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.

CS-MVS-test		98.09 55	99.32 28	96.67 79	95.48 131	99.61 38	99.01 51	92.22 100	99.32 33	93.89 81	99.30 17	98.77 66	99.49 20	99.16 35	99.16 24	99.92 1	99.91 5

DROMVSNet		98.22 50	99.44 17	96.79 76	95.62 120	99.56 52	99.01 51	92.22 100	99.17 54	94.51 67	99.41 13	99.62 53	99.49 20	99.16 35	99.26 14	99.91 2	99.94 1

test2506		97.16 83	96.68 132	97.73 49	96.95 86	99.79 2	98.48 69	94.42 66	99.17 54	97.74 24	99.15 25	80.93 201	98.89 67	99.03 43	99.09 26	99.88 3	99.62 117

test1111		97.09 87	96.83 129	97.39 55	96.92 88	99.81 1	98.44 74	94.45 65	99.17 54	95.85 45	92.10 163	88.97 151	98.78 71	99.02 45	99.11 25	99.88 3	99.63 115

ECVR-MVS		97.27 80	97.09 119	97.48 54	96.95 86	99.79 2	98.48 69	94.42 66	99.17 54	96.28 40	93.54 150	89.39 150	98.89 67	99.03 43	99.09 26	99.88 3	99.61 120

CS-MVS		97.98 59	99.26 33	96.48 89	95.60 123	99.67 16	98.46 72	93.16 95	99.37 26	92.22 110	98.49 60	98.95 65	99.55 14	99.27 27	99.17 23	99.88 3	99.92 2

TSAR-MVS + MP.		99.27 11	99.57 5	98.92 24	98.78 55	99.53 56	99.72 2	98.11 30	99.73 2	97.43 27	99.15 25	99.96 13	99.59 10	99.73 1	99.07 28	99.88 3	99.82 28
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
ETV-MVS		98.05 56	99.25 34	96.65 81	95.61 121	99.61 38	98.26 86	93.52 85	98.90 92	93.74 86	99.32 16	99.20 60	98.90 64	99.21 31	98.72 55	99.87 8	99.79 42

DPE-MVS		99.39 5	99.55 6	99.20 4	99.63 22	99.71 13	99.66 6	98.33 6	99.29 37	98.40 12	99.64 5	99.98 2	99.31 35	99.56 10	98.96 37	99.85 9	99.70 92
Kehua Chen, Zhenlong Yuan, Tianlu Mao, Zhaoqi Wang: Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization. AAAI2025
3Dnovator	96.92 7	98.67 38	99.05 45	98.23 39	99.57 28	99.45 67	99.11 43	94.66 60	99.69 3	96.80 35	96.55 114	99.61 54	99.40 28	98.87 58	99.49 3	99.85 9	99.66 106

APDe-MVS		99.49 1	99.64 1	99.32 2	99.74 4	99.74 9	99.75 1	98.34 4	99.56 11	98.72 7	99.57 7	99.97 8	99.53 17	99.65 2	99.25 15	99.84 11	99.77 56

Vis-MVSNet		96.16 115	98.22 77	93.75 131	95.33 135	99.70 15	97.27 118	90.85 127	98.30 137	85.51 148	95.72 131	96.45 90	93.69 196	98.70 70	99.00 34	99.84 11	99.69 96
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
EPP-MVSNet		97.75 65	98.71 61	96.63 83	95.68 118	99.56 52	97.51 110	93.10 96	99.22 47	94.99 59	97.18 96	97.30 84	98.65 77	98.83 59	98.93 39	99.84 11	99.92 2

IS_MVSNet		97.86 61	98.86 56	96.68 78	96.02 104	99.72 10	98.35 81	93.37 89	98.75 116	94.01 76	96.88 103	98.40 71	98.48 86	99.09 39	99.42 5	99.83 14	99.80 35

tfpn200view9		96.75 97	96.51 138	97.03 68	96.31 96	99.67 16	98.41 75	93.99 75	97.35 170	94.52 65	95.90 125	86.93 160	99.14 48	98.26 95	97.80 107	99.82 15	99.70 92

thres600view7		96.69 101	96.43 145	97.00 73	96.28 99	99.67 16	98.41 75	93.99 75	97.85 159	94.29 74	95.96 123	85.91 172	99.19 42	98.26 95	97.63 112	99.82 15	99.73 76

thres200		96.76 96	96.53 136	97.03 68	96.31 96	99.67 16	98.37 78	93.99 75	97.68 165	94.49 68	95.83 128	86.77 162	99.18 44	98.26 95	97.82 106	99.82 15	99.66 106

MCST-MVS		99.11 21	99.27 32	98.93 23	99.67 14	99.33 89	99.51 21	98.31 9	99.28 38	96.57 38	99.10 32	99.90 33	99.71 2	99.19 32	98.35 76	99.82 15	99.71 90

casdiffmvs		96.93 92	97.43 106	96.34 91	95.70 116	99.50 61	97.75 104	93.22 92	98.98 85	92.64 101	94.97 136	91.71 138	98.93 60	98.62 75	98.52 66	99.82 15	99.72 87

DELS-MVS		98.19 51	98.77 60	97.52 53	98.29 62	99.71 13	99.12 42	94.58 64	98.80 106	95.38 52	96.24 119	98.24 74	97.92 102	99.06 42	99.52 1	99.82 15	99.79 42
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
thres100view900		96.72 99	96.47 141	97.00 73	96.31 96	99.52 59	98.28 84	94.01 73	97.35 170	94.52 65	95.90 125	86.93 160	99.09 53	98.07 108	97.87 103	99.81 21	99.63 115

MVS_0304		98.14 53	99.03 49	97.10 64	98.05 66	99.63 29	99.27 35	94.33 69	99.63 7	93.06 95	97.32 90	99.05 63	98.09 95	98.82 60	98.87 44	99.81 21	99.89 10

MSP-MVS		99.34 7	99.52 10	99.14 8	99.68 13	99.75 7	99.64 8	98.31 9	99.44 21	98.10 14	99.28 18	99.98 2	99.30 37	99.34 23	99.05 30	99.81 21	99.79 42
Zhenlong Yuan, Cong Liu, Fei Shen, Zhaoxin Li, Jingguo luo, Tianlu Mao and Zhaoqi Wang: MSP-MVS: Multi-granularity Segmentation Prior Guided Multi-View Stereo. AAAI2025
UA-Net		97.13 85	99.14 39	94.78 115	97.21 80	99.38 77	97.56 109	92.04 104	98.48 129	88.03 130	98.39 66	99.91 32	94.03 190	99.33 24	99.23 18	99.81 21	99.25 159

SD-MVS		99.25 13	99.50 12	98.96 22	98.79 54	99.55 54	99.33 33	98.29 12	99.75 1	97.96 20	99.15 25	99.95 18	99.61 6	99.17 33	99.06 29	99.81 21	99.84 23
Zhenlong Yuan, Jiakai Cao, Zhaoxin Li, Hao Jiang and Zhaoqi Wang: SD-MVS: Segmentation-driven Deformation Multi-View Stereo with Spherical Refinement and EM optimization. AAAI2024
thres400		96.71 100	96.45 143	97.02 70	96.28 99	99.63 29	98.41 75	94.00 74	97.82 160	94.42 71	95.74 129	86.26 169	99.18 44	98.20 99	97.79 108	99.81 21	99.70 92

baseline1		97.58 70	98.05 84	97.02 70	96.21 101	99.45 67	97.71 105	93.71 83	98.47 130	95.75 46	98.78 49	93.20 132	98.91 63	98.52 85	98.44 68	99.81 21	99.53 132

3Dnovator+	96.92 7	98.71 37	99.05 45	98.32 35	99.53 31	99.34 86	99.06 47	94.61 61	99.65 5	97.49 26	96.75 104	99.86 38	99.44 26	98.78 63	99.30 11	99.81 21	99.67 102

DeepC-MVS_fast	98.34 1	99.17 18	99.45 14	98.85 26	99.55 30	99.37 80	99.64 8	98.05 33	99.53 14	96.58 37	98.93 41	99.92 29	99.49 20	99.46 15	99.32 10	99.80 29	99.64 113
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
APD-MVS		99.25 13	99.38 22	99.09 12	99.69 8	99.58 49	99.56 18	98.32 8	98.85 96	97.87 21	98.91 43	99.92 29	99.30 37	99.45 16	99.38 8	99.79 30	99.58 122
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
Vis-MVSNet (Re-imp)		97.40 77	98.89 55	95.66 107	95.99 107	99.62 33	97.82 100	93.22 92	98.82 103	91.40 115	96.94 100	98.56 69	95.70 159	99.14 37	99.41 6	99.79 30	99.75 68

PHI-MVS		99.08 23	99.43 20	98.67 30	99.15 47	99.59 46	99.11 43	97.35 41	99.14 62	97.30 29	99.44 12	99.96 13	99.32 34	98.89 56	99.39 7	99.79 30	99.58 122

EIA-MVS		97.70 67	98.78 59	96.44 90	95.72 115	99.65 21	98.14 90	93.72 82	98.30 137	92.31 106	98.63 56	97.90 76	98.97 59	98.92 53	98.30 82	99.78 33	99.80 35

CP-MVS		99.27 11	99.44 17	99.08 13	99.62 24	99.58 49	99.53 19	98.16 23	99.21 49	97.79 22	99.15 25	99.96 13	99.59 10	99.54 12	98.86 45	99.78 33	99.74 72

SteuartSystems-ACMMP		99.20 16	99.51 11	98.83 28	99.66 17	99.66 20	99.71 3	98.12 29	99.14 62	96.62 36	99.16 24	99.98 2	99.12 49	99.63 3	99.19 21	99.78 33	99.83 27
Skip Steuart: Steuart Systems R&D Blog.
CSCG		98.90 31	98.93 54	98.85 26	99.75 3	99.72 10	99.49 22	96.58 44	99.38 24	98.05 17	98.97 38	97.87 77	99.49 20	97.78 127	98.92 40	99.78 33	99.90 6

DeepC-MVS	97.63 4	98.33 47	98.57 63	98.04 43	98.62 58	99.65 21	99.45 26	98.15 25	99.51 17	92.80 100	95.74 129	96.44 92	99.46 24	99.37 20	99.50 2	99.78 33	99.81 33
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
NCCC		99.05 26	99.08 42	99.02 20	99.62 24	99.38 77	99.43 29	98.21 21	99.36 29	97.66 25	97.79 81	99.90 33	99.45 25	99.17 33	98.43 70	99.77 38	99.51 139

PVSNet_Blended_VisFu		97.41 76	98.49 67	96.15 94	97.49 72	99.76 4	96.02 150	93.75 81	99.26 43	93.38 91	93.73 148	99.35 58	96.47 142	98.96 48	98.46 67	99.77 38	99.90 6

xxxxxxxxxxxxxcwj		98.14 53	97.38 108	99.03 17	99.65 19	99.41 74	98.87 56	98.24 18	99.14 62	98.73 5	99.11 29	86.38 168	98.92 61	99.22 29	98.84 48	99.76 40	99.56 128

SF-MVS		99.18 17	99.32 28	99.03 17	99.65 19	99.41 74	98.87 56	98.24 18	99.14 62	98.73 5	99.11 29	99.92 29	98.92 61	99.22 29	98.84 48	99.76 40	99.56 128

DCV-MVSNet		97.56 71	98.36 70	96.62 84	96.44 93	98.36 154	98.37 78	91.73 110	99.11 68	94.80 61	98.36 67	96.28 95	98.60 81	98.12 102	98.44 68	99.76 40	99.87 16

canonicalmvs		97.31 78	97.81 94	96.72 77	96.20 102	99.45 67	98.21 87	91.60 113	99.22 47	95.39 51	98.48 61	90.95 140	99.16 47	97.66 134	99.05 30	99.76 40	99.90 6

Anonymous202405211			97.40 107		96.45 92	99.54 55	98.08 95	93.79 78	98.24 141		93.55 149	94.41 118	98.88 69	98.04 113	98.24 85	99.75 44	99.76 61

FC-MVSNet-train		97.04 88	97.91 91	96.03 98	96.00 106	98.41 150	96.53 141	93.42 86	99.04 80	93.02 97	98.03 76	94.32 120	97.47 116	97.93 119	97.77 109	99.75 44	99.88 14

CDPH-MVS		98.41 44	99.10 41	97.61 52	99.32 44	99.36 81	99.49 22	96.15 46	98.82 103	91.82 112	98.41 64	99.66 52	99.10 51	98.93 51	98.97 36	99.75 44	99.58 122

TSAR-MVS + GP.		98.66 40	99.36 24	97.85 47	97.16 82	99.46 65	99.03 49	94.59 63	99.09 70	97.19 31	99.73 3	99.95 18	99.39 29	98.95 49	98.69 56	99.75 44	99.65 109

QAPM		98.62 41	99.04 48	98.13 40	99.57 28	99.48 63	99.17 39	94.78 57	99.57 10	96.16 41	96.73 105	99.80 44	99.33 32	98.79 62	99.29 13	99.75 44	99.64 113

DVP-MVS++		99.41 4	99.64 1	99.14 8	99.69 8	99.75 7	99.64 8	98.33 6	99.67 4	98.10 14	99.66 4	99.99 1	99.33 32	99.62 5	98.86 45	99.74 49	99.90 6

DVP-MVS		99.45 2	99.54 7	99.35 1	99.72 7	99.76 4	99.63 12	98.37 2	99.63 7	99.03 3	98.95 40	99.98 2	99.60 7	99.60 7	99.05 30	99.74 49	99.79 42
Zhenlong Yuan, Jinguo Luo, Fei Shen, Zhaoxin Li, Cong Liu, Tianlu Mao, Zhaoqi Wang: DVP-MVS: Synergize Depth-Edge and Visibility Prior for Multi-View Stereo. AAAI2025
CANet		98.46 43	99.16 38	97.64 51	98.48 59	99.64 26	99.35 32	94.71 59	99.53 14	95.17 55	97.63 87	99.59 55	98.38 88	98.88 57	98.99 35	99.74 49	99.86 19

Effi-MVS+		95.81 121	97.31 115	94.06 126	95.09 138	99.35 84	97.24 120	88.22 164	98.54 125	85.38 149	98.52 58	88.68 152	98.70 74	98.32 93	97.93 98	99.74 49	99.84 23

DI_MVS_plusplus_trai		96.90 93	97.49 101	96.21 93	95.61 121	99.40 76	98.72 63	92.11 102	99.14 62	92.98 99	93.08 160	95.14 108	98.13 94	98.05 112	97.91 101	99.74 49	99.73 76

OpenMVS	96.23 11	97.95 60	98.45 68	97.35 56	99.52 33	99.42 72	98.91 55	94.61 61	98.87 93	92.24 109	94.61 140	99.05 63	99.10 51	98.64 73	99.05 30	99.74 49	99.51 139

ACMM	96.26 9	96.67 103	96.69 131	96.66 80	97.29 79	98.46 145	96.48 142	95.09 52	99.21 49	93.19 93	98.78 49	86.73 163	98.17 90	97.84 125	96.32 151	99.74 49	99.49 142
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
LTVRE_ROB	93.20 16	92.84 175	94.92 162	90.43 189	92.83 164	98.63 133	97.08 129	87.87 168	97.91 155	68.42 215	93.54 150	79.46 211	96.62 137	97.55 140	97.40 126	99.74 49	99.92 2
Andreas Kuhn, Heiko HirschmÃ¼ller, Daniel Scharstein, Helmut Mayer: A TV Prior for High-Quality Scalable Multi-View Stereo Reconstruction. International Journal of Computer Vision 2016
GeoE		95.98 120	97.24 117	94.51 119	95.02 140	99.38 77	98.02 97	87.86 169	98.37 134	87.86 133	92.99 162	93.54 127	98.56 82	98.61 76	97.92 99	99.73 57	99.85 22

tttt0517		97.23 81	98.24 76	96.04 97	95.60 123	99.60 44	96.94 133	93.23 90	99.15 59	92.56 104	98.74 53	96.12 99	98.17 90	98.21 98	96.10 159	99.73 57	99.78 48

PGM-MVS		98.86 32	99.35 27	98.29 36	99.77 1	99.63 29	99.67 5	95.63 47	98.66 119	95.27 53	99.11 29	99.82 43	99.67 4	99.33 24	99.19 21	99.73 57	99.74 72

WR-MVS_H		93.54 165	94.67 169	92.22 155	91.95 179	97.91 169	94.58 183	88.75 157	96.64 190	83.88 153	90.66 173	85.13 178	94.40 184	96.54 167	95.91 166	99.73 57	99.89 10

CNVR-MVS		99.23 15	99.28 31	99.17 5	99.65 19	99.34 86	99.46 25	98.21 21	99.28 38	98.47 9	98.89 45	99.94 26	99.50 18	99.42 18	98.61 60	99.73 57	99.52 135

UGNet		97.66 68	99.07 44	96.01 99	97.19 81	99.65 21	97.09 128	93.39 87	99.35 30	94.40 72	98.79 48	99.59 55	94.24 187	98.04 113	98.29 83	99.73 57	99.80 35
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
HyFIR lowres test		95.99 118	96.56 134	95.32 110	97.99 68	99.65 21	96.54 139	88.86 155	98.44 131	89.77 125	84.14 207	97.05 87	99.03 56	98.55 83	98.19 88	99.73 57	99.86 19

thisisatest0530		97.23 81	98.25 73	96.05 96	95.60 123	99.59 46	96.96 132	93.23 90	99.17 54	92.60 103	98.75 52	96.19 96	98.17 90	98.19 100	96.10 159	99.72 64	99.77 56

Fast-Effi-MVS+		95.38 131	96.52 137	94.05 127	94.15 150	99.14 104	97.24 120	86.79 175	98.53 126	87.62 135	94.51 141	87.06 157	98.76 72	98.60 79	98.04 96	99.72 64	99.77 56

ACMMPR		99.30 10	99.54 7	99.03 17	99.66 17	99.64 26	99.68 4	98.25 15	99.56 11	97.12 32	99.19 22	99.95 18	99.72 1	99.43 17	99.25 15	99.72 64	99.77 56

UniMVSNet_NR-MVSNet		94.59 148	95.47 157	93.55 138	91.85 183	97.89 170	95.03 165	92.00 105	97.33 172	86.12 141	93.19 156	87.29 156	96.60 138	96.12 180	96.70 139	99.72 64	99.80 35

PEN-MVS		92.72 180	93.20 196	92.15 158	91.29 200	97.31 199	94.67 180	89.81 144	96.19 196	81.83 170	88.58 188	79.06 212	95.61 163	95.21 193	96.27 152	99.72 64	99.82 28

UniMVSNet (Re)		94.58 149	95.34 159	93.71 133	92.25 175	98.08 163	94.97 167	91.29 124	97.03 181	87.94 131	93.97 147	86.25 170	96.07 151	96.27 177	95.97 164	99.72 64	99.79 42

PVSNet_BlendedMVS		97.51 73	97.71 95	97.28 59	98.06 64	99.61 38	97.31 116	95.02 53	99.08 72	95.51 49	98.05 74	90.11 143	98.07 96	98.91 54	98.40 71	99.72 64	99.78 48

PVSNet_Blended		97.51 73	97.71 95	97.28 59	98.06 64	99.61 38	97.31 116	95.02 53	99.08 72	95.51 49	98.05 74	90.11 143	98.07 96	98.91 54	98.40 71	99.72 64	99.78 48

IB-MVS	93.96 15	95.02 137	96.44 144	93.36 144	97.05 84	99.28 93	90.43 204	93.39 87	98.02 148	96.02 42	94.92 138	92.07 136	83.52 213	95.38 190	95.82 167	99.72 64	99.59 121
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
CLD-MVS		96.74 98	96.51 138	97.01 72	96.71 90	98.62 134	98.73 62	94.38 68	98.94 88	94.46 69	97.33 89	87.03 158	98.07 96	97.20 153	96.87 136	99.72 64	99.54 131
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
zzz-MVS		99.31 9	99.44 17	99.16 6	99.73 5	99.65 21	99.63 12	98.26 14	99.27 40	98.01 19	99.27 19	99.97 8	99.60 7	99.59 8	98.58 62	99.71 74	99.73 76

Anonymous20231211		97.10 86	97.06 122	97.14 63	96.32 95	99.52 59	98.16 89	93.76 79	98.84 100	95.98 43	90.92 169	94.58 117	98.90 64	97.72 132	98.10 93	99.71 74	99.75 68

anonymousdsp		93.12 171	95.86 153	89.93 194	91.09 203	98.25 157	95.12 164	85.08 186	97.44 168	73.30 205	90.89 170	90.78 141	95.25 175	97.91 120	95.96 165	99.71 74	99.82 28

FC-MVSNet-test		96.07 117	97.94 90	93.89 128	93.60 160	98.67 131	96.62 138	90.30 139	98.76 113	88.62 126	95.57 134	97.63 80	94.48 183	97.97 117	97.48 121	99.71 74	99.52 135

HFP-MVS		99.32 8	99.53 9	99.07 14	99.69 8	99.59 46	99.63 12	98.31 9	99.56 11	97.37 28	99.27 19	99.97 8	99.70 3	99.35 22	99.24 17	99.71 74	99.76 61

MP-MVS		99.07 24	99.36 24	98.74 29	99.63 22	99.57 51	99.66 6	98.25 15	99.00 83	95.62 47	98.97 38	99.94 26	99.54 16	99.51 13	98.79 54	99.71 74	99.73 76
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
CP-MVSNet		93.25 169	94.00 182	92.38 154	91.65 190	97.56 189	94.38 186	89.20 152	96.05 200	83.16 161	89.51 179	81.97 196	96.16 150	96.43 169	96.56 145	99.71 74	99.89 10

WR-MVS		93.43 168	94.48 172	92.21 156	91.52 195	97.69 177	94.66 181	89.98 141	96.86 184	83.43 158	90.12 175	85.03 179	93.94 192	96.02 184	95.82 167	99.71 74	99.82 28

ACMMP		98.74 35	99.03 49	98.40 34	99.36 40	99.64 26	99.20 37	97.75 39	98.82 103	95.24 54	98.85 46	99.87 37	99.17 46	98.74 68	97.50 118	99.71 74	99.76 61
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
SED-MVS		99.44 3	99.58 4	99.28 3	99.69 8	99.76 4	99.62 15	98.35 3	99.51 17	99.05 2	99.60 6	99.98 2	99.28 39	99.61 6	98.83 50	99.70 83	99.77 56

ACMMP_NAP		99.05 26	99.45 14	98.58 32	99.73 5	99.60 44	99.64 8	98.28 13	99.23 46	94.57 64	99.35 15	99.97 8	99.55 14	99.63 3	98.66 57	99.70 83	99.74 72

PS-CasMVS		92.72 180	93.36 194	91.98 163	91.62 192	97.52 191	94.13 190	88.98 154	95.94 203	81.51 172	87.35 197	79.95 208	95.91 155	96.37 172	96.49 147	99.70 83	99.89 10

XVS					97.42 74	99.62 33	98.59 66			93.81 83		99.95 18				99.69 86

X-MVStestdata					97.42 74	99.62 33	98.59 66			93.81 83		99.95 18				99.69 86

NR-MVSNet		94.01 156	94.51 171	93.44 141	92.56 168	97.77 172	95.67 154	91.57 114	97.17 176	85.84 144	93.13 158	80.53 203	95.29 173	97.01 158	96.17 156	99.69 86	99.75 68

TranMVSNet+NR-MVSNet		93.67 164	94.14 176	93.13 147	91.28 202	97.58 187	95.60 157	91.97 106	97.06 179	84.05 150	90.64 174	82.22 195	96.17 149	94.94 199	96.78 137	99.69 86	99.78 48

abl_6				98.09 41	99.33 43	99.22 99	98.79 61	94.96 55	98.52 128	97.00 34	97.30 91	99.86 38	98.76 72			99.69 86	99.41 148

LGP-MVS_train		96.23 112	96.89 126	95.46 109	97.32 76	98.77 121	98.81 60	93.60 84	98.58 122	85.52 147	99.08 34	86.67 164	97.83 109	97.87 123	97.51 117	99.69 86	99.73 76

MVS_111021_HR		98.59 42	99.36 24	97.68 50	99.42 36	99.61 38	98.14 90	94.81 56	99.31 34	95.00 58	99.51 9	99.79 46	99.00 58	98.94 50	98.83 50	99.69 86	99.57 127

LS3D		97.79 62	98.25 73	97.26 61	98.40 60	99.63 29	99.53 19	98.63 1	99.25 45	88.13 129	96.93 101	94.14 122	99.19 42	99.14 37	99.23 18	99.69 86	99.42 147

ACMH	95.42 14	95.27 134	95.96 150	94.45 121	96.83 89	98.78 120	94.72 177	91.67 112	98.95 86	86.82 140	96.42 116	83.67 185	97.00 124	97.48 143	96.68 140	99.69 86	99.76 61
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
SMA-MVS		99.38 6	99.60 3	99.12 10	99.76 2	99.62 33	99.39 30	98.23 20	99.52 16	98.03 18	99.45 11	99.98 2	99.64 5	99.58 9	99.30 11	99.68 95	99.76 61
Yufeng Yin; Xiaoyan Liu; Zichao Zhang: SMA-MVS: Segmentation-Guided Multi-Scale Anchor Deformation Patch Multi-View Stereo. IEEE Transactions on Circuits and Systems for Video Technology
tfpnnormal		93.85 163	94.12 178	93.54 139	93.22 163	98.24 158	95.45 160	91.96 107	94.61 209	83.91 152	90.74 171	81.75 198	97.04 123	97.49 142	96.16 157	99.68 95	99.84 23

X-MVS		98.93 30	99.37 23	98.42 33	99.67 14	99.62 33	99.60 16	98.15 25	99.08 72	93.81 83	98.46 63	99.95 18	99.59 10	99.49 14	99.21 20	99.68 95	99.75 68

Baseline_NR-MVSNet		93.87 161	93.98 183	93.75 131	91.66 188	97.02 202	95.53 158	91.52 117	97.16 178	87.77 134	87.93 195	83.69 184	96.35 144	95.10 196	97.23 129	99.68 95	99.73 76

GBi-Net		96.98 90	98.00 88	95.78 102	93.81 155	97.98 164	98.09 92	91.32 120	98.80 106	93.92 78	97.21 93	95.94 102	97.89 103	98.07 108	98.34 78	99.68 95	99.67 102

test1		96.98 90	98.00 88	95.78 102	93.81 155	97.98 164	98.09 92	91.32 120	98.80 106	93.92 78	97.21 93	95.94 102	97.89 103	98.07 108	98.34 78	99.68 95	99.67 102

FMVSNet2		96.64 104	97.50 100	95.63 108	93.81 155	97.98 164	98.09 92	90.87 126	98.99 84	93.48 89	93.17 157	95.25 107	97.89 103	98.63 74	98.80 53	99.68 95	99.67 102

COLMAP_ROB	96.15 12	97.78 63	98.17 79	97.32 57	98.84 52	99.45 67	99.28 34	95.43 50	99.48 19	91.80 113	94.83 139	98.36 72	98.90 64	98.09 105	97.85 104	99.68 95	99.15 164
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
OPM-MVS		96.22 113	95.85 154	96.65 81	97.75 69	98.54 140	99.00 53	95.53 48	96.88 183	89.88 123	95.95 124	86.46 167	98.07 96	97.65 136	96.63 142	99.67 103	98.83 180
Ray L. Khuboni, Hongjun Xu: Octagram Propagation Matching for Multi-Scale View Stereopsis (OPM-MVS).
TransMVSNet (Re)		93.45 166	94.08 179	92.72 152	92.83 164	97.62 185	94.94 168	91.54 116	95.65 206	83.06 162	88.93 185	83.53 186	94.25 186	97.41 144	97.03 132	99.67 103	98.40 189

DU-MVS		93.98 158	94.44 173	93.44 141	91.66 188	97.77 172	95.03 165	91.57 114	97.17 176	86.12 141	93.13 158	81.13 200	96.60 138	95.10 196	97.01 134	99.67 103	99.80 35

EG-PatchMatch MVS		92.45 184	93.92 185	90.72 186	92.56 168	98.43 149	94.88 171	84.54 191	97.18 175	79.55 183	86.12 204	83.23 189	93.15 200	97.22 152	96.00 161	99.67 103	99.27 158

MSDG		98.27 49	98.29 72	98.24 38	99.20 46	99.22 99	99.20 37	97.82 37	99.37 26	94.43 70	95.90 125	97.31 83	99.12 49	98.76 65	98.35 76	99.67 103	99.14 167

v8		92.87 174	93.87 187	91.72 171	92.05 177	97.50 192	94.79 175	88.20 165	96.85 185	80.11 180	90.01 176	82.86 192	95.48 167	95.15 195	94.90 188	99.66 108	99.80 35

v2v482		92.77 179	93.52 193	91.90 167	91.59 193	97.63 182	94.57 184	90.31 137	96.80 187	79.22 184	88.74 187	81.55 199	96.04 153	95.26 192	94.97 186	99.66 108	99.69 96

MVS_Test		97.30 79	98.54 64	95.87 101	95.74 114	99.28 93	98.19 88	91.40 118	99.18 53	91.59 114	98.17 72	96.18 97	98.63 79	98.61 76	98.55 63	99.66 108	99.78 48

TAPA-MVS	97.53 5	98.41 44	98.84 58	97.91 46	99.08 49	99.33 89	99.15 40	97.13 42	99.34 31	93.20 92	97.75 83	99.19 61	99.20 41	98.66 71	98.13 90	99.66 108	99.48 143
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
v1144		92.81 176	94.03 181	91.40 175	91.68 187	97.60 186	94.73 176	88.40 162	96.71 188	78.48 188	88.14 192	84.46 183	95.45 170	96.31 176	95.22 179	99.65 112	99.76 61

v7n		91.61 196	92.95 197	90.04 191	90.56 206	97.69 177	93.74 191	85.59 184	95.89 204	76.95 192	86.60 202	78.60 214	93.76 195	97.01 158	94.99 185	99.65 112	99.87 16

v10		92.79 178	94.06 180	91.31 177	91.78 185	97.29 201	94.87 172	86.10 182	96.97 182	79.82 182	88.16 191	84.56 182	95.63 161	96.33 175	95.31 176	99.65 112	99.80 35

PLC	97.93 2	99.02 29	98.94 53	99.11 11	99.46 35	99.24 97	99.06 47	97.96 35	99.31 34	99.16 1	97.90 79	99.79 46	99.36 30	98.71 69	98.12 91	99.65 112	99.52 135
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
HPM-MVS++		99.10 22	99.30 30	98.86 25	99.69 8	99.48 63	99.59 17	98.34 4	99.26 43	96.55 39	99.10 32	99.96 13	99.36 30	99.25 28	98.37 75	99.64 116	99.66 106

diffmvs		96.83 94	97.33 111	96.25 92	95.76 113	99.34 86	98.06 96	93.22 92	99.43 22	92.30 107	96.90 102	89.83 148	98.55 83	98.00 116	98.14 89	99.64 116	99.70 92

DTE-MVSNet		92.42 188	92.85 199	91.91 166	90.87 205	96.97 203	94.53 185	89.81 144	95.86 205	81.59 171	88.83 186	77.88 215	95.01 179	94.34 203	96.35 150	99.64 116	99.73 76

CPTT-MVS		99.14 20	99.20 37	99.06 15	99.58 27	99.53 56	99.45 26	97.80 38	99.19 52	98.32 13	98.58 57	99.95 18	99.60 7	99.28 26	98.20 87	99.64 116	99.69 96

AdaColmap		99.06 25	98.98 52	99.15 7	99.60 26	99.30 92	99.38 31	98.16 23	99.02 81	98.55 8	98.71 54	99.57 57	99.58 13	99.09 39	97.84 105	99.64 116	99.36 153

gg-mvs-nofinetune		90.85 197	94.14 176	87.02 203	94.89 143	99.25 95	98.64 64	76.29 217	88.24 218	57.50 222	79.93 213	95.45 105	95.18 176	98.77 64	98.07 94	99.62 121	99.24 160

ACMP	96.25 10	96.62 106	96.72 130	96.50 88	96.96 85	98.75 125	97.80 101	94.30 70	98.85 96	93.12 94	98.78 49	86.61 165	97.23 121	97.73 131	96.61 143	99.62 121	99.71 90
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
ET-MVSNet_ETH3D		96.17 114	96.99 124	95.21 111	88.53 211	98.54 140	98.28 84	92.61 98	98.85 96	93.60 88	99.06 36	90.39 142	98.63 79	95.98 185	96.68 140	99.61 123	99.41 148

thisisatest0515		94.61 147	96.89 126	91.95 164	92.00 178	98.47 144	92.01 199	90.73 132	98.18 142	83.96 151	94.51 141	95.13 109	93.38 197	97.38 145	94.74 193	99.61 123	99.79 42

Effi-MVS+-dtu		95.74 123	98.04 85	93.06 148	93.92 151	99.16 102	97.90 98	88.16 166	99.07 77	82.02 169	98.02 77	94.32 120	96.74 132	98.53 84	97.56 115	99.61 123	99.62 117

train_agg		98.73 36	99.11 40	98.28 37	99.36 40	99.35 84	99.48 24	97.96 35	98.83 101	93.86 82	98.70 55	99.86 38	99.44 26	99.08 41	98.38 73	99.61 123	99.58 122

pm-mvs1		94.27 152	95.57 156	92.75 151	92.58 167	98.13 162	94.87 172	90.71 133	96.70 189	83.78 154	89.94 177	89.85 147	94.96 180	97.58 139	97.07 131	99.61 123	99.72 87

MSLP-MVS++		99.15 19	99.24 35	99.04 16	99.52 33	99.49 62	99.09 45	98.07 31	99.37 26	98.47 9	97.79 81	99.89 35	99.50 18	98.93 51	99.45 4	99.61 123	99.76 61

FMVSNet3		97.02 89	98.12 82	95.73 106	93.59 161	97.98 164	98.34 82	91.32 120	98.80 106	93.92 78	97.21 93	95.94 102	97.63 112	98.61 76	98.62 59	99.61 123	99.65 109

FMVSNet1		95.77 122	96.41 146	95.03 112	93.42 162	97.86 171	97.11 127	89.89 143	98.53 126	92.00 111	89.17 181	93.23 131	98.15 93	98.07 108	98.34 78	99.61 123	99.69 96

v1192		92.43 187	93.61 189	91.05 181	91.53 194	97.43 195	94.61 182	87.99 167	96.60 191	76.72 193	87.11 199	82.74 193	95.85 156	96.35 174	95.30 177	99.60 131	99.74 72

testgi		95.67 124	97.48 102	93.56 137	95.07 139	99.00 107	95.33 163	88.47 161	98.80 106	86.90 139	97.30 91	92.33 134	95.97 154	97.66 134	97.91 101	99.60 131	99.38 152

test0.0.03 1		96.69 101	98.12 82	95.01 113	95.49 129	98.99 109	95.86 152	90.82 128	98.38 133	92.54 105	96.66 108	97.33 82	95.75 157	97.75 130	98.34 78	99.60 131	99.40 151

IterMVS-LS		96.12 116	97.48 102	94.53 118	95.19 137	97.56 189	97.15 124	89.19 153	99.08 72	88.23 128	94.97 136	94.73 114	97.84 108	97.86 124	98.26 84	99.60 131	99.88 14
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
CDS-MVSNet		96.59 107	98.02 87	94.92 114	94.45 148	98.96 112	97.46 112	91.75 109	97.86 158	90.07 121	96.02 122	97.25 85	96.21 146	98.04 113	98.38 73	99.60 131	99.65 109
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
MVS_111021_LR		98.67 38	99.41 21	97.81 48	99.37 38	99.53 56	98.51 68	95.52 49	99.27 40	94.85 60	99.56 8	99.69 51	99.04 55	99.36 21	98.88 43	99.60 131	99.58 122

V42		93.05 172	93.90 186	92.04 160	91.91 180	97.66 179	94.91 169	89.91 142	96.85 185	80.58 176	89.66 178	83.43 188	95.37 171	95.03 198	94.90 188	99.59 137	99.78 48

SixPastTwentyTwo		93.44 167	95.32 160	91.24 178	92.11 176	98.40 151	92.77 195	88.64 160	98.09 146	77.83 190	93.51 152	85.74 173	96.52 141	96.91 160	94.89 190	99.59 137	99.73 76

CHOSEN 1792x2688		96.41 108	96.99 124	95.74 105	98.01 67	99.72 10	97.70 106	90.78 130	99.13 67	90.03 122	87.35 197	95.36 106	98.33 89	98.59 81	98.91 42	99.59 137	99.87 16

ACMH+	95.51 13	95.40 130	96.00 148	94.70 116	96.33 94	98.79 118	96.79 134	91.32 120	98.77 112	87.18 137	95.60 133	85.46 175	96.97 125	97.15 154	96.59 144	99.59 137	99.65 109

pmmvs5		92.71 182	94.27 175	90.90 184	91.42 197	97.74 174	93.23 192	86.66 178	95.99 202	78.96 187	91.45 166	83.44 187	95.55 164	97.30 149	95.05 184	99.58 141	98.93 174

v148		92.36 191	92.88 198	91.75 169	91.63 191	97.66 179	92.64 196	90.55 135	96.09 198	83.34 159	88.19 190	80.00 206	92.74 201	93.98 204	94.58 194	99.58 141	99.69 96

GA-MVS		93.93 160	96.31 147	91.16 180	93.61 159	98.79 118	95.39 162	90.69 134	98.25 140	73.28 206	96.15 120	88.42 153	94.39 185	97.76 129	95.35 175	99.58 141	99.45 145

TSAR-MVS + ACMM		98.77 34	99.45 14	97.98 45	99.37 38	99.46 65	99.44 28	98.13 28	99.65 5	92.30 107	98.91 43	99.95 18	99.05 54	99.42 18	98.95 38	99.58 141	99.82 28

MVSTER		97.16 83	97.71 95	96.52 86	95.97 108	98.48 143	98.63 65	92.10 103	98.68 118	95.96 44	99.23 21	91.79 137	96.87 128	98.76 65	97.37 128	99.57 145	99.68 101

OMC-MVS		98.84 33	99.01 51	98.65 31	99.39 37	99.23 98	99.22 36	96.70 43	99.40 23	97.77 23	97.89 80	99.80 44	99.21 40	99.02 45	98.65 58	99.57 145	99.07 170

PCF-MVS	97.50 6	98.18 52	98.35 71	97.99 44	98.65 57	99.36 81	98.94 54	98.14 27	98.59 121	93.62 87	96.61 110	99.76 49	99.03 56	97.77 128	97.45 123	99.57 145	98.89 178
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
UniMVSNet_ETH3D		93.15 170	92.33 203	94.11 125	93.91 152	98.61 136	94.81 174	90.98 125	97.06 179	87.51 136	82.27 211	76.33 217	97.87 107	94.79 200	97.47 122	99.56 148	99.81 33

HQP-MVS		96.37 109	96.58 133	96.13 95	97.31 78	98.44 147	98.45 73	95.22 51	98.86 94	88.58 127	98.33 68	87.00 159	97.67 111	97.23 151	96.56 145	99.56 148	99.62 117

Fast-Effi-MVS+-dtu		95.38 131	98.20 78	92.09 159	93.91 152	98.87 115	97.35 115	85.01 188	99.08 72	81.09 173	98.10 73	96.36 93	95.62 162	98.43 91	97.03 132	99.55 150	99.50 141

v144192		92.38 189	93.55 192	91.00 182	91.44 196	97.47 194	94.27 187	87.41 172	96.52 193	78.03 189	87.50 196	82.65 194	95.32 172	95.82 188	95.15 181	99.55 150	99.78 48

v1921920		92.36 191	93.57 190	90.94 183	91.39 198	97.39 197	94.70 178	87.63 171	96.60 191	76.63 194	86.98 200	82.89 191	95.75 157	96.26 178	95.14 182	99.55 150	99.73 76

USDC		94.26 153	94.83 165	93.59 136	96.02 104	98.44 147	97.84 99	88.65 159	98.86 94	82.73 166	94.02 145	80.56 202	96.76 131	97.28 150	96.15 158	99.55 150	98.50 184

baseline2		96.36 110	97.82 93	94.65 117	94.60 147	99.09 105	96.45 143	89.63 148	98.36 135	91.29 117	97.60 88	94.13 123	96.37 143	98.45 88	97.70 110	99.54 154	99.41 148

TDRefinement		93.04 173	93.57 190	92.41 153	96.58 91	98.77 121	97.78 103	91.96 107	98.12 145	80.84 174	89.13 183	79.87 209	87.78 209	96.44 168	94.50 195	99.54 154	98.15 191

TinyColmap		94.00 157	94.35 174	93.60 135	95.89 109	98.26 156	97.49 111	88.82 156	98.56 124	83.21 160	91.28 168	80.48 204	96.68 134	97.34 147	96.26 154	99.53 156	98.24 190

v1240		91.99 194	93.33 195	90.44 188	91.29 200	97.30 200	94.25 188	86.79 175	96.43 194	75.49 200	86.34 203	81.85 197	95.29 173	96.42 170	95.22 179	99.52 157	99.73 76

test20.03		90.65 200	93.71 188	87.09 202	90.44 207	96.24 208	89.74 210	85.46 185	95.59 207	72.99 209	90.68 172	85.33 176	84.41 212	95.94 186	95.10 183	99.52 157	97.06 205

CR-MVSNet		94.57 150	97.34 110	91.33 176	94.90 142	98.59 137	97.15 124	79.14 207	97.98 150	80.42 177	96.59 113	93.50 129	96.85 129	98.10 103	97.49 119	99.50 159	99.15 164

RPMNet		94.66 144	97.16 118	91.75 169	94.98 141	98.59 137	97.00 131	78.37 213	97.98 150	83.78 154	96.27 118	94.09 125	96.91 127	97.36 146	96.73 138	99.48 160	99.09 169

CNLPA		99.03 28	99.05 45	99.01 21	99.27 45	99.22 99	99.03 49	97.98 34	99.34 31	99.00 4	98.25 70	99.71 50	99.31 35	98.80 61	98.82 52	99.48 160	99.17 163

CANet_DTU		96.64 104	99.08 42	93.81 130	97.10 83	99.42 72	98.85 58	90.01 140	99.31 34	79.98 181	99.78 2	99.10 62	97.42 117	98.35 92	98.05 95	99.47 162	99.53 132

MIMVSNet1		88.61 205	90.68 207	86.19 206	81.56 218	95.30 214	87.78 213	85.98 183	94.19 212	72.30 211	78.84 214	78.90 213	90.06 206	96.59 164	95.47 172	99.46 163	95.49 212

gm-plane-assit		89.44 204	92.82 201	85.49 207	91.37 199	95.34 213	79.55 221	82.12 196	91.68 217	64.79 219	87.98 193	80.26 205	95.66 160	98.51 87	97.56 115	99.45 164	98.41 186

PatchMatch-RL		97.77 64	98.25 73	97.21 62	99.11 48	99.25 95	97.06 130	94.09 72	98.72 117	95.14 56	98.47 62	96.29 94	98.43 87	98.65 72	97.44 124	99.45 164	98.94 173

pmmvs4		95.09 135	95.90 151	94.14 124	92.29 173	97.70 175	95.45 160	90.31 137	98.60 120	90.70 118	93.25 155	89.90 146	96.67 135	97.13 155	95.42 174	99.44 166	99.28 156

TAMVS		95.53 127	96.50 140	94.39 122	93.86 154	99.03 106	96.67 136	89.55 150	97.33 172	90.64 119	93.02 161	91.58 139	96.21 146	97.72 132	97.43 125	99.43 167	99.36 153

PMMVS		97.52 72	98.39 69	96.51 87	95.82 112	98.73 128	97.80 101	93.05 97	98.76 113	94.39 73	99.07 35	97.03 88	98.55 83	98.31 94	97.61 113	99.43 167	99.21 162

baseline		97.45 75	98.70 62	95.99 100	95.89 109	99.36 81	98.29 83	91.37 119	99.21 49	92.99 98	98.40 65	96.87 89	97.96 100	98.60 79	98.60 61	99.42 169	99.86 19

CVMVSNet		95.33 133	97.09 119	93.27 146	95.23 136	98.39 152	95.49 159	92.58 99	97.71 164	83.00 163	94.44 143	93.28 130	93.92 193	97.79 126	98.54 65	99.41 170	99.45 145

MAR-MVS		97.71 66	98.04 85	97.32 57	99.35 42	98.91 114	97.65 107	91.68 111	98.00 149	97.01 33	97.72 85	94.83 112	98.85 70	98.44 90	98.86 45	99.41 170	99.52 135
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
DPM-MVS		98.31 48	98.53 65	98.05 42	98.76 56	98.77 121	99.13 41	98.07 31	99.10 69	94.27 75	96.70 106	99.84 42	98.70 74	97.90 121	98.11 92	99.40 172	99.28 156

MIMVSNet		94.49 151	97.59 99	90.87 185	91.74 186	98.70 130	94.68 179	78.73 211	97.98 150	83.71 157	97.71 86	94.81 113	96.96 126	97.97 117	97.92 99	99.40 172	98.04 193

TSAR-MVS + COLMAP		96.79 95	96.55 135	97.06 66	97.70 71	98.46 145	99.07 46	96.23 45	99.38 24	91.32 116	98.80 47	85.61 174	98.69 76	97.64 137	96.92 135	99.37 174	99.06 171

pmmvs-eth3d		89.81 202	89.65 209	90.00 192	86.94 213	95.38 212	91.08 200	86.39 180	94.57 210	82.27 168	83.03 210	64.94 220	93.96 191	96.57 166	93.82 199	99.35 175	99.24 160

MS-PatchMatch		95.99 118	97.26 116	94.51 119	97.46 73	98.76 124	97.27 118	86.97 174	99.09 70	89.83 124	93.51 152	97.78 78	96.18 148	97.53 141	95.71 170	99.35 175	98.41 186

test-mter		94.86 141	97.32 112	92.00 162	92.41 171	98.82 117	96.18 149	86.35 181	98.05 147	82.28 167	96.48 115	94.39 119	95.46 169	98.17 101	96.20 155	99.32 177	99.13 168

Anonymous20231206		90.70 199	93.93 184	86.92 204	90.21 209	96.79 205	90.30 206	86.61 179	96.05 200	69.25 213	88.46 189	84.86 181	85.86 211	97.11 156	96.47 148	99.30 178	97.80 197

test_part1		95.56 126	95.38 158	95.78 102	96.07 103	98.16 161	97.57 108	90.78 130	97.43 169	93.04 96	89.12 184	89.41 149	97.93 101	96.38 171	97.38 127	99.29 179	99.78 48

EPNet		98.05 56	98.86 56	97.10 64	99.02 50	99.43 71	98.47 71	94.73 58	99.05 78	95.62 47	98.93 41	97.62 81	95.48 167	98.59 81	98.55 63	99.29 179	99.84 23
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
PatchT		93.96 159	97.36 109	90.00 192	94.76 146	98.65 132	90.11 207	78.57 212	97.96 153	80.42 177	96.07 121	94.10 124	96.85 129	98.10 103	97.49 119	99.26 181	99.15 164

EU-MVSNet		92.80 177	94.76 167	90.51 187	91.88 181	96.74 207	92.48 197	88.69 158	96.21 195	79.00 186	91.51 165	87.82 154	91.83 205	95.87 187	96.27 152	99.21 182	98.92 177

test-LLR		95.50 128	97.32 112	93.37 143	95.49 129	98.74 126	96.44 144	90.82 128	98.18 142	82.75 164	96.60 111	94.67 115	95.54 165	98.09 105	96.00 161	99.20 183	98.93 174

TESTMET0.1,1		94.95 138	97.32 112	92.20 157	92.62 166	98.74 126	96.44 144	86.67 177	98.18 142	82.75 164	96.60 111	94.67 115	95.54 165	98.09 105	96.00 161	99.20 183	98.93 174

PM-MVS		89.55 203	90.30 208	88.67 199	87.06 212	95.60 211	90.88 202	84.51 192	96.14 197	75.75 196	86.89 201	63.47 223	94.64 182	96.85 161	93.89 198	99.17 185	99.29 155

FMVSNet5		95.42 129	96.47 141	94.20 123	92.26 174	95.99 210	95.66 155	87.15 173	97.87 157	93.46 90	96.68 107	93.79 126	97.52 113	97.10 157	97.21 130	99.11 186	96.62 210

IterMVS-SCA-FT		94.89 140	97.87 92	91.42 173	94.86 144	97.70 175	97.24 120	84.88 189	98.93 89	75.74 197	94.26 144	98.25 73	96.69 133	98.52 85	97.68 111	99.10 187	99.73 76

GG-mvs-BLEND		69.11 214	98.13 81	35.26 219	3.49 228	98.20 160	94.89 170	2.38 225	98.42 132	5.82 229	96.37 117	98.60 67	5.97 224	98.75 67	97.98 97	99.01 188	98.61 181

CHOSEN 280x420		97.99 58	99.24 35	96.53 85	98.34 61	99.61 38	98.36 80	89.80 146	99.27 40	95.08 57	99.81 1	98.58 68	98.64 78	99.02 45	98.92 40	98.93 189	99.48 143

IterMVS		94.81 142	97.71 95	91.42 173	94.83 145	97.63 182	97.38 113	85.08 186	98.93 89	75.67 198	94.02 145	97.64 79	96.66 136	98.45 88	97.60 114	98.90 190	99.72 87
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
pmmvs6		91.90 195	92.53 202	91.17 179	91.81 184	97.63 182	93.23 192	88.37 163	93.43 214	80.61 175	77.32 215	87.47 155	94.12 188	96.58 165	95.72 169	98.88 191	99.53 132

MVS-HIRNet		92.51 183	95.97 149	88.48 200	93.73 158	98.37 153	90.33 205	75.36 219	98.32 136	77.78 191	89.15 182	94.87 111	95.14 177	97.62 138	96.39 149	98.51 192	97.11 203

new_pmnet		90.45 201	92.84 200	87.66 201	88.96 210	96.16 209	88.71 212	84.66 190	97.56 166	71.91 212	85.60 205	86.58 166	93.28 198	96.07 182	93.54 201	98.46 193	94.39 214

new-patchmatchnet		86.12 209	87.30 211	84.74 208	86.92 214	95.19 215	83.57 218	84.42 193	92.67 215	65.66 216	80.32 212	64.72 221	89.41 207	92.33 211	89.21 213	98.43 194	96.69 208

EPNet_dtu		96.30 111	98.53 65	93.70 134	98.97 51	98.24 158	97.36 114	94.23 71	98.85 96	79.18 185	99.19 22	98.47 70	94.09 189	97.89 122	98.21 86	98.39 195	98.85 179
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
tpm		92.38 189	94.79 166	89.56 196	94.30 149	97.50 192	94.24 189	78.97 210	97.72 163	74.93 202	97.97 78	82.91 190	96.60 138	93.65 205	94.81 191	98.33 196	98.98 172

pmmvs3		88.19 206	91.27 205	84.60 209	85.60 215	93.66 216	85.68 216	81.13 197	92.36 216	63.66 221	89.51 179	77.10 216	93.22 199	96.37 172	92.40 204	98.30 197	97.46 199

pmnet_mix02		92.44 185	94.68 168	89.83 195	92.46 170	97.65 181	89.92 209	90.49 136	98.76 113	73.05 208	91.78 164	90.08 145	94.86 181	94.53 201	91.94 208	98.21 198	98.01 195

N_pmnet		92.21 193	94.60 170	89.42 197	91.88 181	97.38 198	89.15 211	89.74 147	97.89 156	73.75 204	87.94 194	92.23 135	93.85 194	96.10 181	93.20 202	98.15 199	97.43 200

MDTV_nov1_ep13_2view		92.44 185	95.66 155	88.68 198	91.05 204	97.92 168	92.17 198	79.64 203	98.83 101	76.20 195	91.45 166	93.51 128	95.04 178	95.68 189	93.70 200	97.96 200	98.53 183

MDTV_nov1_ep13		95.57 125	97.48 102	93.35 145	95.43 132	98.97 111	97.19 123	83.72 195	98.92 91	87.91 132	97.75 83	96.12 99	97.88 106	96.84 162	95.64 171	97.96 200	98.10 192

ADS-MVSNet		94.65 145	97.04 123	91.88 168	95.68 118	98.99 109	95.89 151	79.03 209	99.15 59	85.81 145	96.96 99	98.21 75	97.10 122	94.48 202	94.24 196	97.74 202	97.21 202

RPSCF		97.61 69	98.16 80	96.96 75	98.10 63	99.00 107	98.84 59	93.76 79	99.45 20	94.78 62	99.39 14	99.31 59	98.53 85	96.61 163	95.43 173	97.74 202	97.93 196

DeepPCF-MVS	97.74 3	98.34 46	99.46 13	97.04 67	98.82 53	99.33 89	96.28 146	97.47 40	99.58 9	94.70 63	98.99 37	99.85 41	97.24 120	99.55 11	99.34 9	97.73 204	99.56 128

SCA		94.95 138	97.44 105	92.04 160	95.55 126	99.16 102	96.26 147	79.30 206	99.02 81	85.73 146	98.18 71	97.13 86	97.69 110	96.03 183	94.91 187	97.69 205	97.65 198

EPMVS		95.05 136	96.86 128	92.94 150	95.84 111	98.96 112	96.68 135	79.87 202	99.05 78	90.15 120	97.12 97	95.99 101	97.49 115	95.17 194	94.75 192	97.59 206	96.96 206

PMMVS2		77.26 212	79.47 215	74.70 215	76.00 221	88.37 220	74.22 222	76.34 216	78.31 220	54.13 223	69.96 217	52.50 225	70.14 219	84.83 215	88.71 214	97.35 207	93.58 216

PatchmatchNet		94.70 143	97.08 121	91.92 165	95.53 127	98.85 116	95.77 153	79.54 204	98.95 86	85.98 143	98.52 58	96.45 90	97.39 118	95.32 191	94.09 197	97.32 208	97.38 201
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
tpmrst		93.86 162	95.88 152	91.50 172	95.69 117	98.62 134	95.64 156	79.41 205	98.80 106	83.76 156	95.63 132	96.13 98	97.25 119	92.92 206	92.31 205	97.27 209	96.74 207

dps		94.63 146	95.31 161	93.84 129	95.53 127	98.71 129	96.54 139	80.12 201	97.81 162	97.21 30	96.98 98	92.37 133	96.34 145	92.46 209	91.77 209	97.26 210	97.08 204

CostFormer		94.25 154	94.88 164	93.51 140	95.43 132	98.34 155	96.21 148	80.64 199	97.94 154	94.01 76	98.30 69	86.20 171	97.52 113	92.71 207	92.69 203	97.23 211	98.02 194

tpm cat1		94.06 155	94.90 163	93.06 148	95.42 134	98.52 142	96.64 137	80.67 198	97.82 160	92.63 102	93.39 154	95.00 110	96.06 152	91.36 212	91.58 211	96.98 212	96.66 209

MDA-MVSNet-bldmvs		87.84 207	89.22 210	86.23 205	81.74 217	96.77 206	83.74 217	89.57 149	94.50 211	72.83 210	96.64 109	64.47 222	92.71 202	81.43 217	92.28 206	96.81 213	98.47 185

test_method		87.27 208	91.58 204	82.25 211	75.65 222	87.52 221	86.81 215	72.60 220	97.51 167	73.20 207	85.07 206	79.97 207	88.69 208	97.31 148	95.24 178	96.53 214	98.41 186

CMPMVS	70.31 18	90.74 198	91.06 206	90.36 190	97.32 76	97.43 195	92.97 194	87.82 170	93.50 213	75.34 201	83.27 209	84.90 180	92.19 204	92.64 208	91.21 212	96.50 215	94.46 213
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
Gipuma		81.40 211	81.78 213	80.96 213	83.21 216	85.61 222	79.73 220	76.25 218	97.33 172	64.21 220	55.32 219	55.55 224	86.04 210	92.43 210	92.20 207	96.32 216	93.99 215
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
ambc			80.99 214		80.04 220	90.84 217	90.91 201		96.09 198	74.18 203	62.81 218	30.59 229	82.44 214	96.25 179	91.77 209	95.91 217	98.56 182

DeepMVS_CX						96.85 204	87.43 214	89.27 151	98.30 137	75.55 199	95.05 135	79.47 210	92.62 203	89.48 213		95.18 218	95.96 211

tmp_tt				82.25 211	97.73 70	88.71 219	80.18 219	68.65 222	99.15 59	86.98 138	99.47 10	85.31 177	68.35 220	87.51 214	83.81 216	91.64 219

FPMVS		83.82 210	84.61 212	82.90 210	90.39 208	90.71 218	90.85 203	84.10 194	95.47 208	65.15 217	83.44 208	74.46 218	75.48 215	81.63 216	79.42 218	91.42 220	87.14 218

PMVS	72.60 17	76.39 213	77.66 216	74.92 214	81.04 219	69.37 226	68.47 223	80.54 200	85.39 219	65.07 218	73.52 216	72.91 219	65.67 221	80.35 218	76.81 219	88.71 221	85.25 221
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
E-PMN		68.30 215	68.43 217	68.15 216	74.70 224	71.56 225	55.64 225	77.24 214	77.48 222	39.46 225	51.95 222	41.68 227	73.28 217	70.65 220	79.51 217	88.61 222	86.20 220

EMVS		68.12 216	68.11 218	68.14 217	75.51 223	71.76 224	55.38 226	77.20 215	77.78 221	37.79 226	53.59 220	43.61 226	74.72 216	67.05 221	76.70 220	88.27 223	86.24 219

MVE	67.97 19	65.53 217	67.43 219	63.31 218	59.33 225	74.20 223	53.09 227	70.43 221	66.27 223	43.13 224	45.98 223	30.62 228	70.65 218	79.34 219	86.30 215	83.25 224	89.33 217
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
testmvs		31.24 218	40.15 220	20.86 220	12.61 226	17.99 227	25.16 228	13.30 223	48.42 224	24.82 227	53.07 221	30.13 230	28.47 222	42.73 222	37.65 221	20.79 225	51.04 222

test123		26.75 219	34.25 221	18.01 221	7.93 227	17.18 228	24.85 229	12.36 224	44.83 225	16.52 228	41.80 224	18.10 231	28.29 223	33.08 223	34.79 222	18.10 226	49.95 223

uanet_test		0.00 220	0.00 222	0.00 222	0.00 229	0.00 229	0.00 230	0.00 226	0.00 226	0.00 230	0.00 225	0.00 232	0.00 225	0.00 224	0.00 223	0.00 227	0.00 224

sosnet-low-res		0.00 220	0.00 222	0.00 222	0.00 229	0.00 229	0.00 230	0.00 226	0.00 226	0.00 230	0.00 225	0.00 232	0.00 225	0.00 224	0.00 223	0.00 227	0.00 224

sosnet		0.00 220	0.00 222	0.00 222	0.00 229	0.00 229	0.00 230	0.00 226	0.00 226	0.00 230	0.00 225	0.00 232	0.00 225	0.00 224	0.00 223	0.00 227	0.00 224

RE-MVS-def										69.05 214

9.14												99.79 46

SR-MVS					99.67 14			98.25 15				99.94 26

our_test_3					92.30 172	97.58 187	90.09 208

MTAPA										98.09 16		99.97 8

MTMP										98.46 11		99.96 13

Patchmatch-RL test							66.86 224

mPP-MVS					99.53 31							99.89 35

NP-MVS									98.57 123

Patchmtry						98.59 137	97.15 124	79.14 207		80.42 177

High-resolution multi-view benchmark