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.

DELS-MVS		98.19 49	98.77 57	97.52 51	98.29 61	99.71 9	99.12 41	94.58 63	98.80 100	95.38 48	96.24 115	98.24 71	97.92 95	99.06 38	99.52 1	99.82 10	99.79 38
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
DeepC-MVS	97.63 4	98.33 46	98.57 60	98.04 42	98.62 57	99.65 17	99.45 25	98.15 24	99.51 16	92.80 95	95.74 125	96.44 89	99.46 21	99.37 19	99.50 2	99.78 28	99.81 28
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
3Dnovator	96.92 7	98.67 37	99.05 42	98.23 38	99.57 27	99.45 61	99.11 42	94.66 59	99.69 3	96.80 33	96.55 110	99.61 52	99.40 25	98.87 52	99.49 3	99.85 3	99.66 102

MSLP-MVS++		99.15 18	99.24 31	99.04 15	99.52 32	99.49 56	99.09 44	98.07 30	99.37 25	98.47 9	97.79 77	99.89 34	99.50 17	98.93 45	99.45 4	99.61 116	99.76 57

IS_MVSNet		97.86 58	98.86 53	96.68 74	96.02 100	99.72 6	98.35 75	93.37 85	98.75 110	94.01 72	96.88 99	98.40 68	98.48 79	99.09 35	99.42 5	99.83 8	99.80 30

Vis-MVSNet (Re-imp)		97.40 74	98.89 52	95.66 101	95.99 103	99.62 29	97.82 93	93.22 88	98.82 97	91.40 109	96.94 96	98.56 66	95.70 152	99.14 33	99.41 6	99.79 25	99.75 64

PHI-MVS		99.08 22	99.43 18	98.67 29	99.15 46	99.59 41	99.11 42	97.35 40	99.14 55	97.30 27	99.44 11	99.96 12	99.32 30	98.89 50	99.39 7	99.79 25	99.58 115

APD-MVS		99.25 12	99.38 20	99.09 11	99.69 8	99.58 44	99.56 17	98.32 7	98.85 90	97.87 20	98.91 39	99.92 28	99.30 33	99.45 15	99.38 8	99.79 25	99.58 115
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
DeepPCF-MVS	97.74 3	98.34 45	99.46 12	97.04 63	98.82 52	99.33 82	96.28 139	97.47 39	99.58 8	94.70 59	98.99 33	99.85 40	97.24 113	99.55 10	99.34 9	97.73 197	99.56 121

DeepC-MVS_fast	98.34 1	99.17 17	99.45 13	98.85 25	99.55 29	99.37 73	99.64 8	98.05 32	99.53 13	96.58 35	98.93 37	99.92 28	99.49 19	99.46 14	99.32 10	99.80 24	99.64 109
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
SMA-MVS		99.38 5	99.60 2	99.12 9	99.76 2	99.62 29	99.39 29	98.23 19	99.52 15	98.03 17	99.45 10	99.98 1	99.64 5	99.58 8	99.30 11	99.68 88	99.76 57
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
3Dnovator+	96.92 7	98.71 36	99.05 42	98.32 34	99.53 30	99.34 79	99.06 46	94.61 60	99.65 4	97.49 24	96.75 100	99.86 37	99.44 23	98.78 57	99.30 11	99.81 16	99.67 98

QAPM		98.62 40	99.04 45	98.13 39	99.57 27	99.48 57	99.17 38	94.78 56	99.57 9	96.16 38	96.73 101	99.80 43	99.33 29	98.79 56	99.29 13	99.75 39	99.64 109

APDe-MVS		99.49 1	99.64 1	99.32 2	99.74 4	99.74 5	99.75 1	98.34 4	99.56 10	98.72 7	99.57 6	99.97 7	99.53 16	99.65 2	99.25 14	99.84 5	99.77 52

ACMMPR		99.30 9	99.54 6	99.03 16	99.66 16	99.64 22	99.68 4	98.25 14	99.56 10	97.12 30	99.19 19	99.95 17	99.72 1	99.43 16	99.25 14	99.72 57	99.77 52

HFP-MVS		99.32 7	99.53 8	99.07 13	99.69 8	99.59 41	99.63 11	98.31 8	99.56 10	97.37 26	99.27 16	99.97 7	99.70 3	99.35 21	99.24 16	99.71 67	99.76 57

UA-Net		97.13 80	99.14 35	94.78 109	97.21 79	99.38 71	97.56 102	92.04 98	98.48 123	88.03 124	98.39 62	99.91 31	94.03 183	99.33 23	99.23 17	99.81 16	99.25 152

LS3D		97.79 59	98.25 70	97.26 57	98.40 59	99.63 25	99.53 18	98.63 1	99.25 42	88.13 123	96.93 97	94.14 119	99.19 38	99.14 33	99.23 17	99.69 79	99.42 140

X-MVS		98.93 29	99.37 21	98.42 32	99.67 13	99.62 29	99.60 15	98.15 24	99.08 65	93.81 78	98.46 59	99.95 17	99.59 10	99.49 13	99.21 19	99.68 88	99.75 64

PGM-MVS		98.86 31	99.35 25	98.29 35	99.77 1	99.63 25	99.67 5	95.63 46	98.66 113	95.27 49	99.11 25	99.82 42	99.67 4	99.33 23	99.19 20	99.73 51	99.74 68

SteuartSystems-ACMMP		99.20 15	99.51 10	98.83 27	99.66 16	99.66 15	99.71 3	98.12 28	99.14 55	96.62 34	99.16 21	99.98 1	99.12 45	99.63 3	99.19 20	99.78 28	99.83 22
Skip Steuart: Steuart Systems R&D Blog.
TSAR-MVS + MP.		99.27 10	99.57 4	98.92 23	98.78 54	99.53 50	99.72 2	98.11 29	99.73 2	97.43 25	99.15 22	99.96 12	99.59 10	99.73 1	99.07 22	99.88 1	99.82 23
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
SD-MVS		99.25 12	99.50 11	98.96 21	98.79 53	99.55 48	99.33 32	98.29 11	99.75 1	97.96 19	99.15 22	99.95 17	99.61 6	99.17 31	99.06 23	99.81 16	99.84 18
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
DVP-MVS		99.45 2	99.54 6	99.35 1	99.72 7	99.76 1	99.63 11	98.37 2	99.63 6	99.03 3	98.95 36	99.98 1	99.60 7	99.60 6	99.05 24	99.74 44	99.79 38
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
MSP-MVS		99.34 6	99.52 9	99.14 8	99.68 12	99.75 4	99.64 8	98.31 8	99.44 20	98.10 14	99.28 15	99.98 1	99.30 33	99.34 22	99.05 24	99.81 16	99.79 38
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
canonicalmvs		97.31 75	97.81 91	96.72 73	96.20 98	99.45 61	98.21 81	91.60 107	99.22 44	95.39 47	98.48 57	90.95 136	99.16 43	97.66 127	99.05 24	99.76 35	99.90 3

OpenMVS	96.23 11	97.95 57	98.45 65	97.35 52	99.52 32	99.42 66	98.91 53	94.61 60	98.87 87	92.24 104	94.61 136	99.05 61	99.10 47	98.64 67	99.05 24	99.74 44	99.51 132

Vis-MVSNet		96.16 109	98.22 74	93.75 124	95.33 129	99.70 11	97.27 111	90.85 121	98.30 130	85.51 141	95.72 127	96.45 87	93.69 189	98.70 64	99.00 28	99.84 5	99.69 92
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
CANet		98.46 42	99.16 34	97.64 49	98.48 58	99.64 22	99.35 31	94.71 58	99.53 13	95.17 51	97.63 83	99.59 53	98.38 81	98.88 51	98.99 29	99.74 44	99.86 15

CDPH-MVS		98.41 43	99.10 38	97.61 50	99.32 43	99.36 74	99.49 21	96.15 45	98.82 97	91.82 106	98.41 60	99.66 51	99.10 47	98.93 45	98.97 30	99.75 39	99.58 115

DPE-MVS		99.39 4	99.55 5	99.20 4	99.63 21	99.71 9	99.66 6	98.33 6	99.29 34	98.40 12	99.64 4	99.98 1	99.31 31	99.56 9	98.96 31	99.85 3	99.70 88

TSAR-MVS + ACMM		98.77 33	99.45 13	97.98 44	99.37 37	99.46 59	99.44 27	98.13 27	99.65 4	92.30 102	98.91 39	99.95 17	99.05 50	99.42 17	98.95 32	99.58 134	99.82 23

EPP-MVSNet		97.75 62	98.71 58	96.63 78	95.68 115	99.56 47	97.51 103	93.10 92	99.22 44	94.99 55	97.18 92	97.30 81	98.65 71	98.83 53	98.93 33	99.84 5	99.92 1

CHOSEN 280x420		97.99 56	99.24 31	96.53 80	98.34 60	99.61 34	98.36 74	89.80 140	99.27 37	95.08 53	99.81 1	98.58 65	98.64 72	99.02 40	98.92 34	98.93 182	99.48 136

CSCG		98.90 30	98.93 51	98.85 25	99.75 3	99.72 6	99.49 21	96.58 43	99.38 23	98.05 16	98.97 34	97.87 74	99.49 19	97.78 120	98.92 34	99.78 28	99.90 3

CHOSEN 1792x2688		96.41 102	96.99 119	95.74 99	98.01 66	99.72 6	97.70 99	90.78 124	99.13 60	90.03 116	87.35 190	95.36 103	98.33 82	98.59 74	98.91 36	99.59 130	99.87 12

MVS_111021_LR		98.67 37	99.41 19	97.81 47	99.37 37	99.53 50	98.51 66	95.52 48	99.27 37	94.85 56	99.56 7	99.69 50	99.04 51	99.36 20	98.88 37	99.60 124	99.58 115

MVS_0304		98.14 51	99.03 46	97.10 60	98.05 65	99.63 25	99.27 34	94.33 65	99.63 6	93.06 90	97.32 86	99.05 61	98.09 88	98.82 54	98.87 38	99.81 16	99.89 6

CP-MVS		99.27 10	99.44 16	99.08 12	99.62 23	99.58 44	99.53 18	98.16 22	99.21 46	97.79 21	99.15 22	99.96 12	99.59 10	99.54 11	98.86 39	99.78 28	99.74 68

MAR-MVS		97.71 63	98.04 82	97.32 53	99.35 41	98.91 107	97.65 100	91.68 105	98.00 142	97.01 31	97.72 81	94.83 109	98.85 65	98.44 83	98.86 39	99.41 163	99.52 128
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
xxxxxxxxxxxxxcwj		98.14 51	97.38 105	99.03 16	99.65 18	99.41 68	98.87 54	98.24 17	99.14 55	98.73 5	99.11 25	86.38 162	98.92 58	99.22 27	98.84 41	99.76 35	99.56 121

SF-MVS		99.18 16	99.32 26	99.03 16	99.65 18	99.41 68	98.87 54	98.24 17	99.14 55	98.73 5	99.11 25	99.92 28	98.92 58	99.22 27	98.84 41	99.76 35	99.56 121

SED-MVS		99.44 3	99.58 3	99.28 3	99.69 8	99.76 1	99.62 14	98.35 3	99.51 16	99.05 2	99.60 5	99.98 1	99.28 35	99.61 5	98.83 43	99.70 76	99.77 52

MVS_111021_HR		98.59 41	99.36 22	97.68 48	99.42 35	99.61 34	98.14 84	94.81 55	99.31 31	95.00 54	99.51 8	99.79 45	99.00 54	98.94 44	98.83 43	99.69 79	99.57 120

CNLPA		99.03 27	99.05 42	99.01 20	99.27 44	99.22 92	99.03 48	97.98 33	99.34 29	99.00 4	98.25 66	99.71 49	99.31 31	98.80 55	98.82 45	99.48 153	99.17 156

FMVSNet2		96.64 98	97.50 97	95.63 102	93.81 148	97.98 157	98.09 86	90.87 120	98.99 77	93.48 84	93.17 152	95.25 104	97.89 96	98.63 68	98.80 46	99.68 88	99.67 98

MP-MVS		99.07 23	99.36 22	98.74 28	99.63 21	99.57 46	99.66 6	98.25 14	99.00 76	95.62 43	98.97 34	99.94 25	99.54 15	99.51 12	98.79 47	99.71 67	99.73 72
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
CS-MVS		98.06 53	99.12 36	96.82 72	95.83 108	99.66 15	98.93 52	93.12 91	98.95 79	94.29 69	98.55 54	99.05 61	98.94 56	99.05 39	98.78 48	99.83 8	99.80 30

ETV-MVS		98.05 54	99.25 30	96.65 76	95.61 117	99.61 34	98.26 80	93.52 81	98.90 86	93.74 81	99.32 14	99.20 58	98.90 61	99.21 29	98.72 49	99.87 2	99.79 38

TSAR-MVS + GP.		98.66 39	99.36 22	97.85 46	97.16 81	99.46 59	99.03 48	94.59 62	99.09 63	97.19 29	99.73 3	99.95 17	99.39 26	98.95 43	98.69 50	99.75 39	99.65 105

ACMMP_NAP		99.05 25	99.45 13	98.58 31	99.73 5	99.60 39	99.64 8	98.28 12	99.23 43	94.57 60	99.35 13	99.97 7	99.55 14	99.63 3	98.66 51	99.70 76	99.74 68

OMC-MVS		98.84 32	99.01 48	98.65 30	99.39 36	99.23 91	99.22 35	96.70 42	99.40 22	97.77 22	97.89 76	99.80 43	99.21 36	99.02 40	98.65 52	99.57 138	99.07 163

FMVSNet3		97.02 83	98.12 79	95.73 100	93.59 154	97.98 157	98.34 76	91.32 114	98.80 100	93.92 74	97.21 89	95.94 99	97.63 105	98.61 70	98.62 53	99.61 116	99.65 105

CNVR-MVS		99.23 14	99.28 28	99.17 5	99.65 18	99.34 79	99.46 24	98.21 20	99.28 35	98.47 9	98.89 41	99.94 25	99.50 17	99.42 17	98.61 54	99.73 51	99.52 128

baseline		97.45 72	98.70 59	95.99 94	95.89 105	99.36 74	98.29 77	91.37 113	99.21 46	92.99 93	98.40 61	96.87 86	97.96 93	98.60 72	98.60 55	99.42 162	99.86 15

zzz-MVS		99.31 8	99.44 16	99.16 6	99.73 5	99.65 17	99.63 11	98.26 13	99.27 37	98.01 18	99.27 16	99.97 7	99.60 7	99.59 7	98.58 56	99.71 67	99.73 72

MVS_Test		97.30 76	98.54 61	95.87 95	95.74 111	99.28 86	98.19 82	91.40 112	99.18 50	91.59 108	98.17 68	96.18 94	98.63 73	98.61 70	98.55 57	99.66 101	99.78 44

EPNet		98.05 54	98.86 53	97.10 60	99.02 49	99.43 65	98.47 67	94.73 57	99.05 71	95.62 43	98.93 37	97.62 78	95.48 160	98.59 74	98.55 57	99.29 172	99.84 18
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
CVMVSNet		95.33 126	97.09 115	93.27 139	95.23 130	98.39 145	95.49 152	92.58 95	97.71 157	83.00 156	94.44 139	93.28 126	93.92 186	97.79 119	98.54 59	99.41 163	99.45 138

casdiffmvs		96.93 86	97.43 103	96.34 85	95.70 113	99.50 55	97.75 97	93.22 88	98.98 78	92.64 96	94.97 132	91.71 134	98.93 57	98.62 69	98.52 60	99.82 10	99.72 83

PVSNet_Blended_VisFu		97.41 73	98.49 64	96.15 88	97.49 71	99.76 1	96.02 143	93.75 77	99.26 40	93.38 86	93.73 144	99.35 56	96.47 135	98.96 42	98.46 61	99.77 33	99.90 3

DCV-MVSNet		97.56 68	98.36 67	96.62 79	96.44 89	98.36 147	98.37 72	91.73 104	99.11 61	94.80 57	98.36 63	96.28 92	98.60 75	98.12 95	98.44 62	99.76 35	99.87 12

baseline1		97.58 67	98.05 81	97.02 66	96.21 97	99.45 61	97.71 98	93.71 79	98.47 124	95.75 42	98.78 45	93.20 128	98.91 60	98.52 78	98.44 62	99.81 16	99.53 125

NCCC		99.05 25	99.08 39	99.02 19	99.62 23	99.38 71	99.43 28	98.21 20	99.36 27	97.66 23	97.79 77	99.90 32	99.45 22	99.17 31	98.43 64	99.77 33	99.51 132

PVSNet_BlendedMVS		97.51 70	97.71 92	97.28 55	98.06 63	99.61 34	97.31 109	95.02 52	99.08 65	95.51 45	98.05 70	90.11 139	98.07 89	98.91 48	98.40 65	99.72 57	99.78 44

PVSNet_Blended		97.51 70	97.71 92	97.28 55	98.06 63	99.61 34	97.31 109	95.02 52	99.08 65	95.51 45	98.05 70	90.11 139	98.07 89	98.91 48	98.40 65	99.72 57	99.78 44

train_agg		98.73 35	99.11 37	98.28 36	99.36 39	99.35 77	99.48 23	97.96 34	98.83 95	93.86 77	98.70 51	99.86 37	99.44 23	99.08 37	98.38 67	99.61 116	99.58 115

CDS-MVSNet		96.59 101	98.02 84	94.92 108	94.45 141	98.96 105	97.46 105	91.75 103	97.86 151	90.07 115	96.02 118	97.25 82	96.21 139	98.04 106	98.38 67	99.60 124	99.65 105
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
HPM-MVS++		99.10 21	99.30 27	98.86 24	99.69 8	99.48 57	99.59 16	98.34 4	99.26 40	96.55 37	99.10 28	99.96 12	99.36 27	99.25 26	98.37 69	99.64 109	99.66 102

MCST-MVS		99.11 20	99.27 29	98.93 22	99.67 13	99.33 82	99.51 20	98.31 8	99.28 35	96.57 36	99.10 28	99.90 32	99.71 2	99.19 30	98.35 70	99.82 10	99.71 86

MSDG		98.27 48	98.29 69	98.24 37	99.20 45	99.22 92	99.20 36	97.82 36	99.37 25	94.43 65	95.90 121	97.31 80	99.12 45	98.76 59	98.35 70	99.67 96	99.14 160

test0.0.03 1		96.69 95	98.12 79	95.01 107	95.49 124	98.99 102	95.86 145	90.82 122	98.38 127	92.54 100	96.66 104	97.33 79	95.75 150	97.75 123	98.34 72	99.60 124	99.40 144

GBi-Net		96.98 84	98.00 85	95.78 96	93.81 148	97.98 157	98.09 86	91.32 114	98.80 100	93.92 74	97.21 89	95.94 99	97.89 96	98.07 101	98.34 72	99.68 88	99.67 98

test1		96.98 84	98.00 85	95.78 96	93.81 148	97.98 157	98.09 86	91.32 114	98.80 100	93.92 74	97.21 89	95.94 99	97.89 96	98.07 101	98.34 72	99.68 88	99.67 98

FMVSNet1		95.77 115	96.41 139	95.03 106	93.42 155	97.86 164	97.11 120	89.89 137	98.53 120	92.00 105	89.17 174	93.23 127	98.15 86	98.07 101	98.34 72	99.61 116	99.69 92

EIA-MVS		97.70 64	98.78 56	96.44 84	95.72 112	99.65 17	98.14 84	93.72 78	98.30 130	92.31 101	98.63 52	97.90 73	98.97 55	98.92 47	98.30 76	99.78 28	99.80 30

UGNet		97.66 65	99.07 41	96.01 93	97.19 80	99.65 17	97.09 121	93.39 83	99.35 28	94.40 67	98.79 44	99.59 53	94.24 180	98.04 106	98.29 77	99.73 51	99.80 30
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
IterMVS-LS		96.12 110	97.48 99	94.53 112	95.19 131	97.56 182	97.15 117	89.19 147	99.08 65	88.23 122	94.97 132	94.73 111	97.84 101	97.86 117	98.26 78	99.60 124	99.88 10
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
Anonymous202405211			97.40 104		96.45 88	99.54 49	98.08 89	93.79 74	98.24 134		93.55 145	94.41 115	98.88 64	98.04 106	98.24 79	99.75 39	99.76 57

EPNet_dtu		96.30 105	98.53 62	93.70 127	98.97 50	98.24 151	97.36 107	94.23 67	98.85 90	79.18 178	99.19 19	98.47 67	94.09 182	97.89 115	98.21 80	98.39 188	98.85 172
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
CPTT-MVS		99.14 19	99.20 33	99.06 14	99.58 26	99.53 50	99.45 25	97.80 37	99.19 49	98.32 13	98.58 53	99.95 17	99.60 7	99.28 25	98.20 81	99.64 109	99.69 92

HyFIR lowres test		95.99 112	96.56 127	95.32 104	97.99 67	99.65 17	96.54 132	88.86 149	98.44 125	89.77 119	84.14 199	97.05 84	99.03 52	98.55 76	98.19 82	99.73 51	99.86 15

diffmvs		96.83 88	97.33 108	96.25 86	95.76 110	99.34 79	98.06 90	93.22 88	99.43 21	92.30 102	96.90 98	89.83 144	98.55 76	98.00 109	98.14 83	99.64 109	99.70 88

TAPA-MVS	97.53 5	98.41 43	98.84 55	97.91 45	99.08 48	99.33 82	99.15 39	97.13 41	99.34 29	93.20 87	97.75 79	99.19 59	99.20 37	98.66 65	98.13 84	99.66 101	99.48 136
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
PLC	97.93 2	99.02 28	98.94 50	99.11 10	99.46 34	99.24 90	99.06 46	97.96 34	99.31 31	99.16 1	97.90 75	99.79 45	99.36 27	98.71 63	98.12 85	99.65 105	99.52 128
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
DPM-MVS		98.31 47	98.53 62	98.05 41	98.76 55	98.77 114	99.13 40	98.07 30	99.10 62	94.27 71	96.70 102	99.84 41	98.70 68	97.90 114	98.11 86	99.40 165	99.28 149

Anonymous20231211		97.10 81	97.06 117	97.14 59	96.32 91	99.52 53	98.16 83	93.76 75	98.84 94	95.98 40	90.92 162	94.58 114	98.90 61	97.72 125	98.10 87	99.71 67	99.75 64

gg-mvs-nofinetune		90.85 190	94.14 169	87.02 196	94.89 136	99.25 88	98.64 62	76.29 210	88.24 210	57.50 214	79.93 205	95.45 102	95.18 169	98.77 58	98.07 88	99.62 114	99.24 153

CANet_DTU		96.64 98	99.08 39	93.81 123	97.10 82	99.42 66	98.85 56	90.01 134	99.31 31	79.98 174	99.78 2	99.10 60	97.42 110	98.35 85	98.05 89	99.47 155	99.53 125

Fast-Effi-MVS+		95.38 124	96.52 130	94.05 120	94.15 143	99.14 97	97.24 113	86.79 168	98.53 120	87.62 128	94.51 137	87.06 151	98.76 66	98.60 72	98.04 90	99.72 57	99.77 52

GG-mvs-BLEND		69.11 206	98.13 78	35.26 211	3.49 220	98.20 153	94.89 163	2.38 217	98.42 126	5.82 221	96.37 113	98.60 64	5.97 216	98.75 61	97.98 91	99.01 181	98.61 174

Effi-MVS+		95.81 114	97.31 112	94.06 119	95.09 132	99.35 77	97.24 113	88.22 158	98.54 119	85.38 142	98.52 55	88.68 146	98.70 68	98.32 86	97.93 92	99.74 44	99.84 18

MIMVSNet		94.49 144	97.59 96	90.87 178	91.74 179	98.70 123	94.68 172	78.73 204	97.98 143	83.71 150	97.71 82	94.81 110	96.96 119	97.97 110	97.92 93	99.40 165	98.04 185

DI_MVS_plusplus_trai		96.90 87	97.49 98	96.21 87	95.61 117	99.40 70	98.72 61	92.11 96	99.14 55	92.98 94	93.08 155	95.14 105	98.13 87	98.05 105	97.91 94	99.74 44	99.73 72

testgi		95.67 117	97.48 99	93.56 130	95.07 133	99.00 100	95.33 156	88.47 155	98.80 100	86.90 132	97.30 87	92.33 130	95.97 147	97.66 127	97.91 94	99.60 124	99.38 145

thres100view900		96.72 93	96.47 134	97.00 69	96.31 92	99.52 53	98.28 78	94.01 69	97.35 162	94.52 61	95.90 121	86.93 154	99.09 49	98.07 101	97.87 96	99.81 16	99.63 111

COLMAP_ROB	96.15 12	97.78 60	98.17 76	97.32 53	98.84 51	99.45 61	99.28 33	95.43 49	99.48 18	91.80 107	94.83 135	98.36 69	98.90 61	98.09 98	97.85 97	99.68 88	99.15 157
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
AdaColmap		99.06 24	98.98 49	99.15 7	99.60 25	99.30 85	99.38 30	98.16 22	99.02 74	98.55 8	98.71 50	99.57 55	99.58 13	99.09 35	97.84 98	99.64 109	99.36 146

thres200		96.76 90	96.53 129	97.03 64	96.31 92	99.67 12	98.37 72	93.99 71	97.68 158	94.49 63	95.83 124	86.77 156	99.18 40	98.26 88	97.82 99	99.82 10	99.66 102

tfpn200view9		96.75 91	96.51 131	97.03 64	96.31 92	99.67 12	98.41 69	93.99 71	97.35 162	94.52 61	95.90 121	86.93 154	99.14 44	98.26 88	97.80 100	99.82 10	99.70 88

thres400		96.71 94	96.45 136	97.02 66	96.28 95	99.63 25	98.41 69	94.00 70	97.82 153	94.42 66	95.74 125	86.26 163	99.18 40	98.20 92	97.79 101	99.81 16	99.70 88

FC-MVSNet-train		97.04 82	97.91 88	96.03 92	96.00 102	98.41 143	96.53 134	93.42 82	99.04 73	93.02 92	98.03 72	94.32 117	97.47 109	97.93 112	97.77 102	99.75 39	99.88 10

baseline2		96.36 104	97.82 90	94.65 111	94.60 140	99.09 98	96.45 136	89.63 142	98.36 128	91.29 111	97.60 84	94.13 120	96.37 136	98.45 81	97.70 103	99.54 147	99.41 141

IterMVS-SCA-FT		94.89 133	97.87 89	91.42 166	94.86 137	97.70 168	97.24 113	84.88 182	98.93 83	75.74 190	94.26 140	98.25 70	96.69 126	98.52 78	97.68 104	99.10 180	99.73 72

thres600view7		96.69 95	96.43 138	97.00 69	96.28 95	99.67 12	98.41 69	93.99 71	97.85 152	94.29 69	95.96 119	85.91 166	99.19 38	98.26 88	97.63 105	99.82 10	99.73 72

PMMVS		97.52 69	98.39 66	96.51 82	95.82 109	98.73 121	97.80 94	93.05 93	98.76 107	94.39 68	99.07 31	97.03 85	98.55 76	98.31 87	97.61 106	99.43 160	99.21 155

IterMVS		94.81 135	97.71 92	91.42 166	94.83 138	97.63 175	97.38 106	85.08 179	98.93 83	75.67 191	94.02 141	97.64 76	96.66 129	98.45 81	97.60 107	98.90 183	99.72 83
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
Effi-MVS+-dtu		95.74 116	98.04 82	93.06 141	93.92 144	99.16 95	97.90 91	88.16 160	99.07 70	82.02 162	98.02 73	94.32 117	96.74 125	98.53 77	97.56 108	99.61 116	99.62 112

gm-plane-assit		89.44 197	92.82 194	85.49 200	91.37 192	95.34 206	79.55 213	82.12 189	91.68 209	64.79 211	87.98 186	80.26 198	95.66 153	98.51 80	97.56 108	99.45 157	98.41 179

LGP-MVS_train		96.23 106	96.89 121	95.46 103	97.32 75	98.77 114	98.81 58	93.60 80	98.58 116	85.52 140	99.08 30	86.67 158	97.83 102	97.87 116	97.51 110	99.69 79	99.73 72

ACMMP		98.74 34	99.03 46	98.40 33	99.36 39	99.64 22	99.20 36	97.75 38	98.82 97	95.24 50	98.85 42	99.87 36	99.17 42	98.74 62	97.50 111	99.71 67	99.76 57
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
CR-MVSNet		94.57 143	97.34 107	91.33 169	94.90 135	98.59 130	97.15 117	79.14 200	97.98 143	80.42 170	96.59 109	93.50 125	96.85 122	98.10 96	97.49 112	99.50 152	99.15 157

PatchT		93.96 152	97.36 106	90.00 185	94.76 139	98.65 125	90.11 200	78.57 205	97.96 146	80.42 170	96.07 117	94.10 121	96.85 122	98.10 96	97.49 112	99.26 174	99.15 157

FC-MVSNet-test		96.07 111	97.94 87	93.89 121	93.60 153	98.67 124	96.62 131	90.30 133	98.76 107	88.62 120	95.57 130	97.63 77	94.48 176	97.97 110	97.48 114	99.71 67	99.52 128

UniMVSNet_ETH3D		93.15 163	92.33 196	94.11 118	93.91 145	98.61 129	94.81 167	90.98 119	97.06 171	87.51 129	82.27 203	76.33 209	97.87 100	94.79 192	97.47 115	99.56 141	99.81 28

PCF-MVS	97.50 6	98.18 50	98.35 68	97.99 43	98.65 56	99.36 74	98.94 51	98.14 26	98.59 115	93.62 82	96.61 106	99.76 48	99.03 52	97.77 121	97.45 116	99.57 138	98.89 171
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
PatchMatch-RL		97.77 61	98.25 70	97.21 58	99.11 47	99.25 88	97.06 123	94.09 68	98.72 111	95.14 52	98.47 58	96.29 91	98.43 80	98.65 66	97.44 117	99.45 157	98.94 166

TAMVS		95.53 120	96.50 133	94.39 115	93.86 147	99.03 99	96.67 129	89.55 144	97.33 164	90.64 113	93.02 156	91.58 135	96.21 139	97.72 125	97.43 118	99.43 160	99.36 146

LTVRE_ROB	93.20 16	92.84 168	94.92 155	90.43 182	92.83 157	98.63 126	97.08 122	87.87 162	97.91 148	68.42 207	93.54 146	79.46 203	96.62 130	97.55 133	97.40 119	99.74 44	99.92 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
test_part1		95.56 119	95.38 151	95.78 96	96.07 99	98.16 154	97.57 101	90.78 124	97.43 161	93.04 91	89.12 177	89.41 145	97.93 94	96.38 163	97.38 120	99.29 172	99.78 44

MVSTER		97.16 79	97.71 92	96.52 81	95.97 104	98.48 136	98.63 63	92.10 97	98.68 112	95.96 41	99.23 18	91.79 133	96.87 121	98.76 59	97.37 121	99.57 138	99.68 97

Baseline_NR-MVSNet		93.87 154	93.98 176	93.75 124	91.66 181	97.02 195	95.53 151	91.52 111	97.16 170	87.77 127	87.93 188	83.69 178	96.35 137	95.10 188	97.23 122	99.68 88	99.73 72

FMVSNet5		95.42 122	96.47 134	94.20 116	92.26 167	95.99 203	95.66 148	87.15 166	97.87 150	93.46 85	96.68 103	93.79 123	97.52 106	97.10 149	97.21 123	99.11 179	96.62 202

pm-mvs1		94.27 145	95.57 149	92.75 144	92.58 160	98.13 155	94.87 165	90.71 127	96.70 181	83.78 147	89.94 170	89.85 143	94.96 173	97.58 132	97.07 124	99.61 116	99.72 83

Fast-Effi-MVS+-dtu		95.38 124	98.20 75	92.09 152	93.91 145	98.87 108	97.35 108	85.01 181	99.08 65	81.09 166	98.10 69	96.36 90	95.62 155	98.43 84	97.03 125	99.55 143	99.50 134

TransMVSNet (Re)		93.45 159	94.08 172	92.72 145	92.83 157	97.62 178	94.94 161	91.54 110	95.65 198	83.06 155	88.93 178	83.53 180	94.25 179	97.41 137	97.03 125	99.67 96	98.40 181

DU-MVS		93.98 151	94.44 166	93.44 134	91.66 181	97.77 165	95.03 158	91.57 108	97.17 168	86.12 134	93.13 153	81.13 194	96.60 131	95.10 188	97.01 127	99.67 96	99.80 30

TSAR-MVS + COLMAP		96.79 89	96.55 128	97.06 62	97.70 70	98.46 138	99.07 45	96.23 44	99.38 23	91.32 110	98.80 43	85.61 168	98.69 70	97.64 130	96.92 128	99.37 167	99.06 164

CLD-MVS		96.74 92	96.51 131	97.01 68	96.71 86	98.62 127	98.73 60	94.38 64	98.94 82	94.46 64	97.33 85	87.03 152	98.07 89	97.20 145	96.87 129	99.72 57	99.54 124
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
TranMVSNet+NR-MVSNet		93.67 157	94.14 169	93.13 140	91.28 195	97.58 180	95.60 150	91.97 100	97.06 171	84.05 143	90.64 167	82.22 189	96.17 142	94.94 191	96.78 130	99.69 79	99.78 44

RPMNet		94.66 137	97.16 114	91.75 162	94.98 134	98.59 130	97.00 124	78.37 206	97.98 143	83.78 147	96.27 114	94.09 122	96.91 120	97.36 139	96.73 131	99.48 153	99.09 162

UniMVSNet_NR-MVSNet		94.59 141	95.47 150	93.55 131	91.85 176	97.89 163	95.03 158	92.00 99	97.33 164	86.12 134	93.19 151	87.29 150	96.60 131	96.12 172	96.70 132	99.72 57	99.80 30

ET-MVSNet_ETH3D		96.17 108	96.99 119	95.21 105	88.53 204	98.54 133	98.28 78	92.61 94	98.85 90	93.60 83	99.06 32	90.39 138	98.63 73	95.98 177	96.68 133	99.61 116	99.41 141

ACMH	95.42 14	95.27 127	95.96 143	94.45 114	96.83 85	98.78 113	94.72 170	91.67 106	98.95 79	86.82 133	96.42 112	83.67 179	97.00 117	97.48 136	96.68 133	99.69 79	99.76 57
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
OPM-MVS		96.22 107	95.85 147	96.65 76	97.75 68	98.54 133	99.00 50	95.53 47	96.88 175	89.88 117	95.95 120	86.46 161	98.07 89	97.65 129	96.63 135	99.67 96	98.83 173

ACMP	96.25 10	96.62 100	96.72 124	96.50 83	96.96 84	98.75 118	97.80 94	94.30 66	98.85 90	93.12 89	98.78 45	86.61 159	97.23 114	97.73 124	96.61 136	99.62 114	99.71 86
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
ACMH+	95.51 13	95.40 123	96.00 141	94.70 110	96.33 90	98.79 111	96.79 127	91.32 114	98.77 106	87.18 130	95.60 129	85.46 169	96.97 118	97.15 146	96.59 137	99.59 130	99.65 105

CP-MVSNet		93.25 162	94.00 175	92.38 147	91.65 183	97.56 182	94.38 179	89.20 146	96.05 192	83.16 154	89.51 172	81.97 190	96.16 143	96.43 161	96.56 138	99.71 67	99.89 6

HQP-MVS		96.37 103	96.58 126	96.13 89	97.31 77	98.44 140	98.45 68	95.22 50	98.86 88	88.58 121	98.33 64	87.00 153	97.67 104	97.23 143	96.56 138	99.56 141	99.62 112

PS-CasMVS		92.72 173	93.36 187	91.98 156	91.62 185	97.52 184	94.13 183	88.98 148	95.94 195	81.51 165	87.35 190	79.95 200	95.91 148	96.37 164	96.49 140	99.70 76	99.89 6

Anonymous20231206		90.70 192	93.93 177	86.92 197	90.21 202	96.79 198	90.30 199	86.61 172	96.05 192	69.25 205	88.46 182	84.86 175	85.86 203	97.11 148	96.47 141	99.30 171	97.80 189

MVS-HIRNet		92.51 176	95.97 142	88.48 193	93.73 151	98.37 146	90.33 198	75.36 212	98.32 129	77.78 184	89.15 175	94.87 108	95.14 170	97.62 131	96.39 142	98.51 185	97.11 195

DTE-MVSNet		92.42 181	92.85 192	91.91 159	90.87 198	96.97 196	94.53 178	89.81 138	95.86 197	81.59 164	88.83 179	77.88 207	95.01 172	94.34 195	96.35 143	99.64 109	99.73 72

ACMM	96.26 9	96.67 97	96.69 125	96.66 75	97.29 78	98.46 138	96.48 135	95.09 51	99.21 46	93.19 88	98.78 45	86.73 157	98.17 83	97.84 118	96.32 144	99.74 44	99.49 135
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
EU-MVSNet		92.80 170	94.76 160	90.51 180	91.88 174	96.74 200	92.48 190	88.69 152	96.21 187	79.00 179	91.51 158	87.82 148	91.83 198	95.87 179	96.27 145	99.21 175	98.92 170

PEN-MVS		92.72 173	93.20 189	92.15 151	91.29 193	97.31 192	94.67 173	89.81 138	96.19 188	81.83 163	88.58 181	79.06 204	95.61 156	95.21 185	96.27 145	99.72 57	99.82 23

TinyColmap		94.00 150	94.35 167	93.60 128	95.89 105	98.26 149	97.49 104	88.82 150	98.56 118	83.21 153	91.28 161	80.48 197	96.68 127	97.34 140	96.26 147	99.53 149	98.24 182

test-mter		94.86 134	97.32 109	92.00 155	92.41 164	98.82 110	96.18 142	86.35 174	98.05 140	82.28 160	96.48 111	94.39 116	95.46 162	98.17 94	96.20 148	99.32 170	99.13 161

NR-MVSNet		94.01 149	94.51 164	93.44 134	92.56 161	97.77 165	95.67 147	91.57 108	97.17 168	85.84 137	93.13 153	80.53 196	95.29 166	97.01 150	96.17 149	99.69 79	99.75 64

tfpnnormal		93.85 156	94.12 171	93.54 132	93.22 156	98.24 151	95.45 153	91.96 101	94.61 201	83.91 145	90.74 164	81.75 192	97.04 116	97.49 135	96.16 150	99.68 88	99.84 18

USDC		94.26 146	94.83 158	93.59 129	96.02 100	98.44 140	97.84 92	88.65 153	98.86 88	82.73 159	94.02 141	80.56 195	96.76 124	97.28 142	96.15 151	99.55 143	98.50 177

thisisatest0530		97.23 77	98.25 70	96.05 90	95.60 119	99.59 41	96.96 125	93.23 86	99.17 51	92.60 98	98.75 48	96.19 93	98.17 83	98.19 93	96.10 152	99.72 57	99.77 52

tttt0517		97.23 77	98.24 73	96.04 91	95.60 119	99.60 39	96.94 126	93.23 86	99.15 52	92.56 99	98.74 49	96.12 96	98.17 83	98.21 91	96.10 152	99.73 51	99.78 44

test-LLR		95.50 121	97.32 109	93.37 136	95.49 124	98.74 119	96.44 137	90.82 122	98.18 135	82.75 157	96.60 107	94.67 112	95.54 158	98.09 98	96.00 154	99.20 176	98.93 167

TESTMET0.1,1		94.95 131	97.32 109	92.20 150	92.62 159	98.74 119	96.44 137	86.67 170	98.18 135	82.75 157	96.60 107	94.67 112	95.54 158	98.09 98	96.00 154	99.20 176	98.93 167

EG-PatchMatch MVS		92.45 177	93.92 178	90.72 179	92.56 161	98.43 142	94.88 164	84.54 184	97.18 167	79.55 176	86.12 197	83.23 183	93.15 193	97.22 144	96.00 154	99.67 96	99.27 151

UniMVSNet (Re)		94.58 142	95.34 152	93.71 126	92.25 168	98.08 156	94.97 160	91.29 118	97.03 173	87.94 125	93.97 143	86.25 164	96.07 144	96.27 169	95.97 157	99.72 57	99.79 38

anonymousdsp		93.12 164	95.86 146	89.93 187	91.09 196	98.25 150	95.12 157	85.08 179	97.44 160	73.30 198	90.89 163	90.78 137	95.25 168	97.91 113	95.96 158	99.71 67	99.82 23

WR-MVS_H		93.54 158	94.67 162	92.22 148	91.95 172	97.91 162	94.58 176	88.75 151	96.64 182	83.88 146	90.66 166	85.13 172	94.40 177	96.54 159	95.91 159	99.73 51	99.89 6

WR-MVS		93.43 161	94.48 165	92.21 149	91.52 188	97.69 170	94.66 174	89.98 135	96.86 176	83.43 151	90.12 168	85.03 173	93.94 185	96.02 176	95.82 160	99.71 67	99.82 23

IB-MVS	93.96 15	95.02 130	96.44 137	93.36 137	97.05 83	99.28 86	90.43 197	93.39 83	98.02 141	96.02 39	94.92 134	92.07 132	83.52 205	95.38 182	95.82 160	99.72 57	99.59 114
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
pmmvs6		91.90 188	92.53 195	91.17 172	91.81 177	97.63 175	93.23 185	88.37 157	93.43 206	80.61 168	77.32 207	87.47 149	94.12 181	96.58 157	95.72 162	98.88 184	99.53 125

MS-PatchMatch		95.99 112	97.26 113	94.51 113	97.46 72	98.76 117	97.27 111	86.97 167	99.09 63	89.83 118	93.51 147	97.78 75	96.18 141	97.53 134	95.71 163	99.35 168	98.41 179

MDTV_nov1_ep13		95.57 118	97.48 99	93.35 138	95.43 126	98.97 104	97.19 116	83.72 188	98.92 85	87.91 126	97.75 79	96.12 96	97.88 99	96.84 154	95.64 164	97.96 193	98.10 184

MIMVSNet1		88.61 198	90.68 199	86.19 199	81.56 211	95.30 207	87.78 206	85.98 176	94.19 204	72.30 203	78.84 206	78.90 205	90.06 199	96.59 156	95.47 165	99.46 156	95.49 204

RPSCF		97.61 66	98.16 77	96.96 71	98.10 62	99.00 100	98.84 57	93.76 75	99.45 19	94.78 58	99.39 12	99.31 57	98.53 78	96.61 155	95.43 166	97.74 195	97.93 188

pmmvs4		95.09 128	95.90 144	94.14 117	92.29 166	97.70 168	95.45 153	90.31 131	98.60 114	90.70 112	93.25 150	89.90 142	96.67 128	97.13 147	95.42 167	99.44 159	99.28 149

GA-MVS		93.93 153	96.31 140	91.16 173	93.61 152	98.79 111	95.39 155	90.69 128	98.25 133	73.28 199	96.15 116	88.42 147	94.39 178	97.76 122	95.35 168	99.58 134	99.45 138

v10		92.79 171	94.06 173	91.31 170	91.78 178	97.29 194	94.87 165	86.10 175	96.97 174	79.82 175	88.16 184	84.56 176	95.63 154	96.33 167	95.31 169	99.65 105	99.80 30

v1192		92.43 180	93.61 182	91.05 174	91.53 187	97.43 188	94.61 175	87.99 161	96.60 183	76.72 186	87.11 192	82.74 187	95.85 149	96.35 166	95.30 170	99.60 124	99.74 68

v1144		92.81 169	94.03 174	91.40 168	91.68 180	97.60 179	94.73 169	88.40 156	96.71 180	78.48 181	88.14 185	84.46 177	95.45 163	96.31 168	95.22 171	99.65 105	99.76 57

v1240		91.99 187	93.33 188	90.44 181	91.29 193	97.30 193	94.25 181	86.79 168	96.43 186	75.49 193	86.34 196	81.85 191	95.29 166	96.42 162	95.22 171	99.52 150	99.73 72

v144192		92.38 182	93.55 185	91.00 175	91.44 189	97.47 187	94.27 180	87.41 165	96.52 185	78.03 182	87.50 189	82.65 188	95.32 165	95.82 180	95.15 173	99.55 143	99.78 44

v1921920		92.36 184	93.57 183	90.94 176	91.39 191	97.39 190	94.70 171	87.63 164	96.60 183	76.63 187	86.98 193	82.89 185	95.75 150	96.26 170	95.14 174	99.55 143	99.73 72

test20.03		90.65 193	93.71 181	87.09 195	90.44 200	96.24 201	89.74 203	85.46 178	95.59 199	72.99 201	90.68 165	85.33 170	84.41 204	95.94 178	95.10 175	99.52 150	97.06 197

pmmvs5		92.71 175	94.27 168	90.90 177	91.42 190	97.74 167	93.23 185	86.66 171	95.99 194	78.96 180	91.45 159	83.44 181	95.55 157	97.30 141	95.05 176	99.58 134	98.93 167

v7n		91.61 189	92.95 190	90.04 184	90.56 199	97.69 170	93.74 184	85.59 177	95.89 196	76.95 185	86.60 195	78.60 206	93.76 188	97.01 150	94.99 177	99.65 105	99.87 12

v2v482		92.77 172	93.52 186	91.90 160	91.59 186	97.63 175	94.57 177	90.31 131	96.80 179	79.22 177	88.74 180	81.55 193	96.04 146	95.26 184	94.97 178	99.66 101	99.69 92

SCA		94.95 131	97.44 102	92.04 153	95.55 121	99.16 95	96.26 140	79.30 199	99.02 74	85.73 139	98.18 67	97.13 83	97.69 103	96.03 175	94.91 179	97.69 198	97.65 190

v8		92.87 167	93.87 180	91.72 164	92.05 170	97.50 185	94.79 168	88.20 159	96.85 177	80.11 173	90.01 169	82.86 186	95.48 160	95.15 187	94.90 180	99.66 101	99.80 30

V42		93.05 165	93.90 179	92.04 153	91.91 173	97.66 172	94.91 162	89.91 136	96.85 177	80.58 169	89.66 171	83.43 182	95.37 164	95.03 190	94.90 180	99.59 130	99.78 44

SixPastTwentyTwo		93.44 160	95.32 153	91.24 171	92.11 169	98.40 144	92.77 188	88.64 154	98.09 139	77.83 183	93.51 147	85.74 167	96.52 134	96.91 152	94.89 182	99.59 130	99.73 72

tpm		92.38 182	94.79 159	89.56 189	94.30 142	97.50 185	94.24 182	78.97 203	97.72 156	74.93 195	97.97 74	82.91 184	96.60 131	93.65 197	94.81 183	98.33 189	98.98 165

EPMVS		95.05 129	96.86 123	92.94 143	95.84 107	98.96 105	96.68 128	79.87 195	99.05 71	90.15 114	97.12 93	95.99 98	97.49 108	95.17 186	94.75 184	97.59 199	96.96 198

thisisatest0515		94.61 140	96.89 121	91.95 157	92.00 171	98.47 137	92.01 192	90.73 126	98.18 135	83.96 144	94.51 137	95.13 106	93.38 190	97.38 138	94.74 185	99.61 116	99.79 38

v148		92.36 184	92.88 191	91.75 162	91.63 184	97.66 172	92.64 189	90.55 129	96.09 190	83.34 152	88.19 183	80.00 199	92.74 194	93.98 196	94.58 186	99.58 134	99.69 92

TDRefinement		93.04 166	93.57 183	92.41 146	96.58 87	98.77 114	97.78 96	91.96 101	98.12 138	80.84 167	89.13 176	79.87 201	87.78 201	96.44 160	94.50 187	99.54 147	98.15 183

ADS-MVSNet		94.65 138	97.04 118	91.88 161	95.68 115	98.99 102	95.89 144	79.03 202	99.15 52	85.81 138	96.96 95	98.21 72	97.10 115	94.48 194	94.24 188	97.74 195	97.21 194

PatchmatchNet		94.70 136	97.08 116	91.92 158	95.53 122	98.85 109	95.77 146	79.54 197	98.95 79	85.98 136	98.52 55	96.45 87	97.39 111	95.32 183	94.09 189	97.32 201	97.38 193
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
PM-MVS		89.55 196	90.30 200	88.67 192	87.06 205	95.60 204	90.88 195	84.51 185	96.14 189	75.75 189	86.89 194	63.47 215	94.64 175	96.85 153	93.89 190	99.17 178	99.29 148

pmmvs-eth3d		89.81 195	89.65 201	90.00 185	86.94 206	95.38 205	91.08 193	86.39 173	94.57 202	82.27 161	83.03 202	64.94 212	93.96 184	96.57 158	93.82 191	99.35 168	99.24 153

MDTV_nov1_ep13_2view		92.44 178	95.66 148	88.68 191	91.05 197	97.92 161	92.17 191	79.64 196	98.83 95	76.20 188	91.45 159	93.51 124	95.04 171	95.68 181	93.70 192	97.96 193	98.53 176

new_pmnet		90.45 194	92.84 193	87.66 194	88.96 203	96.16 202	88.71 205	84.66 183	97.56 159	71.91 204	85.60 198	86.58 160	93.28 191	96.07 174	93.54 193	98.46 186	94.39 206

N_pmnet		92.21 186	94.60 163	89.42 190	91.88 174	97.38 191	89.15 204	89.74 141	97.89 149	73.75 197	87.94 187	92.23 131	93.85 187	96.10 173	93.20 194	98.15 192	97.43 192

CostFormer		94.25 147	94.88 157	93.51 133	95.43 126	98.34 148	96.21 141	80.64 192	97.94 147	94.01 72	98.30 65	86.20 165	97.52 106	92.71 199	92.69 195	97.23 204	98.02 186

pmmvs3		88.19 199	91.27 197	84.60 202	85.60 208	93.66 209	85.68 208	81.13 190	92.36 208	63.66 213	89.51 172	77.10 208	93.22 192	96.37 164	92.40 196	98.30 190	97.46 191

tpmrst		93.86 155	95.88 145	91.50 165	95.69 114	98.62 127	95.64 149	79.41 198	98.80 100	83.76 149	95.63 128	96.13 95	97.25 112	92.92 198	92.31 197	97.27 202	96.74 199

MDA-MVSNet-bldmvs		87.84 200	89.22 202	86.23 198	81.74 210	96.77 199	83.74 209	89.57 143	94.50 203	72.83 202	96.64 105	64.47 214	92.71 195	81.43 209	92.28 198	96.81 206	98.47 178

Gipuma		81.40 203	81.78 205	80.96 205	83.21 209	85.61 214	79.73 212	76.25 211	97.33 164	64.21 212	55.32 211	55.55 216	86.04 202	92.43 202	92.20 199	96.32 208	93.99 207
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
pmnet_mix02		92.44 178	94.68 161	89.83 188	92.46 163	97.65 174	89.92 202	90.49 130	98.76 107	73.05 200	91.78 157	90.08 141	94.86 174	94.53 193	91.94 200	98.21 191	98.01 187

ambc			80.99 206		80.04 213	90.84 210	90.91 194		96.09 190	74.18 196	62.81 210	30.59 221	82.44 206	96.25 171	91.77 201	95.91 209	98.56 175

dps		94.63 139	95.31 154	93.84 122	95.53 122	98.71 122	96.54 132	80.12 194	97.81 155	97.21 28	96.98 94	92.37 129	96.34 138	92.46 201	91.77 201	97.26 203	97.08 196

tpm cat1		94.06 148	94.90 156	93.06 141	95.42 128	98.52 135	96.64 130	80.67 191	97.82 153	92.63 97	93.39 149	95.00 107	96.06 145	91.36 204	91.58 203	96.98 205	96.66 201

CMPMVS	70.31 18	90.74 191	91.06 198	90.36 183	97.32 75	97.43 188	92.97 187	87.82 163	93.50 205	75.34 194	83.27 201	84.90 174	92.19 197	92.64 200	91.21 204	96.50 207	94.46 205
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
new-patchmatchnet		86.12 201	87.30 203	84.74 201	86.92 207	95.19 208	83.57 210	84.42 186	92.67 207	65.66 208	80.32 204	64.72 213	89.41 200	92.33 203	89.21 205	98.43 187	96.69 200

PMMVS2		77.26 204	79.47 207	74.70 207	76.00 214	88.37 213	74.22 214	76.34 209	78.31 212	54.13 215	69.96 209	52.50 217	70.14 211	84.83 207	88.71 206	97.35 200	93.58 208

MVE	67.97 19	65.53 209	67.43 211	63.31 210	59.33 217	74.20 215	53.09 219	70.43 213	66.27 215	43.13 216	45.98 215	30.62 220	70.65 210	79.34 211	86.30 207	83.25 216	89.33 209
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
tmp_tt				82.25 204	97.73 69	88.71 212	80.18 211	68.65 214	99.15 52	86.98 131	99.47 9	85.31 171	68.35 212	87.51 206	83.81 208	91.64 211

E-PMN		68.30 207	68.43 209	68.15 208	74.70 216	71.56 217	55.64 217	77.24 207	77.48 214	39.46 217	51.95 214	41.68 219	73.28 209	70.65 212	79.51 209	88.61 214	86.20 212

FPMVS		83.82 202	84.61 204	82.90 203	90.39 201	90.71 211	90.85 196	84.10 187	95.47 200	65.15 209	83.44 200	74.46 210	75.48 207	81.63 208	79.42 210	91.42 212	87.14 210

PMVS	72.60 17	76.39 205	77.66 208	74.92 206	81.04 212	69.37 218	68.47 215	80.54 193	85.39 211	65.07 210	73.52 208	72.91 211	65.67 213	80.35 210	76.81 211	88.71 213	85.25 213
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
EMVS		68.12 208	68.11 210	68.14 209	75.51 215	71.76 216	55.38 218	77.20 208	77.78 213	37.79 218	53.59 212	43.61 218	74.72 208	67.05 213	76.70 212	88.27 215	86.24 211

testmvs		31.24 210	40.15 212	20.86 212	12.61 218	17.99 219	25.16 220	13.30 215	48.42 216	24.82 219	53.07 213	30.13 222	28.47 214	42.73 214	37.65 213	20.79 217	51.04 214

test123		26.75 211	34.25 213	18.01 213	7.93 219	17.18 220	24.85 221	12.36 216	44.83 217	16.52 220	41.80 216	18.10 223	28.29 215	33.08 215	34.79 214	18.10 218	49.95 215

uanet_test		0.00 212	0.00 214	0.00 214	0.00 221	0.00 221	0.00 222	0.00 218	0.00 218	0.00 222	0.00 217	0.00 224	0.00 217	0.00 216	0.00 215	0.00 219	0.00 216

sosnet-low-res		0.00 212	0.00 214	0.00 214	0.00 221	0.00 221	0.00 222	0.00 218	0.00 218	0.00 222	0.00 217	0.00 224	0.00 217	0.00 216	0.00 215	0.00 219	0.00 216

sosnet		0.00 212	0.00 214	0.00 214	0.00 221	0.00 221	0.00 222	0.00 218	0.00 218	0.00 222	0.00 217	0.00 224	0.00 217	0.00 216	0.00 215	0.00 219	0.00 216

RE-MVS-def										69.05 206

9.14												99.79 45

SR-MVS					99.67 13			98.25 14				99.94 25

our_test_3					92.30 165	97.58 180	90.09 201

MTAPA										98.09 15		99.97 7

MTMP										98.46 11		99.96 12

Patchmatch-RL test							66.86 216

XVS					97.42 73	99.62 29	98.59 64			93.81 78		99.95 17				99.69 79

X-MVStestdata					97.42 73	99.62 29	98.59 64			93.81 78		99.95 17				99.69 79

abl_6				98.09 40	99.33 42	99.22 92	98.79 59	94.96 54	98.52 122	97.00 32	97.30 87	99.86 37	98.76 66			99.69 79	99.41 141

mPP-MVS					99.53 30							99.89 34

NP-MVS									98.57 117

Patchmtry						98.59 130	97.15 117	79.14 200		80.42 170

DeepMVS_CX						96.85 197	87.43 207	89.27 145	98.30 130	75.55 192	95.05 131	79.47 202	92.62 196	89.48 205		95.18 210	95.96 203

High-resolution multi-view benchmark