95.11 101	99.11 9	95.86 4	95.38 83	95.24 75	92.08 184	96.11 45

ACMH+	89.90 10	96.27 20	97.52 12	94.81 47	95.19 77	97.18 29	97.97 33	92.52 49	96.72 9	90.50 131	97.31 59	99.11 9	94.10 23	98.67 12	97.90 15	98.56 16	95.79 50

anonymousdsp		95.45 46	96.70 25	93.99 66	88.43 208	92.05 161	99.18 1	85.42 189	94.29 36	96.10 16	98.63 13	99.08 11	96.11 1	97.77 37	97.41 29	98.70 8	97.69 7

PS-CasMVS		97.22 5	97.84 7	96.50 5	97.08 23	97.92 6	98.17 28	97.02 2	94.71 27	95.32 24	98.52 15	98.97 12	92.91 42	99.04 4	98.47 6	98.49 19	97.24 12

WR-MVS		97.53 3	98.20 4	96.76 3	96.93 27	98.17 1	98.60 11	96.67 6	96.39 13	94.46 44	99.14 1	98.92 13	94.57 18	99.06 3	98.80 2	99.32 1	96.92 27

v148		92.38 124	92.78 125	91.91 118	92.86 155	92.13 160	94.84 121	87.03 170	91.47 91	93.07 82	96.92 68	98.89 14	90.10 106	92.05 156	89.69 172	93.56 171	88.27 168

WR-MVS_H		97.06 9	97.78 8	96.23 14	96.74 36	98.04 3	98.25 25	97.32 1	94.40 34	93.71 65	98.55 14	98.89 14	92.97 39	98.91 9	98.45 7	98.38 28	97.19 14

LGP-MVS_train		96.10 28	96.29 36	95.87 20	96.72 37	97.35 22	98.43 16	93.83 35	90.81 112	92.67 89	95.05 102	98.86 16	95.01 10	98.11 20	97.37 31	98.52 17	96.50 38

OPM-MVS		95.96 32	96.59 27	95.23 36	96.67 39	96.52 42	97.86 37	93.28 43	95.27 23	93.46 70	96.26 76	98.85 17	92.89 43	97.09 48	96.37 49	97.22 61	95.78 51

DTE-MVSNet		97.16 6	97.75 9	96.47 6	97.40 12	97.95 5	98.20 27	96.89 4	95.30 20	95.15 28	98.66 11	98.80 18	92.77 46	98.97 7	98.27 10	98.44 23	96.28 41

COLMAP_ROB	93.74 2	97.09 8	97.98 5	96.05 18	95.97 58	97.78 9	98.56 12	91.72 73	97.53 7	96.01 17	98.14 26	98.76 19	95.28 8	98.76 11	98.23 11	98.77 5	96.67 36
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
SixPastTwentyTwo		97.36 4	97.73 10	96.92 2	97.36 13	96.15 51	98.29 23	94.43 23	96.50 11	96.96 8	98.74 8	98.74 20	96.04 3	99.03 5	97.74 17	98.44 23	97.22 13

v13		94.54 63	94.93 64	94.09 59	93.81 121	95.44 72	96.99 55	91.67 74	92.43 67	95.20 27	98.33 19	98.73 21	91.87 58	93.67 125	92.26 130	95.00 136	93.63 87

ACMP	89.62 11	95.96 32	96.28 37	95.59 24	96.58 42	97.23 27	98.26 24	93.22 44	92.33 70	92.31 97	94.29 117	98.73 21	94.68 15	98.04 24	97.14 37	98.47 21	96.17 44
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
v12		94.44 64	94.79 70	94.02 63	93.75 124	95.37 77	96.92 56	91.61 75	92.21 72	95.10 29	98.27 23	98.69 23	91.73 62	93.49 127	92.15 135	94.97 140	93.37 92

V9		94.33 66	94.66 73	93.94 70	93.69 128	95.31 78	96.84 61	91.53 78	92.04 78	95.00 33	98.22 24	98.64 24	91.62 64	93.29 129	92.05 137	94.93 141	93.10 97

TranMVSNet+NR-MVSNet		95.72 41	96.42 31	94.91 46	96.21 51	96.77 36	96.90 59	94.99 13	92.62 63	91.92 104	98.51 16	98.63 25	90.82 93	97.27 45	96.83 40	98.63 13	94.31 76

v11		94.32 67	94.62 74	93.97 67	93.95 111	95.31 78	96.83 62	91.30 82	91.95 79	95.51 20	98.32 21	98.61 26	91.44 69	92.83 136	92.23 132	94.77 145	93.08 98

EG-PatchMatch MVS		94.81 56	95.53 54	93.97 67	95.89 63	94.62 101	95.55 112	88.18 149	92.77 62	94.88 39	97.04 67	98.61 26	93.31 33	96.89 55	95.19 77	95.99 108	93.56 89

V14		94.21 70	94.52 77	93.85 73	93.62 129	95.25 81	96.76 65	91.42 79	91.83 82	94.91 38	98.15 25	98.57 28	91.49 68	93.06 134	91.93 141	94.90 142	92.82 102

LTVRE_ROB	95.06 1	97.73 1	98.39 2	96.95 1	96.33 48	96.94 32	98.30 22	94.90 15	98.61 2	97.73 3	97.97 32	98.57 28	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
CSCG		96.07 29	97.15 19	94.81 47	96.06 56	97.58 13	96.52 77	90.98 89	96.51 10	93.60 68	97.13 65	98.55 30	93.01 38	97.17 47	95.36 72	98.68 9	97.78 5

v15		94.09 71	94.37 81	93.77 78	93.56 131	95.18 82	96.68 71	91.34 81	91.64 86	94.83 41	98.09 28	98.51 31	91.37 71	92.84 135	91.80 143	94.85 143	92.53 113

CP-MVSNet		96.97 10	97.42 13	96.44 7	97.06 24	97.82 8	98.12 30	96.98 3	93.50 46	95.21 26	97.98 31	98.44 32	92.83 45	98.93 8	98.37 9	98.46 22	96.91 28

v1		93.48 93	93.57 106	93.37 85	93.48 133	94.18 115	96.41 87	89.61 118	91.46 92	95.03 30	97.82 39	98.43 33	90.95 90	92.00 159	91.37 159	94.75 146	89.70 147

v1141		93.47 94	93.56 107	93.36 87	93.48 133	94.17 116	96.42 85	89.62 116	91.44 95	94.99 35	97.81 40	98.42 34	90.94 91	92.00 159	91.38 157	94.74 148	89.69 149

divwei89l23v2f112		93.47 94	93.56 107	93.37 85	93.48 133	94.17 116	96.42 85	89.62 116	91.46 92	95.00 33	97.81 40	98.42 34	90.94 91	92.00 159	91.38 157	94.75 146	89.70 147

v1144		93.83 82	93.87 91	93.78 77	93.72 126	94.57 105	96.85 60	89.98 106	91.31 99	95.90 19	97.89 35	98.40 36	91.13 81	92.01 158	92.01 139	95.10 132	90.94 135

NR-MVSNet		94.55 62	95.66 53	93.25 96	94.26 98	96.44 45	96.69 69	95.32 11	89.97 121	91.79 109	97.46 56	98.39 37	82.85 152	96.87 57	96.48 47	98.57 15	93.98 82

HSP-MVS		95.04 53	95.45 57	94.57 50	96.87 28	97.77 10	98.71 5	93.88 34	91.21 102	91.48 113	95.36 93	98.37 38	90.73 94	94.37 107	92.98 121	95.77 116	98.08 3

v2v482		93.42 96	93.49 111	93.32 89	93.44 136	94.05 118	96.36 94	89.76 112	91.41 96	95.24 25	97.63 52	98.34 39	90.44 104	91.65 164	91.76 144	94.69 149	89.62 150

pm-mvs1		93.27 100	95.94 46	90.16 137	94.13 102	93.66 129	92.61 168	89.91 110	95.73 18	84.28 174	98.51 16	98.29 40	82.80 153	96.44 67	95.76 63	97.25 59	93.21 95

v144192		93.89 78	93.85 93	93.94 70	93.50 132	94.33 107	97.12 48	89.49 124	90.89 109	96.49 13	97.78 42	98.27 41	91.89 57	92.17 154	91.70 145	95.19 128	91.78 129

v1192		93.98 72	93.94 90	94.01 64	93.91 115	94.63 100	97.00 53	89.75 113	91.01 106	96.50 12	97.93 33	98.26 42	91.74 61	92.06 155	92.05 137	95.18 129	91.66 131

v7		93.65 85	93.73 99	93.57 82	93.38 137	94.60 103	96.83 62	89.92 109	89.69 128	95.02 31	97.89 35	98.24 43	91.27 73	92.38 146	92.18 133	94.99 137	91.12 133

v10		93.96 73	94.12 88	93.77 78	93.37 138	95.45 71	96.83 62	91.13 85	89.70 127	95.02 31	97.88 37	98.23 44	91.27 73	92.39 145	92.18 133	94.99 137	93.00 100

ACMM	90.06 9	96.31 19	96.42 31	96.19 15	97.21 19	97.16 30	98.71 5	93.79 36	94.35 35	93.81 61	92.80 133	98.23 44	95.11 9	98.07 23	97.45 25	98.51 18	96.86 30
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
v6		93.27 100	93.41 114	93.12 100	93.13 144	94.20 112	96.39 88	89.55 121	89.89 124	93.93 59	97.72 44	98.22 46	91.10 82	92.36 147	91.49 151	94.63 152	89.95 144

v1neww		93.27 100	93.40 116	93.12 100	93.13 144	94.20 112	96.39 88	89.56 119	89.87 125	93.95 57	97.71 46	98.21 47	91.09 83	92.36 147	91.49 151	94.62 154	89.96 142

v7new		93.27 100	93.40 116	93.12 100	93.13 144	94.20 112	96.39 88	89.56 119	89.87 125	93.95 57	97.71 46	98.21 47	91.09 83	92.36 147	91.49 151	94.62 154	89.96 142

v8		93.60 87	93.82 95	93.34 88	93.13 144	95.06 86	96.39 88	90.75 93	89.90 123	94.03 55	97.70 48	98.21 47	91.08 85	92.36 147	91.47 155	94.63 152	92.07 121

MIMVSNet1		92.52 120	94.88 67	89.77 145	96.09 53	91.99 162	96.92 56	89.68 115	95.92 17	84.55 171	96.64 72	98.21 47	78.44 189	96.08 72	95.10 78	92.91 180	90.22 140

v17		93.60 87	93.85 93	93.30 92	93.15 143	94.99 91	96.46 80	90.81 91	89.58 130	93.61 67	97.66 50	98.15 51	91.19 77	92.60 142	91.61 147	94.61 156	92.37 115

APDe-MVS		96.23 21	97.22 17	95.08 42	96.66 40	97.56 14	98.63 9	93.69 39	94.62 28	89.80 140	97.73 43	98.13 52	93.84 27	97.79 36	97.63 19	97.87 43	97.08 19

UniMVSNet (Re)		95.46 44	95.86 48	95.00 44	96.09 53	96.60 37	96.68 71	94.99 13	90.36 114	92.13 100	97.64 51	98.13 52	91.38 70	96.90 54	96.74 41	98.73 6	94.63 72

DU-MVS		95.51 43	95.68 51	95.33 32	96.45 46	96.44 45	96.61 74	95.32 11	89.97 121	93.78 62	97.46 56	98.07 54	91.19 77	97.03 49	96.53 44	98.61 14	94.22 77

v1921920		93.90 77	93.82 95	94.00 65	93.74 125	94.31 108	97.12 48	89.33 130	91.13 103	96.77 10	97.90 34	98.06 55	91.95 55	91.93 162	91.54 150	95.10 132	91.85 126

ACMMP		96.12 26	96.27 38	95.93 19	97.20 20	97.60 12	98.64 8	93.74 38	92.47 65	93.13 80	93.23 128	98.06 55	94.51 20	97.99 28	97.57 22	98.39 27	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
RPSCF		95.46 44	96.95 21	93.73 80	95.72 68	95.94 58	95.58 111	88.08 153	95.31 19	91.34 115	96.26 76	98.04 57	93.63 30	98.28 17	97.67 18	98.01 40	97.13 16

CP-MVS		96.21 22	96.16 42	96.27 13	97.56 8	97.13 31	98.43 16	94.70 17	92.62 63	94.13 53	92.71 134	98.03 58	94.54 19	98.00 27	97.60 20	98.23 30	97.05 20

v16		93.53 92	93.80 98	93.20 97	93.10 149	94.98 92	96.43 82	90.81 91	89.39 133	93.12 81	97.63 52	98.01 59	91.19 77	92.60 142	91.65 146	94.58 158	92.36 116

V42		92.67 117	93.50 110	91.71 121	91.41 181	92.96 144	95.71 106	85.00 190	89.67 129	93.22 76	97.67 49	98.01 59	91.02 88	92.65 139	92.12 136	93.86 168	91.42 132

testgi		86.49 182	90.31 148	82.03 205	95.63 70	88.18 182	93.47 146	84.89 192	93.23 54	69.54 228	87.16 186	97.96 61	60.66 222	91.90 163	89.90 170	87.99 196	83.84 186

v18		93.33 99	93.59 105	93.04 105	92.94 152	94.87 94	96.31 95	90.59 97	88.96 135	92.89 84	97.51 54	97.90 62	91.01 89	92.33 151	91.48 154	94.50 159	92.05 122

v1240		93.89 78	93.72 100	94.09 59	93.98 108	94.31 108	97.12 48	89.37 128	90.74 113	96.92 9	98.05 30	97.89 63	92.15 54	91.53 166	91.60 148	94.99 137	91.93 125

TransMVSNet (Re)		93.55 91	96.32 35	90.32 136	94.38 94	94.05 118	93.30 155	89.53 122	97.15 8	85.12 168	98.83 6	97.89 63	82.21 158	96.75 61	96.14 55	97.35 55	93.46 90

Baseline_NR-MVSNet		94.85 55	95.35 59	94.26 56	96.45 46	93.86 126	96.70 67	94.54 18	90.07 119	90.17 136	98.77 7	97.89 63	90.64 98	97.03 49	96.16 53	97.04 68	93.67 85

new-patchmatchnet		84.45 192	88.75 167	79.43 212	93.28 140	81.87 211	81.68 229	83.48 201	94.47 30	71.53 222	98.33 19	97.88 66	58.61 225	90.35 174	77.33 217	87.99 196	81.05 198

Gipuma		95.86 37	96.17 40	95.50 29	95.92 60	94.59 104	94.77 123	92.50 50	97.82 6	97.90 2	95.56 88	97.88 66	94.71 14	98.02 25	94.81 86	97.23 60	94.48 75
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
TDRefinement		97.59 2	98.32 3	96.73 4	95.90 61	98.10 2	99.08 2	93.92 32	98.24 4	96.44 15	98.12 27	97.86 68	96.06 2	99.24 1	98.93 1	99.00 2	97.77 6

OMC-MVS		94.74 57	95.46 56	93.91 72	94.62 88	96.26 48	96.64 73	89.36 129	94.20 37	94.15 52	94.02 122	97.73 69	91.34 72	96.15 71	95.04 80	97.37 54	94.80 66

APD-MVS		95.38 47	95.68 51	95.03 43	97.30 17	96.90 34	97.83 38	93.92 32	89.40 131	90.35 132	95.41 92	97.69 70	92.97 39	97.24 46	97.17 35	97.83 44	95.96 47
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
mPP-MVS					98.24 3							97.65 71

CPTT-MVS		95.00 54	94.52 77	95.57 26	96.84 32	96.78 35	97.88 36	93.67 40	92.20 73	92.35 96	85.87 196	97.56 72	94.98 12	96.96 52	96.07 57	97.70 47	96.18 43

FMVSNet1		92.86 113	95.26 60	90.06 139	92.40 163	95.16 83	94.37 130	92.22 56	93.18 55	82.16 188	96.76 70	97.48 73	81.85 163	95.32 86	94.98 81	97.34 56	93.93 83

CLD-MVS		92.81 114	94.32 83	91.05 126	95.39 73	95.31 78	95.82 102	81.44 211	89.40 131	91.94 103	95.86 83	97.36 74	85.83 136	95.35 84	94.59 98	95.85 113	92.34 118
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
ACMMPR		96.54 14	96.71 24	96.35 11	97.55 9	97.63 11	98.62 10	94.54 18	94.45 31	94.19 50	95.04 104	97.35 75	94.92 13	97.85 32	97.50 23	98.26 29	97.17 15

ESAPD		95.63 42	96.35 33	94.80 49	96.76 35	97.29 24	97.74 42	94.15 28	91.69 84	90.01 137	96.65 71	97.29 76	92.45 48	97.41 42	97.18 34	97.67 50	96.95 25

MTMP										95.43 21		97.25 77

UniMVSNet_NR-MVSNet		95.34 48	95.51 55	95.14 39	95.80 66	96.55 38	96.61 74	94.79 16	90.04 120	93.78 62	97.51 54	97.25 77	91.19 77	96.68 62	96.31 52	98.65 12	94.22 77

PGM-MVS		95.90 35	95.72 50	96.10 16	97.53 10	97.45 19	98.55 13	94.12 29	90.25 115	93.71 65	93.20 129	97.18 79	94.63 16	97.68 40	97.34 32	98.08 36	96.97 22

MVS_0304		93.92 75	93.81 97	94.05 62	96.06 56	96.00 56	96.43 82	92.76 47	85.99 169	94.43 46	94.04 121	97.08 80	88.12 124	94.65 104	94.20 104	96.47 78	94.71 69

EU-MVSNet		91.63 134	92.73 126	90.35 135	88.36 209	87.89 187	96.53 76	81.51 210	92.45 66	91.82 108	96.44 75	97.05 81	93.26 35	94.10 119	88.94 182	90.61 187	92.24 119

pmmvs-eth3d		92.34 125	92.33 129	92.34 112	92.67 158	90.67 173	96.37 92	89.06 133	90.98 107	93.60 68	97.13 65	97.02 82	88.29 121	90.20 175	91.42 156	94.07 166	88.89 158

zzz-MVS		96.18 23	96.01 44	96.38 8	98.30 2	96.18 50	98.51 14	94.48 22	94.56 29	94.81 42	91.73 143	96.96 83	94.30 22	98.09 21	97.83 16	97.91 42	96.73 33

HFP-MVS		96.18 23	96.53 29	95.77 21	97.34 16	97.26 25	98.16 29	94.54 18	94.45 31	92.52 91	95.05 102	96.95 84	93.89 26	97.28 44	97.46 24	98.19 31	97.25 10

DeepC-MVS	92.47 4	96.44 16	96.75 22	96.08 17	97.57 7	97.19 28	97.96 34	94.28 24	95.29 21	94.92 37	98.31 22	96.92 85	93.69 29	96.81 59	96.50 46	98.06 38	96.27 42
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
SteuartSystems-ACMMP		95.96 32	96.13 43	95.76 22	97.06 24	97.36 21	98.40 20	94.24 26	91.49 89	91.91 105	94.50 112	96.89 86	94.99 11	98.01 26	97.44 26	97.97 41	97.25 10
Skip Steuart: Steuart Systems R&D Blog.
MTAPA										94.88 39		96.88 87

TAPA-MVS	88.94 13	93.78 84	94.31 84	93.18 99	94.14 101	95.99 57	95.74 104	86.98 171	93.43 49	93.88 60	90.16 159	96.88 87	91.05 86	94.33 108	93.95 105	97.28 57	95.40 56
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
Anonymous20231206		87.45 179	89.66 154	84.87 197	94.00 104	87.73 189	91.36 186	86.41 180	88.89 138	75.03 213	92.59 135	96.82 89	72.48 211	89.72 180	88.06 184	89.93 190	83.81 187

XVS					96.86 30	97.48 16	98.73 3			93.28 73		96.82 89				98.17 33

X-MVStestdata					96.86 30	97.48 16	98.73 3			93.28 73		96.82 89				98.17 33

X-MVS		95.33 49	95.13 62	95.57 26	97.35 14	97.48 16	98.43 16	94.28 24	92.30 71	93.28 73	86.89 189	96.82 89	91.87 58	97.85 32	97.59 21	98.19 31	96.95 25

SD-MVS		95.77 40	96.17 40	95.30 34	96.72 37	96.19 49	97.01 52	93.04 45	94.03 41	92.71 86	96.45 74	96.78 93	93.91 25	96.79 60	95.89 61	98.42 25	97.09 18

USDC		92.17 129	92.17 132	92.18 116	92.93 153	92.22 158	93.66 143	87.41 164	93.49 47	97.99 1	94.10 119	96.68 94	86.46 132	92.04 157	89.18 178	94.61 156	87.47 171

PM-MVS		92.65 118	93.20 121	92.00 117	92.11 174	90.16 177	95.99 98	84.81 193	91.31 99	92.41 94	95.87 82	96.64 95	92.35 51	93.65 126	92.91 122	94.34 163	91.85 126

TSAR-MVS + MP.		95.99 31	96.57 28	95.31 33	96.87 28	96.50 43	98.71 5	91.58 76	93.25 52	92.71 86	96.86 69	96.57 96	93.92 24	98.09 21	97.91 14	98.08 36	96.81 32

test1235678		81.50 206	84.78 190	77.67 222	87.67 212	80.27 213	90.12 198	77.62 215	80.36 201	69.71 225	90.93 149	96.51 97	56.57 227	88.60 190	84.93 197	84.34 211	71.87 227

testmv		81.49 207	84.76 191	77.67 222	87.67 212	80.25 214	90.12 198	77.62 215	80.34 202	69.71 225	90.92 150	96.47 98	56.57 227	88.58 191	84.92 198	84.33 212	71.86 228

PMMVS2		69.86 230	82.14 201	55.52 232	75.19 233	63.08 236	75.52 235	60.97 230	88.50 143	25.11 240	91.77 141	96.44 99	25.43 234	88.70 187	79.34 212	70.93 233	67.17 229

FC-MVSNet-train		92.75 116	95.40 58	89.66 149	95.21 76	94.82 95	97.00 53	89.40 127	91.13 103	81.71 189	97.72 44	96.43 100	77.57 196	96.89 55	96.72 42	97.05 67	94.09 80

MP-MVS		96.13 25	95.93 47	96.37 9	98.19 4	97.31 23	98.49 15	94.53 21	91.39 97	94.38 47	94.32 116	96.43 100	94.59 17	97.75 38	97.44 26	98.04 39	96.88 29

ambc			94.61 75		98.09 5	95.14 84	91.71 182		94.18 38	96.46 14	96.26 76	96.30 102	91.26 75	94.70 102	92.00 140	93.45 172	93.67 85

Vis-MVSNet		94.39 65	95.85 49	92.68 108	90.91 190	95.88 59	97.62 46	91.41 80	91.95 79	89.20 142	97.29 60	96.26 103	90.60 102	96.95 53	95.91 59	96.32 91	96.71 34
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
HQP-MVS		92.87 112	92.49 128	93.31 90	95.75 67	95.01 90	95.64 109	91.06 87	88.54 142	91.62 112	88.16 175	96.25 104	89.47 110	92.26 153	91.81 142	96.34 88	95.40 56

tfpnnormal		92.45 121	94.77 71	89.74 146	93.95 111	93.44 139	93.25 156	88.49 147	95.27 23	83.20 178	96.51 73	96.23 105	83.17 150	95.47 81	94.52 100	96.38 82	91.97 124

Fast-Effi-MVS+		92.93 110	92.64 127	93.27 93	93.81 121	93.88 125	95.90 99	90.61 96	83.98 181	92.71 86	92.81 132	96.22 106	90.67 96	94.90 100	93.92 107	95.92 110	92.77 105

MVS_111021_LR		93.15 105	93.65 102	92.56 109	93.89 117	92.28 157	95.09 117	86.92 173	91.26 101	92.99 83	94.46 114	96.22 106	90.64 98	95.11 94	93.45 115	95.85 113	92.74 106

MDTV_nov1_ep13_2view		88.22 169	87.85 176	88.65 161	91.40 182	86.75 192	94.07 138	84.97 191	88.86 139	93.20 77	96.11 80	96.21 108	83.70 147	87.29 200	80.29 210	84.56 209	79.46 207

pmmvs5		88.63 161	89.70 153	87.39 181	89.24 202	90.64 174	91.87 176	82.13 206	83.34 184	87.86 153	94.58 110	96.15 109	79.87 175	87.33 199	89.07 181	93.39 174	86.76 177

MSLP-MVS++		93.91 76	94.30 85	93.45 84	95.51 72	95.83 61	93.12 158	91.93 69	91.45 94	91.40 114	87.42 184	96.12 110	93.27 34	96.57 66	96.40 48	95.49 120	96.29 40

CDPH-MVS		93.96 73	93.86 92	94.08 61	96.31 49	95.84 60	96.92 56	91.85 70	87.21 160	91.25 119	92.83 131	96.06 111	91.05 86	95.57 80	94.81 86	97.12 62	94.72 68

MVS_111021_HR		93.82 83	94.26 87	93.31 90	95.01 80	93.97 123	95.73 105	89.75 113	92.06 77	92.49 92	94.01 123	96.05 112	90.61 101	95.95 74	94.78 89	96.28 93	93.04 99

MCST-MVS		93.60 87	93.40 116	93.83 74	95.30 74	95.40 75	96.49 78	90.87 90	90.08 118	91.72 110	90.28 157	95.99 113	91.69 63	93.94 121	92.99 120	96.93 71	95.13 63

Effi-MVS+-dtu		92.32 127	91.66 138	93.09 104	95.13 79	94.73 98	94.57 128	92.14 61	81.74 189	90.33 133	88.13 176	95.91 114	89.24 111	94.23 117	93.65 113	97.12 62	93.23 94

train_agg		93.89 78	93.46 112	94.40 54	97.35 14	93.78 127	97.63 45	92.19 59	88.12 145	90.52 130	93.57 126	95.78 115	92.31 52	94.78 101	93.46 114	96.36 83	94.70 71

N_pmnet		79.33 211	84.22 193	73.62 227	91.72 178	73.72 230	86.11 222	76.36 218	92.38 68	53.38 235	95.54 91	95.62 116	59.14 224	84.23 212	74.84 225	75.03 231	73.25 224

NCCC		93.87 81	93.42 113	94.40 54	96.84 32	95.42 73	96.47 79	92.62 48	92.36 69	92.05 101	83.83 205	95.55 117	91.84 60	95.89 75	95.23 76	96.56 76	95.63 53

PVSNet_BlendedMVS		90.09 144	90.12 149	90.05 140	92.40 163	92.74 148	91.74 179	85.89 182	80.54 199	90.30 134	88.54 171	95.51 118	84.69 141	92.64 140	90.25 168	95.28 125	90.61 137

PVSNet_Blended		90.09 144	90.12 149	90.05 140	92.40 163	92.74 148	91.74 179	85.89 182	80.54 199	90.30 134	88.54 171	95.51 118	84.69 141	92.64 140	90.25 168	95.28 125	90.61 137

UGNet		92.31 128	94.70 72	89.53 151	90.99 189	95.53 70	96.19 96	92.10 64	91.35 98	85.76 163	95.31 95	95.48 120	76.84 200	95.22 91	94.79 88	95.32 122	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
test20.03		88.20 170	91.26 142	84.63 200	96.64 41	89.39 179	90.73 194	89.97 107	91.07 105	72.02 221	94.98 105	95.45 121	69.35 213	92.70 138	91.19 160	89.06 193	84.02 185

DeepC-MVS_fast	91.38 6	94.73 58	94.98 63	94.44 51	96.83 34	96.12 52	96.69 69	92.17 60	92.98 57	93.72 64	94.14 118	95.45 121	90.49 103	95.73 79	95.30 73	96.71 73	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
PMVS	87.16 16	95.88 36	96.47 30	95.19 38	97.00 26	96.02 55	96.70 67	91.57 77	94.43 33	95.33 23	97.16 63	95.37 123	92.39 50	98.89 10	98.72 3	98.17 33	94.71 69
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
CNVR-MVS		94.24 69	94.47 79	93.96 69	96.56 43	95.67 65	96.43 82	91.95 67	92.08 76	91.28 117	90.51 155	95.35 124	91.20 76	96.34 70	95.50 70	96.34 88	95.88 48

Effi-MVS+		92.93 110	92.16 133	93.83 74	94.29 96	93.53 137	95.04 118	92.98 46	85.27 173	94.46 44	90.24 158	95.34 125	89.99 107	93.72 123	94.23 103	96.22 99	92.79 104

PHI-MVS		94.65 59	94.84 68	94.44 51	94.95 82	96.55 38	96.46 80	91.10 86	88.96 135	96.00 18	94.55 111	95.32 126	90.67 96	96.97 51	96.69 43	97.44 52	94.84 65

TinyColmap		93.17 104	93.33 119	93.00 106	93.84 118	92.76 146	94.75 125	88.90 141	93.97 42	97.48 4	95.28 98	95.29 127	88.37 120	95.31 89	91.58 149	94.65 151	89.10 155

DELS-MVS		92.33 126	93.61 104	90.83 129	92.84 156	95.13 85	94.76 124	87.22 169	87.78 152	88.42 151	95.78 85	95.28 128	85.71 137	94.44 106	93.91 108	96.01 107	92.97 101
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
TSAR-MVS + COLMAP		93.06 109	93.65 102	92.36 111	94.62 88	94.28 110	95.36 115	89.46 126	92.18 74	91.64 111	95.55 89	95.27 129	88.60 118	93.24 130	92.50 127	94.46 160	92.55 112

CANet		93.07 108	93.05 122	93.10 103	95.90 61	95.41 74	95.88 100	91.94 68	84.77 176	93.36 71	94.05 120	95.25 130	86.25 134	94.33 108	93.94 106	95.30 123	93.58 88

TSAR-MVS + ACMM		95.17 52	95.95 45	94.26 56	96.07 55	96.46 44	95.67 108	94.21 27	93.84 43	90.99 123	97.18 62	95.24 131	93.55 31	96.60 65	95.61 68	95.06 134	96.69 35

LS3D		95.83 39	96.35 33	95.22 37	96.47 45	97.49 15	97.99 31	92.35 54	94.92 26	94.58 43	94.88 107	95.11 132	91.52 67	98.48 14	98.05 13	98.42 25	95.49 55

HPM-MVS++		95.21 51	94.89 66	95.59 24	97.79 6	95.39 76	97.68 43	94.05 30	91.91 81	94.35 48	93.38 127	95.07 133	92.94 41	96.01 73	95.88 62	96.73 72	96.61 37

UA-Net		96.56 13	96.73 23	96.36 10	98.99 1	97.90 7	97.79 41	95.64 10	92.78 61	92.54 90	96.23 79	95.02 134	94.31 21	98.43 15	98.12 12	98.89 3	98.58 2

MVS_Test		90.19 142	90.58 144	89.74 146	92.12 173	91.74 164	92.51 169	88.54 146	82.80 186	87.50 155	94.62 109	95.02 134	83.97 145	88.69 188	89.32 176	93.79 169	91.85 126

test12356		75.40 226	80.89 207	69.01 228	77.43 232	75.75 223	83.03 227	61.48 229	78.13 217	59.08 234	87.69 181	94.95 136	57.37 226	88.18 194	80.59 209	75.65 229	60.93 232

SMA-MVS		96.11 27	96.61 26	95.53 28	97.49 11	97.41 20	97.62 46	93.78 37	94.14 40	94.18 51	97.16 63	94.67 137	92.42 49	97.74 39	97.33 33	97.70 47	97.79 4

CNLPA		93.14 106	93.67 101	92.53 110	94.62 88	94.73 98	95.00 119	86.57 177	92.85 60	92.43 93	90.94 148	94.67 137	90.35 105	95.41 82	93.70 109	96.23 98	93.37 92

MSDG		92.09 131	92.84 124	91.22 125	92.55 159	92.97 143	93.42 147	85.43 188	90.24 116	91.83 107	94.70 108	94.59 139	88.48 119	94.91 99	93.31 119	95.59 119	89.15 154

abl_6				91.88 119	93.76 123	94.98 92	95.64 109	88.97 136	86.20 167	90.00 138	86.31 193	94.50 140	87.31 126			95.60 118	92.48 114

TSAR-MVS + GP.		94.25 68	94.81 69	93.60 81	96.52 44	95.80 62	94.37 130	92.47 52	90.89 109	88.92 143	95.34 94	94.38 141	92.85 44	96.36 69	95.62 67	96.47 78	95.28 60

MDA-MVSNet-bldmvs		89.75 147	91.67 137	87.50 179	74.25 234	90.88 171	94.68 126	85.89 182	91.64 86	91.03 121	95.86 83	94.35 142	89.10 113	96.87 57	86.37 192	90.04 188	85.72 182

DeepPCF-MVS	90.68 7	94.56 61	94.92 65	94.15 58	94.11 103	95.71 64	97.03 51	90.65 95	93.39 51	94.08 54	95.29 96	94.15 143	93.21 37	95.22 91	94.92 84	95.82 115	95.75 52

ACMMP_Plus		95.86 37	96.18 39	95.47 30	97.11 22	97.26 25	98.37 21	93.48 42	93.49 47	93.99 56	95.61 86	94.11 144	92.49 47	97.87 31	97.44 26	97.40 53	97.52 8

PCF-MVS	87.46 14	92.44 122	91.80 136	93.19 98	94.66 86	95.80 62	96.37 92	90.19 100	87.57 153	92.23 98	89.26 166	93.97 145	89.24 111	91.32 168	90.82 164	96.46 80	93.86 84
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
PLC	87.27 15	93.08 107	92.92 123	93.26 94	94.67 85	95.03 87	94.38 129	90.10 101	91.69 84	92.14 99	87.24 185	93.91 146	91.61 65	95.05 95	94.73 95	96.67 75	92.80 103
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
FC-MVSNet-test		91.49 135	94.43 80	88.07 174	94.97 81	90.53 176	95.42 113	91.18 84	93.24 53	72.94 219	98.37 18	93.86 147	78.78 183	97.82 35	96.13 56	95.13 130	91.05 134

canonicalmvs		93.38 97	94.36 82	92.24 114	93.94 113	96.41 47	94.18 136	90.47 99	93.07 56	88.47 149	88.66 170	93.78 148	88.80 115	95.74 78	95.75 64	97.57 51	97.13 16

3Dnovator+	92.82 3	95.22 50	95.16 61	95.29 35	96.17 52	96.55 38	97.64 44	94.02 31	94.16 39	94.29 49	92.09 140	93.71 149	91.90 56	96.68 62	96.51 45	97.70 47	96.40 39

CANet_DTU		88.95 158	89.51 155	88.29 171	93.12 148	91.22 168	93.61 144	83.47 202	80.07 204	90.71 128	89.19 167	93.68 150	76.27 204	91.44 167	91.17 161	92.59 181	89.83 145

IB-MVS	86.01 17	88.24 168	87.63 177	88.94 157	92.03 177	91.77 163	92.40 172	85.58 186	78.24 215	84.85 169	71.99 231	93.45 151	83.96 146	93.48 128	92.33 128	94.84 144	92.15 120
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
MAR-MVS		91.86 133	91.14 143	92.71 107	94.29 96	94.24 111	94.91 120	91.82 71	81.66 190	93.32 72	84.51 203	93.42 152	86.86 130	95.16 93	94.44 101	95.05 135	94.53 74
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
3Dnovator	91.81 5	93.36 98	94.27 86	92.29 113	92.99 151	95.03 87	95.76 103	87.79 158	93.82 44	92.38 95	92.19 139	93.37 153	88.14 123	95.26 90	94.85 85	96.69 74	95.40 56

testus		78.20 220	81.50 205	74.36 226	85.59 219	79.36 216	86.99 216	65.76 226	76.01 221	73.00 218	77.98 222	93.35 154	51.30 233	86.33 203	82.79 202	83.50 217	74.68 219

Fast-Effi-MVS+-dtu		89.57 149	88.42 171	90.92 128	93.35 139	91.57 165	93.01 160	95.71 9	78.94 213	87.65 154	84.68 202	93.14 155	82.00 160	90.84 171	91.01 162	93.78 170	88.77 159

OpenMVS	89.22 12	91.09 137	91.42 141	90.71 130	92.79 157	93.61 134	92.74 167	85.47 187	86.10 168	90.73 124	85.71 197	93.07 156	86.69 131	94.07 120	93.34 118	95.86 111	94.02 81

FMVSNet2		90.28 141	92.04 134	88.23 172	91.22 185	94.05 118	92.88 163	90.69 94	86.53 164	79.89 199	94.38 115	92.73 157	78.54 186	91.64 165	92.26 130	96.17 102	92.67 107

PVSNet_Blended_VisFu		93.60 87	93.41 114	93.83 74	96.31 49	95.65 66	95.71 106	90.58 98	88.08 148	93.17 78	95.29 96	92.20 158	90.72 95	94.69 103	93.41 117	96.51 77	94.54 73

pmmvs3		81.69 203	83.83 194	79.19 216	78.33 230	78.57 218	89.53 204	58.71 232	78.88 214	84.34 173	88.36 173	91.96 159	77.69 195	87.48 197	82.42 203	86.54 204	79.18 209

QAPM		92.57 119	93.51 109	91.47 122	92.91 154	94.82 95	93.01 160	87.51 162	91.49 89	91.21 120	92.24 137	91.70 160	88.74 116	94.54 105	94.39 102	95.41 121	95.37 59

new_pmnet		76.65 225	83.52 196	68.63 229	82.60 224	72.08 231	76.76 234	64.17 227	84.41 178	49.73 237	91.77 141	91.53 161	56.16 229	86.59 201	83.26 201	82.37 220	75.02 216

PMMVS		81.93 202	83.48 198	80.12 210	72.35 235	75.05 228	88.54 209	64.01 228	77.02 220	82.22 187	87.51 183	91.12 162	79.70 176	86.59 201	86.64 190	93.88 167	80.41 200

CMPMVS	66.55 18	85.55 187	87.46 180	83.32 203	84.99 221	81.97 210	79.19 232	75.93 219	79.32 208	88.82 145	85.09 198	91.07 163	82.12 159	92.56 144	89.63 174	88.84 194	92.56 111
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
AdaColmap		92.41 123	91.49 140	93.48 83	95.96 59	95.02 89	95.37 114	91.73 72	87.97 151	91.28 117	82.82 210	91.04 164	90.62 100	95.82 77	95.07 79	95.95 109	92.67 107

IterMVS-LS		92.10 130	92.33 129	91.82 120	93.18 141	93.66 129	92.80 166	92.27 55	90.82 111	90.59 129	97.19 61	90.97 165	87.76 125	89.60 181	90.94 163	94.34 163	93.16 96
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
MIMVSNet		84.76 191	86.75 182	82.44 204	91.71 179	85.95 194	89.74 203	89.49 124	85.28 172	69.69 227	87.93 178	90.88 166	64.85 219	88.26 193	87.74 186	89.18 192	81.24 196

PatchMatch-RL		89.59 148	88.80 165	90.51 133	92.20 172	88.00 186	91.72 181	86.64 174	84.75 177	88.25 152	87.10 187	90.66 167	89.85 109	93.23 131	92.28 129	94.41 162	85.60 184

pmmvs4		89.95 146	89.32 158	90.69 131	91.60 180	89.17 181	94.37 130	87.63 161	88.07 149	91.02 122	94.50 112	90.50 168	86.13 135	86.33 203	89.40 175	93.39 174	87.29 173

CDS-MVSNet		88.41 163	89.79 151	86.79 186	94.55 92	90.82 172	92.50 170	89.85 111	83.26 185	80.52 195	91.05 145	89.93 169	69.11 214	93.17 133	92.71 125	94.21 165	87.63 170
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
EPNet_dtu		87.40 180	86.27 186	88.72 160	95.68 69	83.37 205	92.09 174	90.08 102	78.11 218	91.29 116	86.33 192	89.74 170	75.39 205	89.07 185	87.89 185	87.81 198	89.38 152
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
FPMVS		90.81 138	91.60 139	89.88 143	92.52 160	88.18 182	93.31 154	83.62 199	91.59 88	88.45 150	88.96 168	89.73 171	86.96 128	96.42 68	95.69 66	94.43 161	90.65 136

CHOSEN 1792x2688		86.64 181	86.62 183	86.65 189	90.33 195	87.86 188	93.19 157	83.30 203	83.95 182	82.32 186	87.93 178	89.34 172	86.92 129	85.64 209	84.95 196	83.85 215	86.68 178

MS-PatchMatch		87.72 176	88.62 170	86.66 188	90.81 192	88.18 182	90.92 190	82.25 205	85.86 170	80.40 198	90.14 160	89.29 173	84.93 138	89.39 184	89.12 179	90.67 186	88.34 167

Vis-MVSNet (Re-imp)		90.68 139	92.18 131	88.92 158	94.63 87	92.75 147	92.91 162	91.20 83	89.21 134	75.01 214	93.96 124	89.07 174	82.72 155	95.88 76	95.30 73	97.08 66	89.08 156

diffmvs		88.28 167	88.88 162	87.58 178	89.51 200	88.07 185	91.88 175	85.83 185	87.31 155	86.34 161	96.01 81	88.90 175	81.90 161	85.49 210	86.61 191	90.04 188	89.77 146

IS_MVSNet		92.76 115	93.25 120	92.19 115	94.91 83	95.56 67	95.86 101	92.12 62	88.10 146	82.71 183	93.15 130	88.30 176	88.86 114	97.29 43	96.95 39	98.66 10	93.38 91

TAMVS		82.96 197	86.15 187	79.24 215	90.57 193	83.12 208	87.29 213	75.12 222	84.06 179	65.81 230	92.22 138	88.27 177	69.11 214	88.72 186	87.26 189	87.56 201	79.38 208

DI_MVS_plusplus_trai		90.68 139	90.40 147	91.00 127	92.43 162	92.61 151	94.17 137	88.98 135	88.32 144	88.76 147	93.67 125	87.58 178	86.44 133	89.74 179	90.33 166	95.24 127	90.56 139

GBi-Net		89.35 151	90.58 144	87.91 175	91.22 185	94.05 118	92.88 163	90.05 103	79.40 205	78.60 205	90.58 152	87.05 179	78.54 186	95.32 86	94.98 81	96.17 102	92.67 107

test1		89.35 151	90.58 144	87.91 175	91.22 185	94.05 118	92.88 163	90.05 103	79.40 205	78.60 205	90.58 152	87.05 179	78.54 186	95.32 86	94.98 81	96.17 102	92.67 107

FMVSNet3		87.90 175	88.63 169	87.04 183	89.78 199	93.46 138	91.62 184	90.05 103	79.40 205	78.60 205	90.58 152	87.05 179	77.07 199	88.03 195	89.86 171	95.12 131	92.04 123

IterMVS		88.32 164	88.25 172	88.41 166	90.83 191	91.24 167	93.07 159	81.69 208	86.77 162	88.55 148	95.61 86	86.91 182	87.01 127	87.38 198	83.77 199	89.29 191	86.06 181
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
EPNet		90.17 143	89.07 160	91.45 123	97.25 18	90.62 175	94.84 121	93.54 41	80.96 192	91.85 106	86.98 188	85.88 183	77.79 193	92.30 152	92.58 126	93.41 173	94.20 79
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
CVMVSNet		88.97 157	89.73 152	88.10 173	87.33 216	85.22 196	94.68 126	78.68 213	88.94 137	86.98 159	95.55 89	85.71 184	89.87 108	91.19 169	89.69 172	91.05 185	91.78 129

HyFIR lowres test		88.19 171	86.56 184	90.09 138	91.24 184	92.17 159	94.30 134	88.79 143	84.06 179	85.45 167	89.52 164	85.64 185	88.64 117	85.40 211	87.28 187	92.14 183	81.87 195

MVE	60.41 19	73.21 227	80.84 208	64.30 230	56.34 236	57.24 237	75.28 236	72.76 223	87.14 161	41.39 238	86.31 193	85.30 186	80.66 171	86.17 205	83.36 200	59.35 235	80.38 201
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
EPP-MVSNet		93.63 86	93.95 89	93.26 94	95.15 78	96.54 41	96.18 97	91.97 66	91.74 83	85.76 163	94.95 106	84.27 187	91.60 66	97.61 41	97.38 30	98.87 4	95.18 62

GG-mvs-BLEND		54.28 232	77.89 221	26.72 234	0.37 240	83.31 206	70.04 237	0.39 238	74.71 225	5.36 241	68.78 233	83.06 188	0.62 238	83.73 215	78.99 215	83.55 216	72.68 226

tfpn111		87.59 178	86.89 181	88.41 166	92.28 167	93.64 131	93.36 149	88.12 150	80.90 193	80.71 193	73.93 228	82.25 189	79.65 178	94.27 112	94.76 90	96.36 83	88.48 163

conf200view11		87.93 174	87.51 178	88.41 166	92.28 167	93.64 131	93.36 149	88.12 150	80.90 193	80.71 193	78.25 219	82.25 189	79.65 178	94.27 112	94.76 90	96.36 83	88.48 163

thres100view900		86.46 183	86.00 188	86.99 184	92.28 167	91.03 169	91.09 188	84.49 195	80.90 193	80.89 191	78.25 219	82.25 189	77.57 196	90.17 176	92.84 123	95.63 117	86.57 179

tfpn200view9		87.94 173	87.51 178	88.44 165	92.28 167	93.63 133	93.35 153	88.11 152	80.90 193	80.89 191	78.25 219	82.25 189	79.65 178	94.27 112	94.76 90	96.36 83	88.48 163

thres200		88.29 166	87.88 175	88.76 159	92.50 161	93.55 135	92.47 171	88.02 154	84.80 174	81.44 190	79.28 218	82.20 193	81.83 164	94.27 112	93.67 110	96.27 95	87.40 172

GA-MVS		88.76 159	88.04 174	89.59 150	92.32 166	91.46 166	92.28 173	86.62 175	83.82 183	89.84 139	92.51 136	81.94 194	83.53 148	89.41 183	89.27 177	92.95 179	87.90 169

test0.0.03 1		81.51 205	83.30 199	79.42 213	93.99 106	86.50 193	85.93 224	87.32 166	78.16 216	61.62 231	80.78 215	81.78 195	59.87 223	88.40 192	87.27 188	87.78 200	80.19 202

E-PMN		77.81 221	77.88 222	77.73 221	88.26 210	70.48 233	80.19 231	71.20 224	86.66 163	72.89 220	88.09 177	81.74 196	78.75 184	90.02 178	68.30 230	75.10 230	59.85 233

thres400		88.54 162	88.15 173	88.98 156	93.17 142	92.84 145	93.56 145	86.93 172	86.45 165	82.37 185	79.96 216	81.46 197	81.83 164	93.21 132	94.76 90	96.04 106	88.39 166

MVS-HIRNet		78.28 219	75.28 229	81.79 207	80.33 227	69.38 234	76.83 233	86.59 176	70.76 230	86.66 160	89.57 163	81.04 198	77.74 194	77.81 228	71.65 227	82.62 218	66.73 230

view600		89.09 154	88.78 166	89.46 153	93.59 130	93.33 141	93.92 141	87.76 159	87.40 154	82.79 180	81.29 213	80.71 199	82.59 157	94.28 111	95.72 65	96.12 105	88.70 160

thres600view7		89.14 153	88.83 163	89.51 152	93.71 127	93.55 135	93.93 140	88.02 154	87.30 157	82.40 184	81.18 214	80.63 200	82.69 156	94.27 112	95.90 60	96.27 95	88.94 157

EMVS		77.65 222	77.49 224	77.83 219	87.75 211	71.02 232	81.13 230	70.54 225	86.38 166	74.52 215	89.38 165	80.19 201	78.22 191	89.48 182	67.13 231	74.83 232	58.84 234

view800		89.42 150	89.11 159	89.78 144	94.00 104	93.71 128	93.96 139	88.47 148	88.10 146	82.91 179	82.61 211	79.85 202	83.10 151	94.92 98	95.38 71	96.26 97	89.19 153

LP		84.09 194	84.31 192	83.85 202	79.40 229	84.34 202	90.26 197	84.02 196	87.99 150	84.66 170	91.61 144	79.13 203	80.58 172	85.90 208	81.59 205	84.16 214	79.59 206

conf0.05thres1000		91.24 136	91.85 135	90.53 132	94.59 91	94.56 106	94.33 133	89.52 123	93.67 45	83.77 176	91.04 146	79.10 204	83.98 144	96.66 64	95.56 69	96.98 70	92.36 116

tfpn_ndepth		85.89 185	86.40 185	85.30 195	91.31 183	92.47 152	90.78 192	87.75 160	84.79 175	71.04 223	76.95 223	78.80 205	74.52 208	92.72 137	93.43 116	96.39 81	85.65 183

tfpn_n400		89.03 155	89.39 156	88.61 162	93.98 108	92.33 154	91.83 177	88.97 136	92.97 58	78.90 201	84.93 199	78.24 206	81.77 166	95.00 96	93.67 110	96.22 99	88.59 161

tfpnconf		89.03 155	89.39 156	88.61 162	93.98 108	92.33 154	91.83 177	88.97 136	92.97 58	78.90 201	84.93 199	78.24 206	81.77 166	95.00 96	93.67 110	96.22 99	88.59 161

tfpnview11		88.74 160	88.95 161	88.50 164	93.91 115	92.43 153	91.70 183	88.90 141	90.93 108	78.90 201	84.93 199	78.24 206	81.71 168	94.32 110	94.60 97	95.86 111	87.23 174

ADS-MVSNet		79.11 215	79.38 211	78.80 218	81.90 226	75.59 224	84.36 225	83.69 198	87.31 155	76.76 212	87.58 182	76.90 209	68.55 216	78.70 225	75.56 221	77.53 225	74.07 222

gg-mvs-nofinetune		88.32 164	88.81 164	87.75 177	93.07 150	89.37 180	89.06 207	95.94 7	95.29 21	87.15 156	97.38 58	76.38 210	68.05 217	91.04 170	89.10 180	93.24 176	83.10 192

CHOSEN 280x420		79.24 213	78.26 217	80.38 209	79.60 228	68.80 235	89.32 206	75.38 220	77.25 219	78.02 210	75.57 225	76.17 211	81.19 170	88.61 189	81.39 206	78.79 224	80.03 204

tfpn1000		88.13 172	88.68 168	87.49 180	93.94 113	92.64 150	91.50 185	88.70 145	90.12 117	74.35 216	86.74 191	75.27 212	80.14 174	94.16 118	94.66 96	96.33 90	87.16 175

MDTV_nov1_ep13		82.33 199	79.66 209	85.45 193	88.83 205	83.88 203	90.09 200	81.98 207	79.07 212	88.82 145	88.70 169	73.77 213	78.41 190	80.29 223	76.08 220	84.56 209	75.83 215

test-mter		78.71 218	78.35 216	79.12 217	84.03 223	76.58 222	88.51 210	59.06 231	71.06 229	78.87 204	83.73 207	71.83 214	76.44 203	83.41 216	80.61 208	87.79 199	81.24 196

test-LLR		80.62 209	77.20 225	84.62 201	93.99 106	75.11 226	87.04 214	87.32 166	70.11 231	78.59 208	83.17 208	71.60 215	73.88 209	82.32 217	79.20 213	86.91 202	78.87 210

TESTMET0.1,1		77.47 223	77.20 225	77.78 220	81.94 225	75.11 226	87.04 214	58.33 233	70.11 231	78.59 208	83.17 208	71.60 215	73.88 209	82.32 217	79.20 213	86.91 202	78.87 210

PatchmatchNet		82.44 198	78.69 214	86.83 185	89.81 197	81.55 212	90.78 192	87.27 168	82.39 188	88.85 144	88.31 174	70.96 217	81.90 161	78.58 226	74.33 226	82.35 221	74.69 218
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
DeepMVS_CX						47.68 238	53.20 238	19.21 234	63.24 235	26.96 239	66.50 234	69.82 218	66.91 218	64.27 234		54.91 236	72.72 225

CostFormer		82.15 201	79.54 210	85.20 196	88.92 204	85.70 195	90.87 191	86.26 181	79.19 211	83.87 175	87.89 180	69.20 219	76.62 202	77.50 229	75.28 222	84.69 208	82.02 194

EPMVS		79.26 212	78.20 219	80.49 208	87.04 217	78.86 217	86.08 223	83.51 200	82.63 187	73.94 217	89.59 162	68.67 220	72.03 212	78.17 227	75.08 223	80.37 223	74.37 220

CR-MVSNet		85.32 188	81.58 204	89.69 148	90.36 194	84.79 199	86.72 218	92.22 56	75.38 223	90.73 124	90.41 156	67.88 221	84.86 139	83.76 213	85.74 194	93.24 176	83.14 190

tpmrst		78.81 217	76.18 227	81.87 206	88.56 207	77.45 221	86.74 217	81.52 209	80.08 203	83.48 177	90.84 151	66.88 222	74.54 206	73.04 232	71.02 228	76.38 228	73.95 223

MVSTER		84.79 190	83.79 195	85.96 191	89.14 203	89.80 178	89.39 205	82.99 204	74.16 227	82.78 181	85.97 195	66.81 223	76.84 200	90.77 172	88.83 183	94.66 150	90.19 141

PatchT		83.44 195	81.10 206	86.18 190	77.92 231	82.58 209	89.87 201	87.39 165	75.88 222	90.73 124	89.86 161	66.71 224	84.86 139	83.76 213	85.74 194	86.33 205	83.14 190

FMVSNet5		79.08 216	78.83 213	79.38 214	87.52 215	86.78 191	87.64 212	78.15 214	69.54 233	70.64 224	65.97 235	65.44 225	63.87 220	90.17 176	90.46 165	88.48 195	83.45 189

tpm		81.58 204	78.84 212	84.79 199	91.11 188	79.50 215	89.79 202	83.75 197	79.30 209	92.05 101	90.98 147	64.78 226	74.54 206	80.50 222	76.67 219	77.49 226	80.15 203

tpmp4_e23		82.16 200	78.26 217	86.70 187	89.92 196	84.82 198	91.17 187	89.95 108	81.21 191	87.10 157	81.91 212	64.01 227	77.88 192	79.89 224	74.99 224	84.18 213	81.00 199

tpm cat1		80.03 210	75.93 228	84.81 198	89.31 201	83.26 207	88.86 208	86.55 178	79.24 210	86.10 162	84.22 204	63.62 228	77.37 198	73.43 231	70.88 229	80.67 222	76.87 213

dps		81.42 208	77.88 222	85.56 192	87.67 212	85.17 197	88.37 211	87.46 163	74.37 226	84.55 171	86.80 190	62.18 229	80.20 173	81.13 221	77.52 216	85.10 206	77.98 212

RPMNet		83.42 196	78.40 215	89.28 154	89.79 198	84.79 199	90.64 195	92.11 63	75.38 223	87.10 157	79.80 217	61.99 230	82.79 154	81.88 219	82.07 204	93.23 178	82.87 193

conf0.01		85.72 186	83.49 197	88.32 169	92.11 174	93.35 140	93.36 149	88.02 154	80.90 193	80.51 196	74.83 226	59.86 231	79.65 178	93.80 122	94.76 90	96.29 92	86.94 176

thresconf0.02		84.34 193	82.02 202	87.06 182	92.23 171	90.93 170	91.05 189	86.43 179	88.83 140	77.65 211	73.93 228	55.81 232	79.68 177	90.62 173	90.28 167	95.30 123	83.73 188

tfpn		87.65 177	85.66 189	89.96 142	94.36 95	93.94 124	93.85 142	89.02 134	88.71 141	82.78 181	83.79 206	53.79 233	83.43 149	95.35 84	94.54 99	96.35 87	89.51 151

testpf		72.68 229	66.81 232	79.53 211	86.52 218	73.89 229	83.56 226	88.74 144	58.70 236	79.68 200	71.31 232	53.64 234	62.23 221	68.68 233	66.64 232	76.46 227	74.82 217

test2356		72.95 228	71.24 231	74.95 225	84.89 222	75.49 225	82.67 228	75.38 220	68.02 234	68.65 229	74.40 227	52.81 235	55.61 230	81.50 220	79.80 211	82.50 219	66.70 231

conf0.002		84.82 189	81.84 203	88.30 170	92.05 176	93.28 142	93.36 149	88.00 157	80.90 193	80.48 197	73.43 230	52.48 236	79.65 178	93.72 123	92.82 124	96.28 93	86.22 180

DWT-MVSNet_training		79.22 214	73.99 230	85.33 194	88.57 206	84.41 201	90.56 196	80.96 212	73.90 228	85.72 165	75.62 224	50.09 237	81.30 169	76.91 230	77.02 218	84.88 207	79.97 205

gm-plane-assit		86.15 184	82.51 200	90.40 134	95.81 65	92.29 156	97.99 31	84.66 194	92.15 75	93.15 79	97.84 38	44.65 238	78.60 185	88.02 196	85.95 193	92.20 182	76.69 214

1111		76.85 224	78.03 220	75.46 224	94.16 99	78.29 219	86.40 220	89.12 131	87.23 158	61.26 232	95.15 99	44.14 239	51.46 231	86.04 206	81.00 207	70.40 234	74.37 220

.test1245		60.07 231	56.75 233	63.93 231	94.16 99	78.29 219	86.40 220	89.12 131	87.23 158	61.26 232	95.15 99	44.14 239	51.46 231	86.04 206	2.51 235	1.21 239	3.92 236

tmp_tt				28.44 233	36.05 237	15.86 239	21.29 239	6.40 235	54.52 237	51.96 236	50.37 236	38.68 241	9.55 235	61.75 235	59.66 233	45.36 237

testmvs		2.38 233	3.35 234	1.26 236	0.83 238	0.96 241	1.53 241	0.83 236	3.59 238	1.63 243	6.03 237	2.93 242	1.55 237	3.49 236	2.51 235	1.21 239	3.92 236

test123		2.16 234	2.82 235	1.41 235	0.62 239	1.18 240	1.53 241	0.82 237	2.78 239	2.27 242	4.18 238	1.98 243	1.64 236	2.58 237	3.01 234	1.56 238	4.00 235

sosnet-low-res		0.00 235	0.00 236	0.00 237	0.00 241	0.00 242	0.00 243	0.00 239	0.00 240	0.00 244	0.00 239	0.00 244	0.00 239	0.00 238	0.00 237	0.00 241	0.00 238

sosnet		0.00 235	0.00 236	0.00 237	0.00 241	0.00 242	0.00 243	0.00 239	0.00 240	0.00 244	0.00 239	0.00 244	0.00 239	0.00 238	0.00 237	0.00 241	0.00 238

Patchmatch-RL test							8.96 240

NP-MVS									85.48 171

Patchmtry						83.74 204	86.72 218	92.22 56		90.73 124

High-resolution multi-view benchmark