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.

HFP-MVS		98.48 7	98.62 8	98.32 9	99.39 15	99.33 15	99.27 9	97.42 15	98.27 5	95.25 21	98.34 9	98.83 22	99.08 1	98.26 25	98.08 21	99.48 22	99.26 27

ACMMPR		98.40 10	98.49 10	98.28 11	99.41 11	99.40 8	99.36 3	97.35 18	98.30 4	95.02 23	97.79 15	98.39 32	99.04 2	98.26 25	98.10 19	99.50 20	99.22 33

SD-MVS		98.52 5	98.77 5	98.23 13	98.15 46	99.26 20	98.79 25	97.59 12	98.52 1	96.25 12	97.99 13	99.75 4	99.01 3	98.27 24	97.97 25	99.59 4	99.63 1

CPTT-MVS		97.78 23	97.54 30	98.05 19	98.91 31	99.05 31	99.00 19	96.96 29	97.14 38	95.92 15	95.50 37	98.78 24	98.99 4	97.20 54	96.07 79	98.54 175	99.04 57

MSLP-MVS++		98.04 20	97.93 29	98.18 14	99.10 24	99.09 30	98.34 34	96.99 28	97.54 26	96.60 10	94.82 44	98.45 31	98.89 5	97.46 49	98.77 4	99.17 94	99.37 14

TSAR-MVS + MP.		98.49 6	98.78 4	98.15 17	98.14 47	99.17 27	99.34 4	97.18 25	98.44 3	95.72 17	97.84 14	99.28 10	98.87 6	99.05 1	98.05 22	99.66 1	99.60 3

zzz-MVS		98.43 9	98.31 20	98.57 2	99.48 5	99.40 8	99.32 7	97.62 10	97.70 17	96.67 7	96.59 29	99.09 17	98.86 7	98.65 11	97.56 38	99.45 31	99.17 42

APDe-MVS		98.87 1	98.96 1	98.77 1	99.58 2	99.53 3	99.44 1	97.81 1	98.22 7	97.33 2	98.70 3	99.33 8	98.86 7	98.96 3	98.40 11	99.63 3	99.57 5

SMA-MVS		98.66 3	98.89 3	98.39 7	99.60 1	99.41 7	99.00 19	97.63 9	97.78 14	95.83 16	98.33 10	99.83 1	98.85 9	98.93 5	98.56 6	99.41 48	99.40 12

CNVR-MVS		98.47 8	98.46 13	98.48 5	99.40 12	99.05 31	99.02 18	97.54 13	97.73 15	96.65 8	97.20 26	99.13 15	98.85 9	98.91 6	98.10 19	99.41 48	99.08 49

PGM-MVS		97.81 22	98.11 25	97.46 27	99.55 3	99.34 14	99.32 7	94.51 41	96.21 60	93.07 35	98.05 12	97.95 37	98.82 11	98.22 28	97.89 30	99.48 22	99.09 48

CP-MVS		98.32 14	98.34 18	98.29 10	99.34 17	99.30 16	99.15 12	97.35 18	97.49 27	95.58 19	97.72 16	98.62 29	98.82 11	98.29 23	97.67 35	99.51 18	99.28 22

MCST-MVS		98.20 15	98.36 15	98.01 20	99.40 12	99.05 31	99.00 19	97.62 10	97.59 25	93.70 31	97.42 24	99.30 9	98.77 13	98.39 21	97.48 40	99.59 4	99.31 21

AdaColmap		97.53 27	96.93 40	98.24 12	99.21 20	98.77 63	98.47 32	97.34 20	96.68 47	96.52 11	95.11 42	96.12 51	98.72 14	97.19 56	96.24 75	99.17 94	98.39 113

ACMMP_Plus		98.20 15	98.49 10	97.85 23	99.50 4	99.40 8	99.26 10	97.64 8	97.47 29	92.62 44	97.59 18	99.09 17	98.71 15	98.82 10	97.86 31	99.40 51	99.19 38

DeepC-MVS_fast	96.13 1	98.13 17	98.27 22	97.97 22	99.16 23	99.03 36	99.05 17	97.24 23	98.22 7	94.17 29	95.82 34	98.07 34	98.69 16	98.83 9	98.80 2	99.52 14	99.10 46
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
HSP-MVS		98.59 4	98.65 7	98.52 4	99.44 10	99.57 1	99.34 4	97.65 7	97.36 32	96.62 9	98.49 6	99.65 6	98.67 17	98.60 12	97.44 42	99.40 51	99.46 10

HPM-MVS++		98.34 13	98.47 12	98.18 14	99.46 8	99.15 28	99.10 15	97.69 6	97.67 20	94.93 24	97.62 17	99.70 5	98.60 18	98.45 17	97.46 41	99.31 68	99.26 27

ESAPD		98.75 2	98.91 2	98.57 2	99.21 20	99.54 2	99.42 2	97.78 4	97.49 27	96.84 5	98.94 1	99.82 3	98.59 19	98.90 7	98.22 15	99.56 9	99.48 8

NCCC		98.10 18	98.05 27	98.17 16	99.38 16	99.05 31	99.00 19	97.53 14	98.04 10	95.12 22	94.80 45	99.18 13	98.58 20	98.49 15	97.78 33	99.39 53	98.98 65

MP-MVS		98.09 19	98.30 21	97.84 24	99.34 17	99.19 26	99.23 11	97.40 16	97.09 40	93.03 38	97.58 19	98.85 21	98.57 21	98.44 19	97.69 34	99.48 22	99.23 31
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
X-MVS		97.84 21	98.19 24	97.42 28	99.40 12	99.35 11	99.06 16	97.25 22	97.38 31	90.85 55	96.06 33	98.72 25	98.53 22	98.41 20	98.15 18	99.46 27	99.28 22

APD-MVS		98.36 12	98.32 19	98.41 6	99.47 6	99.26 20	99.12 13	97.77 5	96.73 45	96.12 13	97.27 25	98.88 20	98.46 23	98.47 16	98.39 12	99.52 14	99.22 33
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
v1.0		90.94 131	84.57 218	98.39 7	99.46 8	99.50 5	99.11 14	97.80 2	97.20 36	96.06 14	98.56 4	99.83 1	98.43 24	98.84 8	98.03 24	99.45 31	0.00 246

train_agg		97.65 26	98.06 26	97.18 31	98.94 29	98.91 56	98.98 23	97.07 27	96.71 46	90.66 60	97.43 23	99.08 19	98.20 25	97.96 37	97.14 50	99.22 86	99.19 38

DeepC-MVS	94.87 4	96.76 42	96.50 46	97.05 33	98.21 45	99.28 18	98.67 26	97.38 17	97.31 33	90.36 66	89.19 99	93.58 62	98.19 26	98.31 22	98.50 7	99.51 18	99.36 15
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
SteuartSystems-ACMMP		98.38 11	98.71 6	97.99 21	99.34 17	99.46 6	99.34 4	97.33 21	97.31 33	94.25 27	98.06 11	99.17 14	98.13 27	98.98 2	98.46 9	99.55 10	99.54 6
Skip Steuart: Steuart Systems R&D Blog.
CSCG		97.44 29	97.18 36	97.75 25	99.47 6	99.52 4	98.55 30	95.41 36	97.69 19	95.72 17	94.29 49	95.53 55	98.10 28	96.20 104	97.38 45	99.24 79	99.62 2

3Dnovator+	93.91 7	97.23 32	97.22 33	97.24 30	98.89 32	98.85 60	98.26 35	93.25 53	97.99 11	95.56 20	90.01 95	98.03 36	98.05 29	97.91 38	98.43 10	99.44 41	99.35 16

TSAR-MVS + GP.		97.45 28	98.36 15	96.39 39	95.56 78	98.93 50	97.74 45	93.31 50	97.61 23	94.24 28	98.44 8	99.19 12	98.03 30	97.60 45	97.41 44	99.44 41	99.33 18

PLC	94.95 3	97.37 30	96.77 43	98.07 18	98.97 28	98.21 93	97.94 42	96.85 31	97.66 21	97.58 1	93.33 56	96.84 43	98.01 31	97.13 58	96.20 78	99.09 106	98.01 131
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
3Dnovator	93.79 8	97.08 34	97.20 34	96.95 35	99.09 25	99.03 36	98.20 36	93.33 49	97.99 11	93.82 30	90.61 89	96.80 44	97.82 32	97.90 39	98.78 3	99.47 25	99.26 27

LS3D		95.46 51	95.14 67	95.84 47	97.91 50	98.90 58	98.58 29	97.79 3	97.07 41	83.65 118	88.71 102	88.64 93	97.82 32	97.49 48	97.42 43	99.26 78	97.72 154

CNLPA		96.90 38	96.28 49	97.64 26	98.56 39	98.63 76	96.85 60	96.60 32	97.73 15	97.08 4	89.78 97	96.28 50	97.80 34	96.73 75	96.63 61	98.94 121	98.14 127

ACMMP		97.37 30	97.48 32	97.25 29	98.88 33	99.28 18	98.47 32	96.86 30	97.04 42	92.15 46	97.57 20	96.05 53	97.67 35	97.27 52	95.99 84	99.46 27	99.14 45
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
canonicalmvs		95.25 57	95.45 61	95.00 62	95.27 93	98.72 67	96.89 58	89.82 105	96.51 49	90.84 58	93.72 52	86.01 107	97.66 36	95.78 117	97.94 27	99.54 12	99.50 7

QAPM		96.78 41	97.14 37	96.36 40	99.05 26	99.14 29	98.02 39	93.26 51	97.27 35	90.84 58	91.16 81	97.31 39	97.64 37	97.70 43	98.20 16	99.33 63	99.18 41

TAPA-MVS	94.18 5	96.38 43	96.49 47	96.25 41	98.26 44	98.66 71	98.00 40	94.96 39	97.17 37	89.48 76	92.91 61	96.35 47	97.53 38	96.59 82	95.90 87	99.28 72	97.82 144
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
PHI-MVS		97.78 23	98.44 14	97.02 34	98.73 34	99.25 22	98.11 37	95.54 35	96.66 48	92.79 41	98.52 5	99.38 7	97.50 39	97.84 40	98.39 12	99.45 31	99.03 58

OMC-MVS		97.00 36	96.92 41	97.09 32	98.69 35	98.66 71	97.85 43	95.02 38	98.09 9	94.47 25	93.15 58	96.90 41	97.38 40	97.16 57	96.82 59	99.13 101	97.65 155

MAR-MVS		95.50 49	95.60 56	95.39 55	98.67 36	98.18 95	95.89 93	89.81 106	94.55 99	91.97 48	92.99 59	90.21 78	97.30 41	96.79 70	97.49 39	98.72 160	98.99 63
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
abl_6				96.82 36	98.60 38	98.74 64	97.74 45	93.73 45	96.25 58	94.37 26	94.55 48	98.60 30	97.25 42			99.27 74	98.61 92

MVS_111021_LR		97.16 33	98.01 28	96.16 43	98.47 40	98.98 41	96.94 57	93.89 44	97.64 22	91.44 50	98.89 2	96.41 46	97.20 43	98.02 36	97.29 49	99.04 115	98.85 79

CDPH-MVS		96.84 39	97.49 31	96.09 44	98.92 30	98.85 60	98.61 27	95.09 37	96.00 67	87.29 103	95.45 39	97.42 38	97.16 44	97.83 41	97.94 27	99.44 41	98.92 70

thres400		93.56 95	92.43 117	94.87 69	95.40 83	98.91 56	96.70 70	92.38 64	92.93 124	88.19 93	86.69 120	77.35 153	97.13 45	96.75 74	95.85 91	99.42 47	98.56 95

thres200		93.62 92	92.54 111	94.88 66	95.36 84	98.93 50	96.75 69	92.31 65	92.84 125	88.28 89	86.99 115	77.81 152	97.13 45	96.82 65	95.92 85	99.45 31	98.49 105

tfpn111		94.05 80	93.34 105	94.88 66	95.33 85	98.94 44	96.82 61	92.31 65	92.63 127	88.26 90	92.61 65	78.01 148	97.12 47	96.82 65	95.85 91	99.45 31	98.56 95

conf0.01		93.33 103	91.89 130	95.00 62	95.32 89	98.94 44	96.82 61	92.41 63	92.63 127	88.91 84	88.02 110	72.75 189	97.12 47	96.78 71	95.85 91	99.44 41	98.27 120

conf0.002		93.20 106	91.63 133	95.02 60	95.31 90	98.94 44	96.82 61	92.43 62	92.63 127	88.99 83	88.16 109	70.49 208	97.12 47	96.77 72	96.30 67	99.44 41	98.16 126

conf200view11		93.64 89	92.57 109	94.88 66	95.33 85	98.94 44	96.82 61	92.31 65	92.63 127	88.26 90	87.21 112	78.01 148	97.12 47	96.82 65	95.85 91	99.45 31	98.56 95

tfpn200view9		93.64 89	92.57 109	94.89 65	95.33 85	98.94 44	96.82 61	92.31 65	92.63 127	88.29 87	87.21 112	78.01 148	97.12 47	96.82 65	95.85 91	99.45 31	98.56 95

view600		93.50 99	92.39 120	94.80 72	95.41 82	98.93 50	96.60 73	92.30 70	93.09 121	87.96 94	86.67 121	76.97 155	97.12 47	96.83 64	95.64 101	99.43 46	98.62 91

thres600view7		93.49 100	92.37 121	94.79 73	95.42 79	98.93 50	96.58 75	92.31 65	93.04 122	87.88 95	86.62 122	76.94 156	97.09 53	96.82 65	95.63 102	99.45 31	98.63 90

view800		93.45 102	92.37 121	94.71 74	95.42 79	98.92 54	96.51 78	92.19 73	93.14 120	87.62 97	86.72 119	76.54 159	97.08 54	96.86 63	95.74 98	99.45 31	98.70 88

tfpn		92.91 108	91.44 137	94.63 76	95.42 79	98.92 54	96.41 81	92.10 74	93.19 118	87.34 102	86.85 116	69.20 216	97.01 55	96.88 62	96.28 71	99.47 25	98.75 87

TSAR-MVS + COLMAP		94.79 64	94.51 79	95.11 57	96.50 65	97.54 103	97.99 41	94.54 40	97.81 13	85.88 108	96.73 28	81.28 137	96.99 56	96.29 100	95.21 114	98.76 157	96.73 182

PCF-MVS	93.95 6	95.65 47	95.14 67	96.25 41	97.73 53	98.73 66	97.59 48	97.13 26	92.50 135	89.09 82	89.85 96	96.65 45	96.90 57	94.97 137	94.89 120	99.08 107	98.38 114
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
TSAR-MVS + ACMM		97.71 25	98.60 9	96.66 37	98.64 37	99.05 31	98.85 24	97.23 24	98.45 2	89.40 78	97.51 21	99.27 11	96.88 58	98.53 13	97.81 32	98.96 120	99.59 4

casdiffmvs1		95.62 48	96.09 52	95.08 59	95.16 98	98.67 70	96.95 56	89.70 108	97.61 23	92.42 45	94.68 47	90.01 79	96.85 59	95.45 126	97.02 52	99.37 58	99.21 35

OpenMVS	92.33 11	95.50 49	95.22 66	95.82 48	98.98 27	98.97 42	97.67 47	93.04 59	94.64 97	89.18 80	84.44 139	94.79 57	96.79 60	97.23 53	97.61 36	99.24 79	98.88 75

thres100view900		93.55 98	92.47 116	94.81 71	95.33 85	98.74 64	96.78 68	92.30 70	92.63 127	88.29 87	87.21 112	78.01 148	96.78 61	96.38 95	95.92 85	99.38 55	98.40 112

Anonymous20231211		93.49 100	92.33 123	94.84 70	94.78 113	98.00 100	96.11 87	91.85 81	94.86 95	90.91 54	74.69 184	89.18 87	96.73 62	94.82 138	95.51 107	98.67 166	99.24 30

Effi-MVS+		92.93 107	93.86 92	91.86 113	94.07 127	98.09 99	95.59 100	85.98 155	94.27 103	79.54 140	91.12 84	81.81 134	96.71 63	96.67 78	96.06 80	99.27 74	98.98 65

MVS_111021_HR		97.04 35	98.20 23	95.69 49	98.44 42	99.29 17	96.59 74	93.20 54	97.70 17	89.94 71	98.46 7	96.89 42	96.71 63	98.11 33	97.95 26	99.27 74	99.01 61

Fast-Effi-MVS+		91.87 117	92.08 126	91.62 118	92.91 142	97.21 111	94.93 114	84.60 174	93.61 113	81.49 131	83.50 145	78.95 143	96.62 65	96.55 86	96.22 76	99.16 97	98.51 103

Anonymous202405211			92.18 124		95.04 106	98.20 94	96.14 86	91.79 85	93.93 107		74.60 185	88.38 96	96.48 66	95.17 133	95.82 97	99.00 116	99.15 44

MVS_Test		94.82 62	95.66 55	93.84 89	94.79 112	98.35 86	96.49 79	89.10 118	96.12 63	87.09 104	92.58 67	90.61 75	96.48 66	96.51 91	96.89 56	99.11 104	98.54 99

ACMM	92.75 10	94.41 75	93.84 93	95.09 58	96.41 68	96.80 120	94.88 117	93.54 47	96.41 51	90.16 67	92.31 71	83.11 129	96.32 68	96.22 103	94.65 126	99.22 86	97.35 162
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
conf0.05thres1000		92.47 113	91.39 138	93.73 91	95.21 96	98.52 79	95.66 99	91.56 86	90.87 157	84.27 113	82.79 150	76.12 160	96.29 69	96.59 82	95.68 100	99.39 53	99.19 38

OPM-MVS		93.61 93	92.43 117	95.00 62	96.94 62	97.34 108	97.78 44	94.23 42	89.64 171	85.53 109	88.70 103	82.81 130	96.28 70	96.28 101	95.00 119	99.24 79	97.22 166

CANet		96.84 39	97.20 34	96.42 38	97.92 49	99.24 24	98.60 28	93.51 48	97.11 39	93.07 35	91.16 81	97.24 40	96.21 71	98.24 27	98.05 22	99.22 86	99.35 16

diffmvs1		94.84 61	95.60 56	93.97 85	94.90 109	98.40 84	96.53 77	88.81 120	97.41 30	88.52 85	92.91 61	88.92 91	96.20 72	96.37 96	96.41 65	98.83 130	98.76 85

casdiffmvs		94.87 60	95.25 65	94.43 79	94.87 111	98.52 79	95.98 90	89.42 114	96.32 54	91.05 53	93.18 57	87.33 100	96.06 73	96.11 108	96.38 66	99.23 85	98.92 70

PMMVS		94.61 69	95.56 58	93.50 96	94.30 123	96.74 124	94.91 116	89.56 111	95.58 80	87.72 96	96.15 32	92.86 65	96.06 73	95.47 124	95.02 117	98.43 183	97.09 169

CLD-MVS		94.79 64	94.36 83	95.30 56	95.21 96	97.46 105	97.23 52	92.24 72	96.43 50	91.77 49	92.69 64	84.31 118	96.06 73	95.52 122	95.03 116	99.31 68	99.06 53
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
PatchMatch-RL		94.69 68	94.41 81	95.02 60	97.63 54	98.15 97	94.50 126	91.99 78	95.32 85	91.31 51	95.47 38	83.44 124	96.02 76	96.56 85	95.23 113	98.69 165	96.67 183

Anonymous20240521		94.76 67	95.12 69	94.35 80	95.10 102	95.81 153	96.46 80	89.49 112	96.33 53	90.16 67	92.55 68	90.26 77	95.83 77	95.52 122	96.03 82	99.06 112	99.33 18

MVS_0304		96.31 44	96.91 42	95.62 50	97.21 59	99.20 25	98.55 30	93.10 56	97.04 42	89.73 73	90.30 91	96.35 47	95.71 78	98.14 30	97.93 29	99.38 55	99.40 12

HyFIR lowres test		92.03 115	91.55 135	92.58 110	97.13 60	98.72 67	94.65 122	86.54 147	93.58 115	82.56 122	67.75 227	90.47 76	95.67 79	95.87 113	95.54 106	98.91 124	98.93 69

CHOSEN 280x420		95.46 51	97.01 38	93.66 93	97.28 58	97.98 101	96.40 82	85.39 162	96.10 65	91.07 52	96.53 30	96.34 49	95.61 80	97.65 44	96.95 55	96.21 209	97.49 157

HQP-MVS		94.43 73	94.57 77	94.27 81	96.41 68	97.23 110	96.89 58	93.98 43	95.94 69	83.68 117	95.01 43	84.46 117	95.58 81	95.47 124	94.85 124	99.07 109	99.00 62

MSDG		94.82 62	93.73 96	96.09 44	98.34 43	97.43 107	97.06 53	96.05 33	95.84 73	90.56 61	86.30 129	89.10 89	95.55 82	96.13 107	95.61 104	99.00 116	95.73 193

DeepPCF-MVS	95.28 2	97.00 36	98.35 17	95.42 54	97.30 57	98.94 44	94.82 118	96.03 34	98.24 6	92.11 47	95.80 35	98.64 28	95.51 83	98.95 4	98.66 5	96.78 205	99.20 37

DI_MVS_plusplus_trai		94.01 82	93.63 98	94.44 78	94.54 118	98.26 91	97.51 49	90.63 96	95.88 71	89.34 79	80.54 160	89.36 84	95.48 84	96.33 99	96.27 72	99.17 94	98.78 84

EPP-MVSNet		95.27 56	96.18 51	94.20 82	94.88 110	98.64 74	94.97 113	90.70 95	95.34 84	89.67 75	91.66 78	93.84 60	95.42 85	97.32 51	97.00 53	99.58 6	99.47 9

diffmvs		94.16 77	94.72 74	93.52 95	94.55 117	98.32 87	96.06 89	88.85 119	96.39 52	87.54 99	92.31 71	86.19 105	95.40 86	95.39 128	95.85 91	98.71 161	98.59 94

RPSCF		94.05 80	94.00 89	94.12 83	96.20 70	96.41 134	96.61 72	91.54 87	95.83 74	89.73 73	96.94 27	92.80 66	95.35 87	91.63 201	90.44 209	95.27 221	93.94 208

LGP-MVS_train		94.12 79	94.62 76	93.53 94	96.44 67	97.54 103	97.40 51	91.84 82	94.66 96	81.09 133	95.70 36	83.36 128	95.10 88	96.36 98	95.71 99	99.32 65	99.03 58

DELS-MVS		96.06 46	96.04 53	96.07 46	97.77 51	99.25 22	98.10 38	93.26 51	94.42 100	92.79 41	88.52 106	93.48 63	95.06 89	98.51 14	98.83 1	99.45 31	99.28 22
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
ACMP	92.88 9	94.43 73	94.38 82	94.50 77	96.01 73	97.69 102	95.85 96	92.09 75	95.74 76	89.12 81	95.14 41	82.62 132	94.77 90	95.73 118	94.67 125	99.14 100	99.06 53
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
thisisatest0530		94.54 70	95.47 60	93.46 97	94.51 119	98.65 73	94.66 121	90.72 93	95.69 78	86.90 106	93.80 50	89.44 83	94.74 91	96.98 61	94.86 121	99.19 93	98.85 79

tttt0517		94.52 71	95.44 62	93.44 98	94.51 119	98.68 69	94.61 124	90.72 93	95.61 79	86.84 107	93.78 51	89.26 86	94.74 91	97.02 60	94.86 121	99.20 92	98.87 77

PVSNet_BlendedMVS		95.41 53	95.28 63	95.57 51	97.42 55	99.02 38	95.89 93	93.10 56	96.16 61	93.12 33	91.99 74	85.27 112	94.66 93	98.09 34	97.34 46	99.24 79	99.08 49

PVSNet_Blended		95.41 53	95.28 63	95.57 51	97.42 55	99.02 38	95.89 93	93.10 56	96.16 61	93.12 33	91.99 74	85.27 112	94.66 93	98.09 34	97.34 46	99.24 79	99.08 49

FC-MVSNet-train		93.85 85	93.91 90	93.78 90	94.94 108	96.79 123	94.29 129	91.13 90	93.84 111	88.26 90	90.40 90	85.23 114	94.65 95	96.54 87	95.31 111	99.38 55	99.28 22

CANet_DTU		93.92 84	96.57 45	90.83 125	95.63 76	98.39 85	96.99 55	87.38 139	96.26 57	71.97 196	96.31 31	93.02 64	94.53 96	97.38 50	96.83 58	98.49 178	97.79 146

FMVSNet1		91.54 124	90.93 143	92.26 112	90.35 165	95.27 178	95.22 110	87.16 142	91.37 151	87.62 97	75.45 173	83.84 122	94.43 97	96.52 88	96.30 67	98.82 131	97.74 153

IterMVS-LS		92.56 112	93.18 106	91.84 114	93.90 129	94.97 186	94.99 112	86.20 151	94.18 105	82.68 121	85.81 131	87.36 99	94.43 97	95.31 129	96.02 83	98.87 127	98.60 93
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
GBi-Net		93.81 86	94.18 86	93.38 99	91.34 154	95.86 149	96.22 83	88.68 121	95.23 88	90.40 62	86.39 125	91.16 70	94.40 99	96.52 88	96.30 67	99.21 89	97.79 146

test1		93.81 86	94.18 86	93.38 99	91.34 154	95.86 149	96.22 83	88.68 121	95.23 88	90.40 62	86.39 125	91.16 70	94.40 99	96.52 88	96.30 67	99.21 89	97.79 146

FMVSNet2		93.30 104	93.36 104	93.22 103	91.34 154	95.86 149	96.22 83	88.24 126	95.15 92	89.92 72	81.64 154	89.36 84	94.40 99	96.77 72	96.98 54	99.21 89	97.79 146

IS_MVSNet		95.28 55	96.43 48	93.94 86	95.30 91	99.01 40	95.90 91	91.12 91	94.13 106	87.50 100	91.23 80	94.45 59	94.17 102	98.45 17	98.50 7	99.65 2	99.23 31

FMVSNet3		93.79 88	94.17 88	93.35 101	91.21 157	95.99 142	96.62 71	88.68 121	95.23 88	90.40 62	86.39 125	91.16 70	94.11 103	95.96 110	96.67 60	99.07 109	97.79 146

CHOSEN 1792x2688		92.66 111	92.49 114	92.85 106	97.13 60	98.89 59	95.90 91	88.50 124	95.32 85	83.31 119	71.99 212	88.96 90	94.10 104	96.69 76	96.49 63	98.15 187	99.10 46

EPMVS		90.88 132	92.12 125	89.44 145	94.71 114	97.24 109	93.55 135	76.81 216	95.89 70	81.77 128	91.49 79	86.47 102	93.87 105	90.21 210	90.07 211	95.92 211	93.49 214

COLMAP_ROB	90.49 14	93.27 105	92.71 108	93.93 87	97.75 52	97.44 106	96.07 88	93.17 55	95.40 83	83.86 116	83.76 144	88.72 92	93.87 105	94.25 149	94.11 141	98.87 127	95.28 199
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
ACMH+	90.88 12	91.41 126	91.13 140	91.74 116	95.11 101	96.95 114	93.13 143	89.48 113	92.42 137	79.93 137	85.13 134	78.02 147	93.82 107	93.49 161	93.88 147	98.94 121	97.99 132

ACMH	90.77 13	91.51 125	91.63 133	91.38 119	95.62 77	96.87 117	91.76 185	89.66 109	91.58 149	78.67 142	86.73 118	78.12 146	93.77 108	94.59 140	94.54 133	98.78 151	98.98 65
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
CostFormer		90.69 133	90.48 148	90.93 123	94.18 124	96.08 141	94.03 131	78.20 212	93.47 116	89.96 70	90.97 86	80.30 139	93.72 109	87.66 223	88.75 215	95.51 217	96.12 187

MVSTER		94.89 59	95.07 70	94.68 75	94.71 114	96.68 126	97.00 54	90.57 97	95.18 91	93.05 37	95.21 40	86.41 103	93.72 109	97.59 46	95.88 89	99.00 116	98.50 104

USDC		90.69 133	90.52 147	90.88 124	94.17 125	96.43 133	95.82 97	86.76 145	93.92 108	76.27 156	86.49 124	74.30 171	93.67 111	95.04 136	93.36 156	98.61 171	94.13 206

DWT-MVSNet_training		91.30 127	89.73 151	93.13 104	94.64 116	96.87 117	94.93 114	86.17 152	94.22 104	93.18 32	89.11 100	73.28 183	93.59 112	88.00 220	90.73 207	96.26 208	95.87 190

Effi-MVS+-dtu		91.78 119	93.59 100	89.68 143	92.44 146	97.11 112	94.40 127	84.94 170	92.43 136	75.48 160	91.09 85	83.75 123	93.55 113	96.61 79	95.47 108	97.24 201	98.67 89

PatchmatchNet		90.56 135	92.49 114	88.31 163	93.83 132	96.86 119	92.42 154	76.50 220	95.96 68	78.31 143	91.96 76	89.66 82	93.48 114	90.04 212	89.20 214	95.32 219	93.73 212
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
TinyColmap		89.42 150	88.58 161	90.40 132	93.80 133	95.45 165	93.96 133	86.54 147	92.24 143	76.49 153	80.83 158	70.44 209	93.37 115	94.45 144	93.30 159	98.26 186	93.37 216

tfpn_ndepth		94.36 76	94.64 75	94.04 84	95.16 98	98.51 81	95.58 101	92.09 75	95.78 75	88.52 85	92.38 70	85.74 109	93.34 116	96.39 93	95.90 87	99.54 12	97.79 146

LTVRE_ROB	87.32 16	87.55 187	88.25 165	86.73 198	90.66 160	95.80 154	93.05 144	84.77 171	83.35 224	60.32 229	83.12 148	67.39 221	93.32 117	94.36 147	94.86 121	98.28 185	98.87 77
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
ADS-MVSNet		89.80 147	91.33 139	88.00 175	94.43 121	96.71 125	92.29 164	74.95 228	96.07 66	77.39 146	88.67 104	86.09 106	93.26 118	88.44 218	89.57 213	95.68 214	93.81 211

MDTV_nov1_ep13		91.57 123	93.18 106	89.70 141	93.39 136	96.97 113	93.53 136	80.91 204	95.70 77	81.86 127	92.40 69	89.93 80	93.25 119	91.97 199	90.80 206	95.25 222	94.46 203

UniMVSNet_NR-MVSNet		90.35 139	89.96 149	90.80 126	89.66 175	95.83 152	92.48 152	90.53 98	90.96 156	79.57 138	79.33 164	77.14 154	93.21 120	92.91 170	94.50 136	99.37 58	99.05 55

DU-MVS		89.67 149	88.84 159	90.63 129	89.26 199	95.61 158	92.48 152	89.91 103	91.22 152	79.57 138	77.72 168	71.18 205	93.21 120	92.53 174	94.57 130	99.35 60	99.05 55

tpmp4_e23		89.82 146	89.31 157	90.42 131	94.01 128	95.45 165	94.63 123	78.37 209	93.59 114	87.09 104	86.62 122	76.59 158	93.06 122	88.50 217	88.52 216	95.36 218	95.88 189

pmmvs4		90.55 136	89.91 150	91.30 120	90.26 167	94.95 187	92.73 148	87.94 132	93.44 117	85.35 110	82.28 153	76.09 162	93.02 123	93.56 159	92.26 198	98.51 177	96.77 181

thresconf0.02		93.57 94	93.84 93	93.25 102	95.03 107	98.16 96	95.80 98	92.46 61	96.12 63	83.88 115	92.61 65	80.39 138	92.83 124	96.11 108	96.21 77	99.49 21	97.28 165

tpmrst		88.86 162	89.62 152	87.97 176	94.33 122	95.98 143	92.62 150	76.36 221	94.62 98	76.94 150	85.98 130	82.80 131	92.80 125	86.90 224	87.15 225	94.77 226	93.93 209

tfpn1000		94.14 78	94.54 78	93.67 92	95.27 93	98.50 82	95.36 107	91.84 82	96.31 56	87.38 101	92.98 60	84.04 119	92.60 126	96.49 92	95.62 103	99.55 10	97.82 144

RPMNet		90.19 141	92.03 128	88.05 172	93.46 134	95.95 146	93.41 138	74.59 229	92.40 138	75.91 158	84.22 140	86.41 103	92.49 127	94.42 145	93.85 149	98.44 181	96.96 176

FMVSNet5		90.36 138	90.93 143	89.70 141	87.99 216	92.25 213	92.03 179	83.51 181	92.20 144	84.13 114	85.59 132	86.48 101	92.43 128	94.61 139	94.52 134	98.13 188	90.85 223

dps		90.11 144	89.37 156	90.98 122	93.89 130	96.21 138	93.49 137	77.61 214	91.95 146	92.74 43	88.85 101	78.77 145	92.37 129	87.71 222	87.71 221	95.80 212	94.38 204

Baseline_NR-MVSNet		89.27 154	88.01 169	90.73 128	89.26 199	93.71 208	92.71 149	89.78 107	90.73 159	81.28 132	73.53 204	72.85 184	92.30 130	92.53 174	93.84 150	99.07 109	98.88 75

tfpnview11		93.63 91	94.42 80	92.71 107	95.08 103	98.26 91	95.58 101	92.06 77	96.32 54	81.88 124	93.44 53	83.43 125	92.14 131	96.58 84	95.88 89	99.52 14	97.07 173

CR-MVSNet		90.16 142	91.96 129	88.06 171	93.32 137	95.95 146	93.36 139	75.99 223	92.40 138	75.19 165	83.18 147	85.37 111	92.05 132	95.21 131	94.56 131	98.47 180	97.08 171

PatchT		89.13 157	91.71 131	86.11 207	92.92 141	95.59 160	83.64 223	75.09 227	91.87 147	75.19 165	82.63 151	85.06 116	92.05 132	95.21 131	94.56 131	97.76 196	97.08 171

tfpn_n400		93.56 95	94.36 83	92.63 108	95.07 104	98.28 88	95.50 105	91.98 79	95.48 81	81.88 124	93.44 53	83.43 125	92.01 134	96.60 80	96.27 72	99.34 61	97.04 174

tfpnconf		93.56 95	94.36 83	92.63 108	95.07 104	98.28 88	95.50 105	91.98 79	95.48 81	81.88 124	93.44 53	83.43 125	92.01 134	96.60 80	96.27 72	99.34 61	97.04 174

v2v482		88.25 169	87.71 180	88.88 152	89.23 203	95.28 176	92.10 175	87.89 133	88.69 185	73.31 190	75.32 174	71.64 201	91.89 136	92.10 193	92.92 166	98.86 129	97.99 132

tfpnnormal		88.50 163	87.01 199	90.23 133	91.36 153	95.78 155	92.74 147	90.09 101	83.65 223	76.33 155	71.46 217	69.58 214	91.84 137	95.54 121	94.02 144	99.06 112	99.03 58

TranMVSNet+NR-MVSNet		89.23 155	88.48 163	90.11 139	89.07 205	95.25 179	92.91 146	90.43 99	90.31 164	77.10 149	76.62 171	71.57 203	91.83 138	92.12 187	94.59 129	99.32 65	98.92 70

EPNet		96.27 45	96.97 39	95.46 53	98.47 40	98.28 88	97.41 50	93.67 46	95.86 72	92.86 40	97.51 21	93.79 61	91.76 139	97.03 59	97.03 51	98.61 171	99.28 22
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
Fast-Effi-MVS+-dtu		91.19 128	93.64 97	88.33 162	92.19 149	96.46 132	93.99 132	81.52 203	92.59 133	71.82 197	92.17 73	85.54 110	91.68 140	95.73 118	94.64 127	98.80 138	98.34 115

tpm		87.95 176	89.44 155	86.21 205	92.53 145	94.62 199	91.40 188	76.36 221	91.46 150	69.80 212	87.43 111	75.14 166	91.55 141	89.85 215	90.60 208	95.61 215	96.96 176

tpm cat1		88.90 160	87.78 179	90.22 134	93.88 131	95.39 174	93.79 134	78.11 213	92.55 134	89.43 77	81.31 156	79.84 141	91.40 142	84.95 228	86.34 230	94.68 229	94.09 207

v7		88.18 171	88.01 169	88.39 160	89.45 184	95.14 182	92.36 156	85.37 163	89.29 175	72.94 193	73.98 199	72.77 187	91.38 143	93.59 155	92.87 168	98.82 131	98.42 109

v10		88.00 175	87.96 174	88.05 172	89.44 185	94.68 195	92.36 156	83.35 184	89.37 174	72.96 191	73.98 199	72.79 186	91.35 144	93.59 155	92.88 167	98.81 135	98.42 109

v1192		87.51 188	87.31 189	87.74 182	89.04 206	94.87 193	92.07 176	85.03 168	88.49 187	70.32 206	72.65 209	70.35 210	91.21 145	93.59 155	92.80 172	98.78 151	98.42 109

UniMVSNet (Re)		90.03 145	89.61 153	90.51 130	89.97 171	96.12 140	92.32 160	89.26 115	90.99 155	80.95 134	78.25 167	75.08 168	91.14 146	93.78 154	93.87 148	99.41 48	99.21 35

v1921920		87.31 197	87.13 196	87.52 191	88.87 209	94.72 194	91.96 183	84.59 175	88.28 195	69.86 211	72.50 210	70.03 213	91.10 147	93.33 163	92.61 190	98.71 161	98.44 106

v11		87.58 186	87.50 188	87.67 185	89.34 193	91.91 221	92.22 173	81.63 200	89.01 179	72.95 192	74.11 197	72.51 197	91.08 148	94.01 153	93.00 164	98.77 155	97.93 135

v1144		87.92 180	87.79 178	88.07 169	89.27 198	95.15 181	92.17 174	85.62 159	88.52 186	71.52 198	73.80 202	72.40 198	91.06 149	93.54 160	92.80 172	98.81 135	98.33 116

MIMVSNet		88.99 159	91.07 141	86.57 200	86.78 224	95.62 157	91.20 193	75.40 226	90.65 161	76.57 152	84.05 141	82.44 133	91.01 150	95.84 114	95.38 110	98.48 179	93.50 213

test-LLR		91.62 122	93.56 101	89.35 147	93.31 138	96.57 129	92.02 180	87.06 143	92.34 141	75.05 168	90.20 92	88.64 93	90.93 151	96.19 105	94.07 142	97.75 197	96.90 179

TESTMET0.1,1		91.07 129	93.56 101	88.17 166	90.43 162	96.57 129	92.02 180	82.83 188	92.34 141	75.05 168	90.20 92	88.64 93	90.93 151	96.19 105	94.07 142	97.75 197	96.90 179

SixPastTwentyTwo		88.37 167	89.47 154	87.08 195	90.01 170	95.93 148	87.41 212	85.32 164	90.26 166	70.26 207	86.34 128	71.95 199	90.93 151	92.89 171	91.72 202	98.55 174	97.22 166

test-mter		90.95 130	93.54 103	87.93 177	90.28 166	96.80 120	91.44 187	82.68 190	92.15 145	74.37 176	89.57 98	88.23 97	90.88 154	96.37 96	94.31 138	97.93 194	97.37 161

PVSNet_Blended_VisFu		94.77 66	95.54 59	93.87 88	96.48 66	98.97 42	94.33 128	91.84 82	94.93 94	90.37 65	85.04 135	94.99 56	90.87 155	98.12 32	97.30 48	99.30 70	99.45 11

CP-MVSNet		87.89 181	87.27 190	88.62 158	89.30 195	95.06 183	90.60 198	85.78 157	87.43 204	75.98 157	74.60 185	68.14 220	90.76 156	93.07 168	93.60 153	99.30 70	98.98 65

v144192		87.40 193	87.20 192	87.64 186	88.89 207	94.88 192	91.65 186	84.70 173	87.80 199	71.17 203	73.20 207	70.91 206	90.75 157	92.69 172	92.49 191	98.71 161	98.43 107

pmmvs5		87.83 183	88.09 167	87.51 192	89.59 182	95.48 163	89.75 206	84.73 172	86.07 216	71.44 199	80.57 159	70.09 212	90.74 158	94.47 143	92.87 168	98.82 131	97.10 168

v18		87.93 177	87.61 185	88.31 163	89.74 173	92.04 214	92.59 151	82.71 189	89.70 168	75.32 163	75.23 175	73.55 177	90.74 158	92.11 190	92.77 178	98.78 151	97.87 140

v16		87.87 182	87.60 186	88.19 165	89.70 174	92.01 216	92.37 155	82.54 192	89.67 170	75.00 170	75.02 179	73.65 175	90.73 160	92.14 186	92.80 172	98.77 155	97.90 137

v17		87.83 183	87.56 187	88.13 167	89.65 176	92.02 215	92.34 159	82.55 191	89.38 173	74.76 171	75.14 176	73.59 176	90.70 161	92.15 185	92.78 176	98.78 151	97.89 138

v1neww		88.41 165	88.00 172	88.89 150	89.61 179	95.44 168	92.31 161	87.65 135	89.09 176	74.30 177	75.02 179	73.42 181	90.68 162	92.12 187	92.77 178	98.79 144	98.18 123

v7new		88.41 165	88.00 172	88.89 150	89.61 179	95.44 168	92.31 161	87.65 135	89.09 176	74.30 177	75.02 179	73.42 181	90.68 162	92.12 187	92.77 178	98.79 144	98.18 123

v8		88.21 170	87.94 176	88.51 159	89.62 177	95.01 185	92.31 161	84.99 169	88.94 180	74.70 172	75.03 178	73.51 178	90.67 164	92.11 190	92.74 184	98.80 138	98.24 121

v1240		86.89 199	86.75 203	87.06 196	88.75 211	94.65 197	91.30 192	84.05 178	87.49 203	68.94 215	71.96 213	68.86 219	90.65 165	93.33 163	92.72 186	98.67 166	98.24 121

gm-plane-assit		83.26 218	85.29 213	80.89 219	89.52 183	89.89 228	70.26 236	78.24 211	77.11 233	58.01 233	74.16 195	66.90 223	90.63 166	97.20 54	96.05 81	98.66 168	95.68 194

v6		88.43 164	88.01 169	88.92 149	89.60 181	95.43 170	92.36 156	87.66 134	89.07 178	74.50 174	75.06 177	73.47 179	90.59 167	92.11 190	92.76 182	98.79 144	98.18 123

MS-PatchMatch		91.82 118	92.51 112	91.02 121	95.83 75	96.88 115	95.05 111	84.55 177	93.85 110	82.01 123	82.51 152	91.71 68	90.52 168	95.07 135	93.03 163	98.13 188	94.52 201

v1141		88.17 172	87.69 181	88.74 155	89.44 185	95.41 171	92.25 169	87.98 129	88.38 190	73.54 188	74.43 189	72.71 193	90.45 169	92.08 194	92.72 186	98.79 144	98.09 128

V14		87.47 190	87.19 193	87.80 180	89.37 192	91.95 218	92.25 169	82.12 196	88.39 189	73.83 181	74.31 192	72.84 185	90.44 170	92.20 182	92.78 176	98.80 138	97.84 142

divwei89l23v2f112		88.17 172	87.69 181	88.74 155	89.44 185	95.41 171	92.26 167	87.97 131	88.29 194	73.57 187	74.45 188	72.75 189	90.42 171	92.08 194	92.72 186	98.81 135	98.09 128

CDS-MVSNet		92.77 109	93.60 99	91.80 115	92.63 144	96.80 120	95.24 109	89.14 117	90.30 165	84.58 112	86.76 117	90.65 74	90.42 171	95.89 112	96.49 63	98.79 144	98.32 118
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
TAMVS		90.54 137	90.87 145	90.16 135	91.48 152	96.61 128	93.26 141	86.08 153	87.71 200	81.66 130	83.11 149	84.04 119	90.42 171	94.54 141	94.60 128	98.04 192	95.48 197

v15		87.46 191	87.16 194	87.81 179	89.41 190	91.96 217	92.26 167	82.28 195	88.42 188	73.72 183	74.29 194	72.73 192	90.41 174	92.17 184	92.76 182	98.79 144	97.83 143

V42		88.31 168	87.95 175	88.73 157	89.44 185	95.34 175	92.23 171	87.21 141	88.83 182	74.49 175	74.89 183	73.43 180	90.41 174	92.08 194	92.77 178	98.60 173	98.33 116

V9		87.41 192	87.15 195	87.72 183	89.33 194	91.93 219	92.23 171	82.02 197	88.35 191	73.59 186	74.13 196	72.77 187	90.37 176	92.21 181	92.80 172	98.79 144	97.86 141

v1		88.17 172	87.66 183	88.77 154	89.44 185	95.40 173	92.29 164	87.98 129	88.21 197	73.75 182	74.41 191	72.75 189	90.36 177	92.07 197	92.71 189	98.80 138	98.09 128

v13		87.34 195	87.11 198	87.62 187	89.30 195	91.91 221	92.04 178	81.86 199	88.35 191	73.36 189	73.88 201	72.69 194	90.34 178	92.23 179	92.82 170	98.80 138	97.88 139

v12		87.38 194	87.13 196	87.68 184	89.30 195	91.92 220	92.01 182	81.94 198	88.35 191	73.69 184	74.10 198	72.57 196	90.33 179	92.23 179	92.82 170	98.80 138	97.91 136

anonymousdsp		88.90 160	91.00 142	86.44 203	88.74 212	95.97 144	90.40 200	82.86 187	88.77 184	67.33 217	81.18 157	81.44 136	90.22 180	96.23 102	94.27 139	99.12 103	99.16 43

PS-CasMVS		87.33 196	86.68 204	88.10 168	89.22 204	94.93 188	90.35 201	85.70 158	86.44 210	74.01 179	73.43 205	66.59 226	90.04 181	92.92 169	93.52 154	99.28 72	98.91 73

IterMVS		90.20 140	92.43 117	87.61 188	92.82 143	94.31 204	94.11 130	81.54 202	92.97 123	69.90 210	84.71 137	88.16 98	89.96 182	95.25 130	94.17 140	97.31 200	97.46 158
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
gg-mvs-nofinetune		86.17 206	88.57 162	83.36 215	93.44 135	98.15 97	96.58 75	72.05 234	74.12 235	49.23 240	64.81 230	90.85 73	89.90 183	97.83 41	96.84 57	98.97 119	97.41 160

GA-MVS		89.28 153	90.75 146	87.57 189	91.77 151	96.48 131	92.29 164	87.58 137	90.61 162	65.77 220	84.48 138	76.84 157	89.46 184	95.84 114	93.68 152	98.52 176	97.34 163

PEN-MVS		87.22 198	86.50 208	88.07 169	88.88 208	94.44 201	90.99 195	86.21 149	86.53 209	73.66 185	74.97 182	66.56 227	89.42 185	91.20 203	93.48 155	99.24 79	98.31 119

NR-MVSNet		89.34 152	88.66 160	90.13 138	90.40 163	95.61 158	93.04 145	89.91 103	91.22 152	78.96 141	77.72 168	68.90 218	89.16 186	94.24 150	93.95 145	99.32 65	98.99 63

pm-mvs1		89.19 156	89.02 158	89.38 146	90.40 163	95.74 156	92.05 177	88.10 128	86.13 214	77.70 144	73.72 203	79.44 142	88.97 187	95.81 116	94.51 135	99.08 107	97.78 152

MVS-HIRNet		85.36 210	86.89 200	83.57 214	90.13 168	94.51 200	83.57 224	72.61 232	88.27 196	71.22 201	68.97 223	81.81 134	88.91 188	93.08 167	91.94 199	94.97 225	89.64 227

PM-MVS		84.72 213	84.47 219	85.03 210	84.67 226	91.57 224	86.27 218	82.31 194	87.65 201	70.62 205	76.54 172	56.41 238	88.75 189	92.59 173	89.85 212	97.54 199	96.66 184

v52		86.57 202	86.63 205	86.50 201	87.47 221	94.89 191	89.90 203	83.39 182	86.36 211	71.17 203	71.53 215	71.65 200	88.34 190	91.14 204	92.32 194	98.74 159	98.52 101

V4		86.56 203	86.61 206	86.50 201	87.49 220	94.90 190	89.87 204	83.39 182	86.25 212	71.20 202	71.57 214	71.58 202	88.30 191	91.14 204	92.31 195	98.75 158	98.52 101

Vis-MVSNet (Re-imp)		94.46 72	96.24 50	92.40 111	95.23 95	98.64 74	95.56 103	90.99 92	94.42 100	85.02 111	90.88 87	94.65 58	88.01 192	98.17 29	98.37 14	99.57 8	98.53 100

v7n		86.43 204	86.52 207	86.33 204	87.91 217	94.93 188	90.15 202	83.05 185	86.57 208	70.21 208	71.48 216	66.78 224	87.72 193	94.19 152	92.96 165	98.92 123	98.76 85

pmmvs6		85.98 207	84.89 217	87.25 194	88.83 210	94.35 203	89.36 207	85.30 166	78.51 232	75.44 161	62.71 233	75.41 165	87.65 194	93.58 158	92.40 193	96.89 203	97.29 164

DTE-MVSNet		86.67 201	86.09 209	87.35 193	88.45 214	94.08 205	90.65 197	86.05 154	86.13 214	72.19 195	74.58 187	66.77 225	87.61 195	90.31 209	93.12 161	99.13 101	97.62 156

MDTV_nov1_ep13_2view		86.30 205	88.27 164	84.01 212	87.71 219	94.67 196	88.08 210	76.78 217	90.59 163	68.66 216	80.46 161	80.12 140	87.58 196	89.95 214	88.20 218	95.25 222	93.90 210

pmmvs-eth3d		84.33 215	82.94 223	85.96 209	84.16 229	90.94 225	86.55 217	83.79 179	84.25 221	75.85 159	70.64 221	56.43 237	87.44 197	92.20 182	90.41 210	97.97 193	95.68 194

LP		84.43 214	85.10 215	83.66 213	92.31 148	93.89 206	87.13 213	72.88 231	90.81 158	67.08 218	70.65 220	75.76 164	86.87 198	86.43 227	87.15 225	95.70 213	90.98 222

v148		87.51 188	86.79 201	88.36 161	89.39 191	95.21 180	89.84 205	88.20 127	87.61 202	77.56 145	73.38 206	70.32 211	86.80 199	90.70 207	92.31 195	98.37 184	97.98 134

TransMVSNet (Re)		87.73 185	86.79 201	88.83 153	90.76 159	94.40 202	91.33 191	89.62 110	84.73 219	75.41 162	72.73 208	71.41 204	86.80 199	94.53 142	93.93 146	99.06 112	95.83 191

WR-MVS_H		87.93 177	87.85 177	88.03 174	89.62 177	95.58 162	90.47 199	85.55 160	87.20 206	76.83 151	74.42 190	72.67 195	86.37 201	93.22 165	93.04 162	99.33 63	98.83 81

UGNet		94.92 58	96.63 44	92.93 105	96.03 72	98.63 76	94.53 125	91.52 88	96.23 59	90.03 69	92.87 63	96.10 52	86.28 202	96.68 77	96.60 62	99.16 97	99.32 20
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
UA-Net		93.96 83	95.95 54	91.64 117	96.06 71	98.59 78	95.29 108	90.00 102	91.06 154	82.87 120	90.64 88	98.06 35	86.06 203	98.14 30	98.20 16	99.58 6	96.96 176

test0.0.03 1		91.97 116	93.91 90	89.72 140	93.31 138	96.40 135	91.34 190	87.06 143	93.86 109	81.67 129	91.15 83	89.16 88	86.02 204	95.08 134	95.09 115	98.91 124	96.64 185

thisisatest0515		90.12 143	92.06 127	87.85 178	90.03 169	96.17 139	87.83 211	87.45 138	91.71 148	77.15 148	85.40 133	84.01 121	85.74 205	95.41 127	93.30 159	98.88 126	98.43 107

FC-MVSNet-test		91.63 121	93.82 95	89.08 148	92.02 150	96.40 135	93.26 141	87.26 140	93.72 112	77.26 147	88.61 105	89.86 81	85.50 206	95.72 120	95.02 117	99.16 97	97.44 159

CMPMVS	65.18 17	84.76 212	83.10 222	86.69 199	95.29 92	95.05 184	88.37 209	85.51 161	80.27 230	71.31 200	68.37 225	73.85 173	85.25 207	87.72 221	87.75 220	94.38 230	88.70 228
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
N_pmnet		84.80 211	85.10 215	84.45 211	89.25 202	92.86 211	84.04 222	86.21 149	88.78 183	66.73 219	72.41 211	74.87 170	85.21 208	88.32 219	86.45 228	95.30 220	92.04 218

WR-MVS		87.93 177	88.09 167	87.75 181	89.26 199	95.28 176	90.81 196	86.69 146	88.90 181	75.29 164	74.31 192	73.72 174	85.19 209	92.26 177	93.32 158	99.27 74	98.81 82

TDRefinement		89.07 158	88.15 166	90.14 137	95.16 98	96.88 115	95.55 104	90.20 100	89.68 169	76.42 154	76.67 170	74.30 171	84.85 210	93.11 166	91.91 200	98.64 170	94.47 202

CVMVSNet		89.77 148	91.66 132	87.56 190	93.21 140	95.45 165	91.94 184	89.22 116	89.62 172	69.34 214	83.99 142	85.90 108	84.81 211	94.30 148	95.28 112	96.85 204	97.09 169

pmmvs3		79.16 225	80.12 227	78.05 226	79.36 234	86.59 234	78.13 234	73.87 230	76.42 234	57.51 234	70.59 222	57.02 235	84.66 212	90.10 211	88.32 217	94.75 227	91.77 220

Vis-MVSNet		92.77 109	95.00 72	90.16 135	94.10 126	98.79 62	94.76 120	88.26 125	92.37 140	79.95 136	88.19 108	91.58 69	84.38 213	97.59 46	97.58 37	99.52 14	98.91 73
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
EG-PatchMatch MVS		86.68 200	87.24 191	86.02 208	90.58 161	96.26 137	91.08 194	81.59 201	84.96 218	69.80 212	71.35 218	75.08 168	84.23 214	94.24 150	93.35 157	98.82 131	95.46 198

testgi		89.42 150	91.50 136	87.00 197	92.40 147	95.59 160	89.15 208	85.27 167	92.78 126	72.42 194	91.75 77	76.00 163	84.09 215	94.38 146	93.82 151	98.65 169	96.15 186

EPNet_dtu		92.45 114	95.02 71	89.46 144	98.02 48	95.47 164	94.79 119	92.62 60	94.97 93	70.11 209	94.76 46	92.61 67	84.07 216	95.94 111	95.56 105	97.15 202	95.82 192
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
v748		85.88 208	85.66 211	86.14 206	88.03 215	94.63 198	87.02 216	84.59 175	84.30 220	74.56 173	70.94 219	67.27 222	83.94 217	90.96 206	92.74 184	98.71 161	98.81 82

MDA-MVSNet-bldmvs		80.11 223	80.24 226	79.94 221	77.01 239	93.21 209	78.86 233	85.94 156	82.71 227	60.86 226	79.71 163	51.77 240	83.71 218	75.60 237	86.37 229	93.28 233	92.35 217

new_pmnet		81.53 220	82.68 224	80.20 220	83.47 231	89.47 229	82.21 228	78.36 210	87.86 198	60.14 231	67.90 226	69.43 215	82.03 219	89.22 216	87.47 222	94.99 224	87.39 229

testpf		83.57 217	85.70 210	81.08 218	90.99 158	88.96 230	82.71 226	65.32 242	90.22 167	73.86 180	81.58 155	76.10 161	81.19 220	84.14 232	85.41 232	92.43 235	93.45 215

DeepMVS_CX						86.86 233	79.50 232	70.43 236	90.73 159	63.66 223	80.36 162	60.83 229	79.68 221	76.23 236		89.46 238	86.53 232

EU-MVSNet		85.62 209	87.65 184	83.24 216	88.54 213	92.77 212	87.12 214	85.32 164	86.71 207	64.54 222	78.52 166	75.11 167	78.35 222	92.25 178	92.28 197	95.58 216	95.93 188

IB-MVS	89.56 15	91.71 120	92.50 113	90.79 127	95.94 74	98.44 83	87.05 215	91.38 89	93.15 119	92.98 39	84.78 136	85.14 115	78.27 223	92.47 176	94.44 137	99.10 105	99.08 49
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
MIMVSNet1		80.03 224	80.93 225	78.97 223	72.46 242	90.73 226	80.81 229	82.44 193	80.39 229	63.64 224	57.57 235	64.93 228	76.37 224	91.66 200	91.55 203	98.07 191	89.70 226

new-patchmatchnet		78.49 226	78.19 228	78.84 224	84.13 230	90.06 227	77.11 235	80.39 206	79.57 231	59.64 232	66.01 228	55.65 239	75.62 225	84.55 231	80.70 234	96.14 210	90.77 224

Anonymous20231206		83.84 216	85.19 214	82.26 217	87.38 222	92.87 210	85.49 220	83.65 180	86.07 216	63.44 225	68.42 224	69.01 217	75.45 226	93.34 162	92.44 192	98.12 190	94.20 205

test2356		81.26 222	84.10 221	77.95 227	84.35 228	87.38 232	79.56 231	79.53 208	86.17 213	54.14 238	83.24 146	60.71 230	73.77 227	90.01 213	91.18 204	96.33 207	90.01 225

1111		73.35 228	74.40 230	72.12 229	78.22 235	82.24 237	65.06 239	65.61 240	70.28 236	55.42 235	56.30 236	57.35 233	73.66 228	86.73 225	88.16 219	94.75 227	79.76 237

.test1245		56.65 236	56.09 238	57.30 237	78.22 235	82.24 237	65.06 239	65.61 240	70.28 236	55.42 235	56.30 236	57.35 233	73.66 228	86.73 225	15.01 243	5.84 247	24.75 243

ambc			73.83 233		76.23 240	85.13 235	82.27 227		84.16 222	65.58 221	52.82 239	23.31 249	73.55 230	91.41 202	85.26 233	92.97 234	94.70 200

testus		81.33 221	84.13 220	78.06 225	84.54 227	87.72 231	79.66 230	80.42 205	87.36 205	54.13 239	83.83 143	56.63 236	73.21 231	90.51 208	91.74 201	96.40 206	91.11 221

Gipuma		68.35 232	66.71 235	70.27 232	74.16 241	68.78 245	63.93 244	71.77 235	83.34 225	54.57 237	34.37 241	31.88 245	68.69 232	83.30 233	85.53 231	88.48 239	79.78 236
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
test20.03		82.92 219	85.52 212	79.90 222	87.75 218	91.84 223	82.80 225	82.99 186	82.65 228	60.32 229	78.90 165	70.50 207	67.10 233	92.05 198	90.89 205	98.44 181	91.80 219

FPMVS		75.84 227	74.59 229	77.29 228	86.92 223	83.89 236	85.01 221	80.05 207	82.91 226	60.61 228	65.25 229	60.41 231	63.86 234	75.60 237	73.60 239	87.29 240	80.47 235

testmv		72.66 229	74.40 230	70.62 230	80.64 232	81.51 239	64.99 241	76.60 218	68.76 238	44.81 241	63.78 231	48.00 241	62.52 235	84.74 229	87.17 223	94.19 231	86.86 230

test1235678		72.65 230	74.40 230	70.62 230	80.64 232	81.50 240	64.99 241	76.59 219	68.74 239	44.81 241	63.78 231	47.99 242	62.51 236	84.73 230	87.17 223	94.19 231	86.85 231

EMVS		49.98 239	46.76 242	53.74 240	64.96 245	51.29 248	37.81 248	69.35 238	51.83 243	22.69 248	29.57 244	25.06 247	57.28 237	44.81 244	56.11 242	70.32 245	68.64 242

E-PMN		50.67 238	47.85 241	53.96 239	64.13 246	50.98 249	38.06 247	69.51 237	51.40 244	24.60 247	29.46 245	24.39 248	56.07 238	48.17 243	59.70 241	71.40 244	70.84 241

PMVS	63.12 18	67.27 233	66.39 236	68.30 233	77.98 237	60.24 246	59.53 245	76.82 215	66.65 241	60.74 227	54.39 238	59.82 232	51.24 239	73.92 240	70.52 240	83.48 242	79.17 238
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
tmp_tt				66.88 235	86.07 225	73.86 243	68.22 237	33.38 244	96.88 44	80.67 135	88.23 107	78.82 144	49.78 240	82.68 234	77.47 236	83.19 243

MVE	50.86 19	49.54 240	51.43 240	47.33 241	44.14 247	59.20 247	36.45 249	60.59 243	41.47 246	31.14 246	29.58 243	17.06 250	48.52 241	62.22 241	74.63 238	63.12 246	75.87 240
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
PMMVS2		64.36 235	65.94 237	62.52 236	67.37 244	77.44 242	64.39 243	69.32 239	61.47 242	34.59 245	46.09 240	41.03 243	48.02 242	74.56 239	78.23 235	91.43 237	82.76 234

test12356		69.55 231	71.53 234	67.24 234	77.70 238	78.48 241	65.92 238	75.55 225	68.39 240	44.26 243	61.80 234	40.70 244	47.92 243	81.45 235	87.01 227	92.09 236	82.89 233

no-one		55.96 237	55.63 239	56.35 238	68.48 243	73.29 244	43.03 246	72.52 233	44.01 245	34.80 244	32.83 242	29.11 246	35.21 244	56.63 242	75.72 237	84.04 241	77.79 239

testmvs		12.09 241	16.94 243	6.42 243	3.15 248	6.08 250	9.51 251	3.84 245	21.46 247	5.31 249	27.49 246	6.76 251	10.89 245	17.06 245	15.01 243	5.84 247	24.75 243

test123		9.58 242	13.53 244	4.97 244	1.31 250	5.47 251	8.32 252	2.95 246	18.14 248	2.03 251	20.82 247	2.34 252	10.60 246	10.00 246	14.16 245	4.60 249	23.77 245

GG-mvs-BLEND		66.17 234	94.91 73	32.63 242	1.32 249	96.64 127	91.40 188	0.85 247	94.39 102	2.20 250	90.15 94	95.70 54	2.27 247	96.39 93	95.44 109	97.78 195	95.68 194

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

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

our_test_3					89.78 172	93.84 207	85.59 219

MTAPA										96.83 6		99.12 16

MTMP										97.18 3		98.83 22

Patchmatch-RL test							34.61 250

XVS					96.60 63	99.35 11	96.82 61			90.85 55		98.72 25				99.46 27

X-MVStestdata					96.60 63	99.35 11	96.82 61			90.85 55		98.72 25				99.46 27

mPP-MVS					99.21 20							98.29 33

NP-MVS									95.32 85

Patchmtry						95.96 145	93.36 139	75.99 223		75.19 165

High-resolution multi-view benchmark