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.

TSAR-MVS + MP.		98.49 9	98.78 8	98.15 21	98.14 52	99.17 32	99.34 6	97.18 31	98.44 5	95.72 21	97.84 17	99.28 12	98.87 7	99.05 1	98.05 27	99.66 1	99.60 7
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
IS_MVSNet		95.28 65	96.43 57	93.94 91	95.30 93	99.01 47	95.90 99	91.12 91	94.13 114	87.50 106	91.23 85	94.45 69	94.17 109	98.45 19	98.50 7	99.65 2	99.23 37

CS-MVS		96.14 55	97.39 39	94.68 81	94.63 115	98.89 59	96.46 84	90.44 99	96.88 48	88.52 97	93.58 60	96.27 58	98.41 27	98.43 22	98.14 23	99.63 3	99.52 12

APDe-MVS		98.87 3	98.96 4	98.77 1	99.58 2	99.53 6	99.44 1	97.81 2	98.22 10	97.33 4	98.70 5	99.33 10	98.86 8	98.96 6	98.40 13	99.63 3	99.57 9

DROMVSNet		96.49 49	97.63 34	95.16 65	94.75 109	98.69 72	97.39 57	88.97 119	96.34 59	92.02 52	96.04 38	96.46 52	98.21 28	98.41 24	97.96 33	99.61 5	99.55 10

test2506		94.32 87	93.00 115	95.87 52	96.16 77	99.39 14	96.96 63	92.80 67	95.22 95	94.47 29	91.55 83	70.45 195	95.25 91	98.29 29	97.98 30	99.59 6	98.10 128

ECVR-MVS		94.14 89	92.96 116	95.52 59	96.16 77	99.39 14	96.96 63	92.80 67	95.22 95	92.38 49	81.48 153	80.31 151	95.25 91	98.29 29	97.98 30	99.59 6	98.05 129

SD-MVS		98.52 8	98.77 9	98.23 17	98.15 51	99.26 25	98.79 27	97.59 17	98.52 3	96.25 17	97.99 16	99.75 6	99.01 3	98.27 33	97.97 32	99.59 6	99.63 2
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
MCST-MVS		98.20 19	98.36 19	98.01 24	99.40 16	99.05 36	99.00 21	97.62 14	97.59 30	93.70 36	97.42 28	99.30 11	98.77 14	98.39 27	97.48 52	99.59 6	99.31 27

test1111		93.94 94	92.78 117	95.29 64	96.14 79	99.42 9	96.79 72	92.85 66	95.08 99	91.39 57	80.69 158	79.86 154	95.00 95	98.28 32	98.00 29	99.58 10	98.11 127

UA-Net		93.96 93	95.95 64	91.64 118	96.06 80	98.59 81	95.29 109	90.00 103	91.06 158	82.87 124	90.64 93	98.06 41	86.06 189	98.14 41	98.20 19	99.58 10	96.96 163

EPP-MVSNet		95.27 66	96.18 61	94.20 89	94.88 104	98.64 77	94.97 115	90.70 95	95.34 88	89.67 85	91.66 81	93.84 70	95.42 89	97.32 64	97.00 66	99.58 10	99.47 15

ETV-MVS		96.31 51	97.47 38	94.96 71	94.79 106	98.78 65	96.08 93	91.41 88	96.16 65	90.50 68	95.76 42	96.20 59	97.39 47	98.42 23	97.82 40	99.57 13	99.18 47

Vis-MVSNet (Re-imp)		94.46 82	96.24 59	92.40 110	95.23 96	98.64 77	95.56 107	90.99 92	94.42 109	85.02 116	90.88 92	94.65 68	88.01 178	98.17 38	98.37 16	99.57 13	98.53 105

DPE-MVS		98.75 5	98.91 6	98.57 5	99.21 25	99.54 5	99.42 2	97.78 6	97.49 32	96.84 10	98.94 1	99.82 5	98.59 22	98.90 10	98.22 18	99.56 15	99.48 14
Kehua Chen, Zhenlong Yuan, Tianlu Mao, Zhaoqi Wang: Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization. AAAI2025
CS-MVS-test		96.19 54	97.34 40	94.85 75	94.52 117	98.20 90	97.39 57	88.97 119	96.83 50	90.45 69	95.29 47	95.41 65	98.21 28	98.41 24	97.73 43	99.56 15	99.47 15

SteuartSystems-ACMMP		98.38 15	98.71 10	97.99 25	99.34 22	99.46 8	99.34 6	97.33 26	97.31 36	94.25 32	98.06 14	99.17 19	98.13 32	98.98 5	98.46 9	99.55 17	99.54 11
Skip Steuart: Steuart Systems R&D Blog.
canonicalmvs		95.25 67	95.45 71	95.00 69	95.27 95	98.72 70	96.89 66	89.82 107	96.51 56	90.84 64	93.72 59	86.01 118	97.66 43	95.78 119	97.94 35	99.54 18	99.50 13

casdiffmvs		94.38 86	94.15 96	94.64 83	94.70 113	98.51 82	96.03 96	91.66 83	95.70 81	89.36 90	86.48 123	85.03 128	96.60 69	97.40 62	97.30 60	99.52 19	98.67 97

baseline1		94.59 79	94.47 86	94.72 79	95.16 98	97.97 99	96.07 94	91.94 78	94.86 102	89.98 79	91.60 82	85.87 120	95.64 82	97.07 72	96.90 69	99.52 19	97.06 162

APD-MVS		98.36 16	98.32 23	98.41 9	99.47 6	99.26 25	99.12 16	97.77 7	96.73 52	96.12 18	97.27 29	98.88 25	98.46 26	98.47 18	98.39 14	99.52 19	99.22 39
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
Vis-MVSNet		92.77 112	95.00 81	90.16 137	94.10 126	98.79 64	94.76 121	88.26 128	92.37 143	79.95 138	88.19 113	91.58 81	84.38 199	97.59 58	97.58 49	99.52 19	98.91 84
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
DeepC-MVS_fast	96.13 1	98.13 21	98.27 26	97.97 26	99.16 28	99.03 43	99.05 19	97.24 28	98.22 10	94.17 34	95.82 40	98.07 40	98.69 17	98.83 11	98.80 2	99.52 19	99.10 54
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
CP-MVS		98.32 18	98.34 22	98.29 14	99.34 22	99.30 21	99.15 15	97.35 23	97.49 32	95.58 23	97.72 19	98.62 34	98.82 12	98.29 29	97.67 45	99.51 24	99.28 28

DeepC-MVS	94.87 4	96.76 48	96.50 55	97.05 37	98.21 50	99.28 23	98.67 28	97.38 22	97.31 36	90.36 74	89.19 104	93.58 72	98.19 31	98.31 28	98.50 7	99.51 24	99.36 21
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
ACMMPR		98.40 13	98.49 13	98.28 15	99.41 15	99.40 11	99.36 4	97.35 23	98.30 6	95.02 27	97.79 18	98.39 38	99.04 2	98.26 34	98.10 24	99.50 26	99.22 39

HFP-MVS		98.48 10	98.62 11	98.32 13	99.39 19	99.33 20	99.27 11	97.42 20	98.27 7	95.25 25	98.34 10	98.83 27	99.08 1	98.26 34	98.08 26	99.48 27	99.26 33

MP-MVS		98.09 23	98.30 25	97.84 28	99.34 22	99.19 31	99.23 14	97.40 21	97.09 43	93.03 42	97.58 23	98.85 26	98.57 24	98.44 21	97.69 44	99.48 27	99.23 37
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
PGM-MVS		97.81 26	98.11 29	97.46 31	99.55 3	99.34 19	99.32 9	94.51 47	96.21 64	93.07 39	98.05 15	97.95 43	98.82 12	98.22 37	97.89 38	99.48 27	99.09 56

3Dnovator	93.79 8	97.08 38	97.20 42	96.95 39	99.09 30	99.03 43	98.20 41	93.33 55	97.99 15	93.82 35	90.61 94	96.80 50	97.82 39	97.90 50	98.78 3	99.47 30	99.26 33

XVS					96.60 69	99.35 16	96.82 69			90.85 61		98.72 30				99.46 31

X-MVStestdata					96.60 69	99.35 16	96.82 69			90.85 61		98.72 30				99.46 31

X-MVS		97.84 25	98.19 28	97.42 32	99.40 16	99.35 16	99.06 18	97.25 27	97.38 35	90.85 61	96.06 37	98.72 30	98.53 25	98.41 24	98.15 22	99.46 31	99.28 28

ACMMP		97.37 34	97.48 37	97.25 33	98.88 38	99.28 23	98.47 35	96.86 36	97.04 45	92.15 50	97.57 24	96.05 62	97.67 42	97.27 65	95.99 93	99.46 31	99.14 53
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
xxxxxxxxxxxxxcwj		97.07 39	95.99 63	98.33 11	99.45 10	99.05 36	98.27 38	97.65 9	97.73 19	97.02 7	98.18 12	81.99 146	98.11 33	98.15 39	97.62 46	99.45 35	99.19 43

SF-MVS		98.39 14	98.45 17	98.33 11	99.45 10	99.05 36	98.27 38	97.65 9	97.73 19	97.02 7	98.18 12	99.25 15	98.11 33	98.15 39	97.62 46	99.45 35	99.19 43

tfpn200view9		93.64 100	92.57 120	94.89 72	95.33 91	98.94 51	96.82 69	92.31 71	92.63 134	88.29 98	87.21 115	78.01 162	97.12 56	96.82 77	95.85 98	99.45 35	98.56 102

zzz-MVS		98.43 12	98.31 24	98.57 5	99.48 5	99.40 11	99.32 9	97.62 14	97.70 23	96.67 12	96.59 33	99.09 22	98.86 8	98.65 13	97.56 50	99.45 35	99.17 49

thres600view7		93.49 105	92.37 131	94.79 78	95.42 88	98.93 53	96.58 80	92.31 71	93.04 128	87.88 103	86.62 121	76.94 168	97.09 57	96.82 77	95.63 103	99.45 35	98.63 99

thres200		93.62 101	92.54 121	94.88 73	95.36 90	98.93 53	96.75 74	92.31 71	92.84 131	88.28 100	86.99 117	77.81 164	97.13 54	96.82 77	95.92 94	99.45 35	98.49 108

DELS-MVS		96.06 56	96.04 62	96.07 51	97.77 57	99.25 27	98.10 43	93.26 57	94.42 109	92.79 45	88.52 111	93.48 73	95.06 94	98.51 16	98.83 1	99.45 35	99.28 28
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
PHI-MVS		97.78 27	98.44 18	97.02 38	98.73 39	99.25 27	98.11 42	95.54 41	96.66 55	92.79 45	98.52 6	99.38 9	97.50 46	97.84 51	98.39 14	99.45 35	99.03 68

CDPH-MVS		96.84 45	97.49 36	96.09 49	98.92 35	98.85 62	98.61 29	95.09 43	96.00 72	87.29 107	95.45 46	97.42 44	97.16 53	97.83 52	97.94 35	99.44 43	98.92 81

TSAR-MVS + GP.		97.45 32	98.36 19	96.39 44	95.56 87	98.93 53	97.74 50	93.31 56	97.61 29	94.24 33	98.44 9	99.19 17	98.03 37	97.60 57	97.41 55	99.44 43	99.33 24

3Dnovator+	93.91 7	97.23 36	97.22 41	97.24 34	98.89 37	98.85 62	98.26 40	93.25 59	97.99 15	95.56 24	90.01 100	98.03 42	98.05 36	97.91 49	98.43 10	99.44 43	99.35 22

DVP-MVS++		98.92 1	99.18 1	98.61 4	99.47 6	99.61 2	99.39 3	97.82 1	98.80 1	96.86 9	98.90 2	99.92 1	98.67 18	99.02 2	98.20 19	99.43 46	99.82 1

EIA-MVS		95.50 58	96.19 60	94.69 80	94.83 105	98.88 61	95.93 98	91.50 87	94.47 108	89.43 87	93.14 63	92.72 77	97.05 58	97.82 54	97.13 64	99.43 46	99.15 51

thres400		93.56 103	92.43 128	94.87 74	95.40 89	98.91 56	96.70 76	92.38 70	92.93 130	88.19 102	86.69 120	77.35 165	97.13 54	96.75 82	95.85 98	99.42 48	98.56 102

SMA-MVS		98.66 7	98.89 7	98.39 10	99.60 1	99.41 10	99.00 21	97.63 13	97.78 18	95.83 20	98.33 11	99.83 4	98.85 10	98.93 8	98.56 6	99.41 49	99.40 18
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
UniMVSNet (Re)		90.03 149	89.61 157	90.51 133	89.97 173	96.12 143	92.32 159	89.26 115	90.99 159	80.95 136	78.25 167	75.08 175	91.14 148	93.78 155	93.87 153	99.41 49	99.21 41

CNVR-MVS		98.47 11	98.46 16	98.48 8	99.40 16	99.05 36	99.02 20	97.54 18	97.73 19	96.65 13	97.20 30	99.13 20	98.85 10	98.91 9	98.10 24	99.41 49	99.08 57

MSP-MVS		98.73 6	98.93 5	98.50 7	99.44 13	99.57 4	99.36 4	97.65 9	98.14 12	96.51 16	98.49 7	99.65 8	98.67 18	98.60 14	98.42 11	99.40 52	99.63 2
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
ACMMP_NAP		98.20 19	98.49 13	97.85 27	99.50 4	99.40 11	99.26 12	97.64 12	97.47 34	92.62 48	97.59 21	99.09 22	98.71 16	98.82 12	97.86 39	99.40 52	99.19 43

NCCC		98.10 22	98.05 31	98.17 20	99.38 20	99.05 36	99.00 21	97.53 19	98.04 14	95.12 26	94.80 53	99.18 18	98.58 23	98.49 17	97.78 42	99.39 54	98.98 75

thres100view900		93.55 104	92.47 127	94.81 77	95.33 91	98.74 67	96.78 73	92.30 74	92.63 134	88.29 98	87.21 115	78.01 162	96.78 64	96.38 97	95.92 94	99.38 55	98.40 114

MVS_0304		96.31 51	96.91 50	95.62 56	97.21 65	99.20 30	98.55 32	93.10 62	97.04 45	89.73 83	90.30 96	96.35 54	95.71 80	98.14 41	97.93 37	99.38 55	99.40 18

FC-MVSNet-train		93.85 96	93.91 98	93.78 95	94.94 103	96.79 125	94.29 129	91.13 90	93.84 119	88.26 101	90.40 95	85.23 125	94.65 102	96.54 91	95.31 112	99.38 55	99.28 28

DVP-MVS		98.86 4	98.97 3	98.75 2	99.43 14	99.63 1	99.25 13	97.81 2	98.62 2	97.69 1	97.59 21	99.90 2	98.93 5	98.99 4	98.42 11	99.37 58	99.62 4
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
UniMVSNet_NR-MVSNet		90.35 142	89.96 154	90.80 129	89.66 176	95.83 155	92.48 155	90.53 98	90.96 160	79.57 140	79.33 164	77.14 166	93.21 128	92.91 172	94.50 140	99.37 58	99.05 65

SED-MVS		98.90 2	99.07 2	98.69 3	99.38 20	99.61 2	99.33 8	97.80 4	98.25 8	97.60 2	98.87 4	99.89 3	98.67 18	99.02 2	98.26 17	99.36 60	99.61 6

DU-MVS		89.67 152	88.84 163	90.63 132	89.26 186	95.61 161	92.48 155	89.91 104	91.22 156	79.57 140	77.72 168	71.18 192	93.21 128	92.53 176	94.57 134	99.35 61	99.05 65

WR-MVS_H		87.93 174	87.85 177	88.03 167	89.62 177	95.58 165	90.47 187	85.55 159	87.20 192	76.83 154	74.42 182	72.67 186	86.37 187	93.22 167	93.04 167	99.33 62	98.83 92

QAPM		96.78 47	97.14 45	96.36 45	99.05 31	99.14 34	98.02 44	93.26 57	97.27 38	90.84 64	91.16 86	97.31 45	97.64 44	97.70 55	98.20 19	99.33 62	99.18 47

NR-MVSNet		89.34 155	88.66 164	90.13 140	90.40 165	95.61 161	93.04 147	89.91 104	91.22 156	78.96 143	77.72 168	68.90 204	89.16 174	94.24 152	93.95 150	99.32 64	98.99 73

TranMVSNet+NR-MVSNet		89.23 158	88.48 167	90.11 141	89.07 192	95.25 176	92.91 148	90.43 100	90.31 166	77.10 152	76.62 171	71.57 190	91.83 140	92.12 182	94.59 133	99.32 64	98.92 81

LGP-MVS_train		94.12 90	94.62 83	93.53 98	96.44 73	97.54 104	97.40 56	91.84 80	94.66 104	81.09 135	95.70 43	83.36 139	95.10 93	96.36 100	95.71 102	99.32 64	99.03 68

HPM-MVS++		98.34 17	98.47 15	98.18 18	99.46 9	99.15 33	99.10 17	97.69 8	97.67 26	94.93 28	97.62 20	99.70 7	98.60 21	98.45 19	97.46 53	99.31 67	99.26 33

CLD-MVS		94.79 73	94.36 89	95.30 63	95.21 97	97.46 107	97.23 59	92.24 75	96.43 57	91.77 54	92.69 68	84.31 131	96.06 75	95.52 125	95.03 119	99.31 67	99.06 62
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
CP-MVSNet		87.89 177	87.27 183	88.62 154	89.30 184	95.06 179	90.60 186	85.78 155	87.43 191	75.98 160	74.60 179	68.14 207	90.76 157	93.07 170	93.60 158	99.30 69	98.98 75

PVSNet_Blended_VisFu		94.77 75	95.54 69	93.87 93	96.48 72	98.97 49	94.33 128	91.84 80	94.93 101	90.37 73	85.04 134	94.99 66	90.87 156	98.12 43	97.30 60	99.30 69	99.45 17

PS-CasMVS		87.33 184	86.68 193	88.10 161	89.22 191	94.93 184	90.35 189	85.70 156	86.44 196	74.01 175	73.43 189	66.59 213	90.04 168	92.92 171	93.52 159	99.28 71	98.91 84

TAPA-MVS	94.18 5	96.38 50	96.49 56	96.25 46	98.26 49	98.66 74	98.00 45	94.96 45	97.17 40	89.48 86	92.91 66	96.35 54	97.53 45	96.59 88	95.90 96	99.28 71	97.82 134
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
Effi-MVS+		92.93 111	93.86 100	91.86 114	94.07 127	98.09 96	95.59 106	85.98 153	94.27 112	79.54 142	91.12 89	81.81 147	96.71 66	96.67 86	96.06 89	99.27 73	98.98 75

WR-MVS		87.93 174	88.09 171	87.75 172	89.26 186	95.28 173	90.81 184	86.69 144	88.90 174	75.29 166	74.31 183	73.72 181	85.19 195	92.26 179	93.32 163	99.27 73	98.81 93

abl_6				96.82 41	98.60 43	98.74 67	97.74 50	93.73 51	96.25 62	94.37 31	94.55 55	98.60 35	97.25 50			99.27 73	98.61 100

MVS_111021_HR		97.04 40	98.20 27	95.69 55	98.44 47	99.29 22	96.59 79	93.20 60	97.70 23	89.94 81	98.46 8	96.89 48	96.71 66	98.11 44	97.95 34	99.27 73	99.01 71

LS3D		95.46 61	95.14 76	95.84 53	97.91 56	98.90 58	98.58 31	97.79 5	97.07 44	83.65 122	88.71 107	88.64 104	97.82 39	97.49 60	97.42 54	99.26 77	97.72 142

OPM-MVS		93.61 102	92.43 128	95.00 69	96.94 68	97.34 110	97.78 49	94.23 48	89.64 170	85.53 114	88.70 108	82.81 142	96.28 73	96.28 103	95.00 122	99.24 78	97.22 155
Ray L. Khuboni, Hongjun Xu: Octagram Propagation Matching for Multi-Scale View Stereopsis (OPM-MVS).
PEN-MVS		87.22 186	86.50 195	88.07 162	88.88 195	94.44 194	90.99 183	86.21 148	86.53 195	73.66 176	74.97 176	66.56 214	89.42 173	91.20 193	93.48 160	99.24 78	98.31 121

PVSNet_BlendedMVS		95.41 63	95.28 73	95.57 57	97.42 61	99.02 45	95.89 101	93.10 62	96.16 65	93.12 37	91.99 74	85.27 123	94.66 100	98.09 45	97.34 58	99.24 78	99.08 57

PVSNet_Blended		95.41 63	95.28 73	95.57 57	97.42 61	99.02 45	95.89 101	93.10 62	96.16 65	93.12 37	91.99 74	85.27 123	94.66 100	98.09 45	97.34 58	99.24 78	99.08 57

CSCG		97.44 33	97.18 44	97.75 29	99.47 6	99.52 7	98.55 32	95.41 42	97.69 25	95.72 21	94.29 56	95.53 64	98.10 35	96.20 107	97.38 57	99.24 78	99.62 4

OpenMVS	92.33 11	95.50 58	95.22 75	95.82 54	98.98 32	98.97 49	97.67 52	93.04 65	94.64 105	89.18 93	84.44 139	94.79 67	96.79 63	97.23 66	97.61 48	99.24 78	98.88 86

CANet		96.84 45	97.20 42	96.42 43	97.92 55	99.24 29	98.60 30	93.51 54	97.11 42	93.07 39	91.16 86	97.24 46	96.21 74	98.24 36	98.05 27	99.22 84	99.35 22

train_agg		97.65 30	98.06 30	97.18 35	98.94 34	98.91 56	98.98 25	97.07 33	96.71 53	90.66 66	97.43 27	99.08 24	98.20 30	97.96 48	97.14 63	99.22 84	99.19 43

ACMM	92.75 10	94.41 85	93.84 101	95.09 67	96.41 74	96.80 122	94.88 118	93.54 53	96.41 58	90.16 75	92.31 72	83.11 140	96.32 72	96.22 105	94.65 129	99.22 84	97.35 152
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
GBi-Net		93.81 97	94.18 92	93.38 102	91.34 157	95.86 152	96.22 88	88.68 122	95.23 92	90.40 70	86.39 124	91.16 82	94.40 106	96.52 92	96.30 80	99.21 87	97.79 135

test1		93.81 97	94.18 92	93.38 102	91.34 157	95.86 152	96.22 88	88.68 122	95.23 92	90.40 70	86.39 124	91.16 82	94.40 106	96.52 92	96.30 80	99.21 87	97.79 135

FMVSNet2		93.30 108	93.36 111	93.22 106	91.34 157	95.86 152	96.22 88	88.24 129	95.15 98	89.92 82	81.64 151	89.36 96	94.40 106	96.77 81	96.98 67	99.21 87	97.79 135

tttt0517		94.52 81	95.44 72	93.44 101	94.51 118	98.68 73	94.61 124	90.72 93	95.61 85	86.84 111	93.78 58	89.26 98	94.74 97	97.02 75	94.86 124	99.20 90	98.87 88

GeoE		92.52 116	92.64 119	92.39 111	93.96 128	97.76 101	96.01 97	85.60 158	93.23 126	83.94 119	81.56 152	84.80 129	95.63 83	96.22 105	95.83 100	99.19 91	99.07 61

thisisatest0530		94.54 80	95.47 70	93.46 100	94.51 118	98.65 76	94.66 122	90.72 93	95.69 83	86.90 110	93.80 57	89.44 95	94.74 97	96.98 76	94.86 124	99.19 91	98.85 90

DI_MVS_plusplus_trai		94.01 92	93.63 105	94.44 85	94.54 116	98.26 88	97.51 54	90.63 96	95.88 76	89.34 91	80.54 160	89.36 96	95.48 88	96.33 101	96.27 83	99.17 93	98.78 95

MSLP-MVS++		98.04 24	97.93 33	98.18 18	99.10 29	99.09 35	98.34 37	96.99 34	97.54 31	96.60 14	94.82 52	98.45 36	98.89 6	97.46 61	98.77 4	99.17 93	99.37 20

AdaColmap		97.53 31	96.93 48	98.24 16	99.21 25	98.77 66	98.47 35	97.34 25	96.68 54	96.52 15	95.11 50	96.12 60	98.72 15	97.19 69	96.24 84	99.17 93	98.39 115

Fast-Effi-MVS+		91.87 120	92.08 135	91.62 120	92.91 143	97.21 114	94.93 116	84.60 172	93.61 122	81.49 133	83.50 144	78.95 157	96.62 68	96.55 90	96.22 85	99.16 96	98.51 106

FC-MVSNet-test		91.63 124	93.82 102	89.08 150	92.02 152	96.40 137	93.26 143	87.26 138	93.72 120	77.26 150	88.61 110	89.86 93	85.50 192	95.72 123	95.02 120	99.16 96	97.44 149

UGNet		94.92 68	96.63 53	92.93 107	96.03 81	98.63 79	94.53 125	91.52 86	96.23 63	90.03 78	92.87 67	96.10 61	86.28 188	96.68 85	96.60 76	99.16 96	99.32 26
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
ACMP	92.88 9	94.43 83	94.38 88	94.50 84	96.01 82	97.69 102	95.85 104	92.09 76	95.74 80	89.12 94	95.14 49	82.62 144	94.77 96	95.73 121	94.67 128	99.14 99	99.06 62
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
DTE-MVSNet		86.67 189	86.09 196	87.35 181	88.45 201	94.08 200	90.65 185	86.05 152	86.13 197	72.19 180	74.58 181	66.77 212	87.61 181	90.31 196	93.12 166	99.13 100	97.62 145

OMC-MVS		97.00 41	96.92 49	97.09 36	98.69 40	98.66 74	97.85 48	95.02 44	98.09 13	94.47 29	93.15 62	96.90 47	97.38 48	97.16 70	96.82 73	99.13 100	97.65 143

anonymousdsp		88.90 163	91.00 147	86.44 189	88.74 199	95.97 147	90.40 188	82.86 182	88.77 177	67.33 201	81.18 155	81.44 149	90.22 167	96.23 104	94.27 144	99.12 102	99.16 50

MVS_Test		94.82 71	95.66 66	93.84 94	94.79 106	98.35 85	96.49 83	89.10 118	96.12 68	87.09 109	92.58 69	90.61 88	96.48 70	96.51 95	96.89 70	99.11 103	98.54 104

IB-MVS	89.56 15	91.71 123	92.50 123	90.79 130	95.94 83	98.44 83	87.05 200	91.38 89	93.15 127	92.98 43	84.78 135	85.14 126	78.27 207	92.47 178	94.44 141	99.10 104	99.08 57
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
PLC	94.95 3	97.37 34	96.77 52	98.07 22	98.97 33	98.21 89	97.94 47	96.85 37	97.66 27	97.58 3	93.33 61	96.84 49	98.01 38	97.13 71	96.20 86	99.09 105	98.01 130
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
pm-mvs1		89.19 159	89.02 162	89.38 148	90.40 165	95.74 159	92.05 167	88.10 131	86.13 197	77.70 147	73.72 187	79.44 156	88.97 175	95.81 118	94.51 139	99.08 106	97.78 140

PCF-MVS	93.95 6	95.65 57	95.14 76	96.25 46	97.73 59	98.73 69	97.59 53	97.13 32	92.50 138	89.09 95	89.85 101	96.65 51	96.90 61	94.97 139	94.89 123	99.08 106	98.38 116
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
Baseline_NR-MVSNet		89.27 157	88.01 173	90.73 131	89.26 186	93.71 202	92.71 152	89.78 109	90.73 161	81.28 134	73.53 188	72.85 184	92.30 135	92.53 176	93.84 155	99.07 108	98.88 86

FMVSNet3		93.79 99	94.17 94	93.35 104	91.21 160	95.99 145	96.62 77	88.68 122	95.23 92	90.40 70	86.39 124	91.16 82	94.11 110	95.96 112	96.67 74	99.07 108	97.79 135

HQP-MVS		94.43 83	94.57 84	94.27 88	96.41 74	97.23 113	96.89 66	93.98 49	95.94 74	83.68 121	95.01 51	84.46 130	95.58 85	95.47 127	94.85 127	99.07 108	99.00 72

ET-MVSNet_ETH3D		93.34 107	94.33 90	92.18 113	83.26 214	97.66 103	96.72 75	89.89 106	95.62 84	87.17 108	96.00 39	83.69 137	96.99 59	93.78 155	95.34 111	99.06 111	98.18 125

DCV-MVSNet		94.76 76	95.12 78	94.35 87	95.10 101	95.81 156	96.46 84	89.49 113	96.33 60	90.16 75	92.55 70	90.26 90	95.83 79	95.52 125	96.03 91	99.06 111	99.33 24

tfpnnormal		88.50 166	87.01 188	90.23 135	91.36 156	95.78 158	92.74 150	90.09 102	83.65 206	76.33 158	71.46 200	69.58 201	91.84 139	95.54 124	94.02 149	99.06 111	99.03 68

TransMVSNet (Re)		87.73 179	86.79 190	88.83 152	90.76 161	94.40 195	91.33 179	89.62 111	84.73 203	75.41 165	72.73 192	71.41 191	86.80 184	94.53 144	93.93 151	99.06 111	95.83 176

diffmvs		94.31 88	94.21 91	94.42 86	94.64 114	98.28 86	96.36 87	91.56 84	96.77 51	88.89 96	88.97 105	84.23 132	96.01 78	96.05 111	96.41 79	99.05 115	98.79 94

MVS_111021_LR		97.16 37	98.01 32	96.16 48	98.47 45	98.98 48	96.94 65	93.89 50	97.64 28	91.44 55	98.89 3	96.41 53	97.20 52	98.02 47	97.29 62	99.04 116	98.85 90

Anonymous202405211			92.18 133		95.04 102	98.20 90	96.14 91	91.79 82	93.93 115		74.60 179	88.38 107	96.48 70	95.17 135	95.82 101	99.00 117	99.15 51

MVSTER		94.89 69	95.07 79	94.68 81	94.71 111	96.68 128	97.00 61	90.57 97	95.18 97	93.05 41	95.21 48	86.41 115	93.72 118	97.59 58	95.88 97	99.00 117	98.50 107

MSDG		94.82 71	93.73 103	96.09 49	98.34 48	97.43 109	97.06 60	96.05 39	95.84 78	90.56 67	86.30 128	89.10 101	95.55 86	96.13 110	95.61 104	99.00 117	95.73 178

gg-mvs-nofinetune		86.17 192	88.57 166	83.36 200	93.44 136	98.15 94	96.58 80	72.05 214	74.12 218	49.23 222	64.81 212	90.85 86	89.90 171	97.83 52	96.84 71	98.97 120	97.41 150

TSAR-MVS + ACMM		97.71 29	98.60 12	96.66 42	98.64 42	99.05 36	98.85 26	97.23 29	98.45 4	89.40 89	97.51 25	99.27 14	96.88 62	98.53 15	97.81 41	98.96 121	99.59 8

DPM-MVS		96.86 44	96.82 51	96.91 40	98.08 53	98.20 90	98.52 34	97.20 30	97.24 39	91.42 56	91.84 78	98.45 36	97.25 50	97.07 72	97.40 56	98.95 122	97.55 146

CNLPA		96.90 43	96.28 58	97.64 30	98.56 44	98.63 79	96.85 68	96.60 38	97.73 19	97.08 6	89.78 102	96.28 57	97.80 41	96.73 83	96.63 75	98.94 123	98.14 126

ACMH+	90.88 12	91.41 129	91.13 145	91.74 117	95.11 100	96.95 117	93.13 145	89.48 114	92.42 140	79.93 139	85.13 133	78.02 161	93.82 116	93.49 162	93.88 152	98.94 123	97.99 131

v7n		86.43 190	86.52 194	86.33 190	87.91 203	94.93 184	90.15 190	83.05 180	86.57 194	70.21 191	71.48 199	66.78 211	87.72 179	94.19 154	92.96 169	98.92 125	98.76 96

test0.0.03 1		91.97 119	93.91 98	89.72 142	93.31 139	96.40 137	91.34 178	87.06 141	93.86 117	81.67 131	91.15 88	89.16 100	86.02 190	95.08 136	95.09 117	98.91 126	96.64 172

HyFIR lowres test		92.03 118	91.55 142	92.58 109	97.13 66	98.72 70	94.65 123	86.54 146	93.58 123	82.56 126	67.75 208	90.47 89	95.67 81	95.87 115	95.54 106	98.91 126	98.93 80

thisisatest0515		90.12 147	92.06 136	87.85 171	90.03 171	96.17 142	87.83 197	87.45 136	91.71 152	77.15 151	85.40 132	84.01 134	85.74 191	95.41 129	93.30 164	98.88 128	98.43 110

IterMVS-LS		92.56 115	93.18 112	91.84 115	93.90 129	94.97 182	94.99 114	86.20 150	94.18 113	82.68 125	85.81 130	87.36 111	94.43 104	95.31 131	96.02 92	98.87 129	98.60 101
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
COLMAP_ROB	90.49 14	93.27 109	92.71 118	93.93 92	97.75 58	97.44 108	96.07 94	93.17 61	95.40 87	83.86 120	83.76 143	88.72 103	93.87 114	94.25 151	94.11 146	98.87 129	95.28 184
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
v2v482		88.25 170	87.71 180	88.88 151	89.23 190	95.28 173	92.10 165	87.89 133	88.69 178	73.31 177	75.32 174	71.64 189	91.89 138	92.10 184	92.92 170	98.86 131	97.99 131

UniMVSNet_ETH3D		88.47 167	86.00 197	91.35 122	91.55 154	96.29 139	92.53 154	88.81 121	85.58 201	82.33 127	67.63 209	66.87 210	94.04 112	91.49 191	95.24 114	98.84 132	98.92 81

pmmvs5		87.83 178	88.09 171	87.51 180	89.59 179	95.48 166	89.75 192	84.73 170	86.07 199	71.44 184	80.57 159	70.09 199	90.74 159	94.47 145	92.87 172	98.82 133	97.10 157

EG-PatchMatch MVS		86.68 188	87.24 184	86.02 193	90.58 163	96.26 140	91.08 182	81.59 187	84.96 202	69.80 196	71.35 201	75.08 175	84.23 200	94.24 152	93.35 162	98.82 133	95.46 183

FMVSNet1		91.54 127	90.93 148	92.26 112	90.35 167	95.27 175	95.22 112	87.16 140	91.37 155	87.62 105	75.45 173	83.84 135	94.43 104	96.52 92	96.30 80	98.82 133	97.74 141

v1144		87.92 176	87.79 178	88.07 162	89.27 185	95.15 178	92.17 164	85.62 157	88.52 179	71.52 183	73.80 186	72.40 187	91.06 150	93.54 161	92.80 173	98.81 136	98.33 118

v10		88.00 172	87.96 174	88.05 165	89.44 181	94.68 189	92.36 158	83.35 179	89.37 172	72.96 178	73.98 185	72.79 185	91.35 146	93.59 157	92.88 171	98.81 136	98.42 112

Fast-Effi-MVS+-dtu		91.19 131	93.64 104	88.33 158	92.19 151	96.46 134	93.99 132	81.52 189	92.59 136	71.82 182	92.17 73	85.54 121	91.68 142	95.73 121	94.64 130	98.80 138	98.34 117

v8		88.21 171	87.94 176	88.51 155	89.62 177	95.01 181	92.31 160	84.99 167	88.94 173	74.70 172	75.03 175	73.51 182	90.67 160	92.11 183	92.74 176	98.80 138	98.24 122

CDS-MVSNet		92.77 112	93.60 106	91.80 116	92.63 147	96.80 122	95.24 111	89.14 117	90.30 167	84.58 117	86.76 118	90.65 87	90.42 164	95.89 114	96.49 77	98.79 140	98.32 120
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
v1192		87.51 181	87.31 182	87.74 173	89.04 193	94.87 187	92.07 166	85.03 166	88.49 180	70.32 189	72.65 193	70.35 197	91.21 147	93.59 157	92.80 173	98.78 141	98.42 112

ACMH	90.77 13	91.51 128	91.63 141	91.38 121	95.62 86	96.87 120	91.76 173	89.66 110	91.58 153	78.67 144	86.73 119	78.12 160	93.77 117	94.59 142	94.54 137	98.78 141	98.98 75
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
TSAR-MVS + COLMAP		94.79 73	94.51 85	95.11 66	96.50 71	97.54 104	97.99 46	94.54 46	97.81 17	85.88 113	96.73 32	81.28 150	96.99 59	96.29 102	95.21 116	98.76 143	96.73 169

MAR-MVS		95.50 58	95.60 67	95.39 62	98.67 41	98.18 93	95.89 101	89.81 108	94.55 107	91.97 53	92.99 64	90.21 91	97.30 49	96.79 80	97.49 51	98.72 144	98.99 73
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
v144192		87.40 183	87.20 185	87.64 174	88.89 194	94.88 186	91.65 174	84.70 171	87.80 186	71.17 187	73.20 191	70.91 193	90.75 158	92.69 174	92.49 179	98.71 145	98.43 110

v1921920		87.31 185	87.13 186	87.52 179	88.87 196	94.72 188	91.96 171	84.59 173	88.28 182	69.86 195	72.50 195	70.03 200	91.10 149	93.33 164	92.61 178	98.71 145	98.44 109

PatchMatch-RL		94.69 77	94.41 87	95.02 68	97.63 60	98.15 94	94.50 126	91.99 77	95.32 89	91.31 58	95.47 45	83.44 138	96.02 77	96.56 89	95.23 115	98.69 147	96.67 170

Anonymous20231211		93.49 105	92.33 132	94.84 76	94.78 108	98.00 97	96.11 92	91.85 79	94.86 102	90.91 60	74.69 178	89.18 99	96.73 65	94.82 140	95.51 107	98.67 148	99.24 36

v1240		86.89 187	86.75 192	87.06 184	88.75 198	94.65 191	91.30 180	84.05 175	87.49 190	68.94 199	71.96 198	68.86 205	90.65 161	93.33 164	92.72 177	98.67 148	98.24 122

baseline2		93.01 110	94.17 94	91.64 118	92.83 145	97.49 106	93.40 140	87.53 135	93.67 121	86.07 112	91.83 79	86.58 112	91.36 145	96.38 97	95.06 118	98.67 148	98.20 124

gm-plane-assit		83.26 202	85.29 199	80.89 203	89.52 180	89.89 213	70.26 219	78.24 195	77.11 216	58.01 219	74.16 184	66.90 209	90.63 162	97.20 67	96.05 90	98.66 151	95.68 179

testgi		89.42 153	91.50 143	87.00 185	92.40 150	95.59 163	89.15 194	85.27 165	92.78 132	72.42 179	91.75 80	76.00 171	84.09 201	94.38 148	93.82 156	98.65 152	96.15 173

TDRefinement		89.07 161	88.15 170	90.14 139	95.16 98	96.88 118	95.55 108	90.20 101	89.68 169	76.42 157	76.67 170	74.30 178	84.85 196	93.11 168	91.91 186	98.64 153	94.47 187

EPNet		96.27 53	96.97 47	95.46 60	98.47 45	98.28 86	97.41 55	93.67 52	95.86 77	92.86 44	97.51 25	93.79 71	91.76 141	97.03 74	97.03 65	98.61 154	99.28 28
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
USDC		90.69 136	90.52 152	90.88 127	94.17 125	96.43 135	95.82 105	86.76 143	93.92 116	76.27 159	86.49 122	74.30 178	93.67 121	95.04 138	93.36 161	98.61 154	94.13 191

V42		88.31 169	87.95 175	88.73 153	89.44 181	95.34 172	92.23 163	87.21 139	88.83 175	74.49 173	74.89 177	73.43 183	90.41 166	92.08 185	92.77 175	98.60 156	98.33 118

SixPastTwentyTwo		88.37 168	89.47 158	87.08 183	90.01 172	95.93 151	87.41 198	85.32 162	90.26 168	70.26 190	86.34 127	71.95 188	90.93 152	92.89 173	91.72 187	98.55 157	97.22 155

CPTT-MVS		97.78 27	97.54 35	98.05 23	98.91 36	99.05 36	99.00 21	96.96 35	97.14 41	95.92 19	95.50 44	98.78 29	98.99 4	97.20 67	96.07 88	98.54 158	99.04 67

GA-MVS		89.28 156	90.75 151	87.57 177	91.77 153	96.48 133	92.29 161	87.58 134	90.61 164	65.77 203	84.48 138	76.84 169	89.46 172	95.84 116	93.68 157	98.52 159	97.34 153

pmmvs4		90.55 139	89.91 155	91.30 123	90.26 169	94.95 183	92.73 151	87.94 132	93.44 125	85.35 115	82.28 150	76.09 170	93.02 130	93.56 160	92.26 184	98.51 160	96.77 168

CANet_DTU		93.92 95	96.57 54	90.83 128	95.63 85	98.39 84	96.99 62	87.38 137	96.26 61	71.97 181	96.31 35	93.02 74	94.53 103	97.38 63	96.83 72	98.49 161	97.79 135

MIMVSNet		88.99 162	91.07 146	86.57 188	86.78 208	95.62 160	91.20 181	75.40 208	90.65 163	76.57 155	84.05 141	82.44 145	91.01 151	95.84 116	95.38 110	98.48 162	93.50 200

CR-MVSNet		90.16 146	91.96 138	88.06 164	93.32 138	95.95 149	93.36 141	75.99 206	92.40 141	75.19 167	83.18 145	85.37 122	92.05 136	95.21 133	94.56 135	98.47 163	97.08 160

test20.03		82.92 203	85.52 198	79.90 206	87.75 204	91.84 208	82.80 210	82.99 181	82.65 211	60.32 215	78.90 165	70.50 194	67.10 214	92.05 186	90.89 189	98.44 164	91.80 206

RPMNet		90.19 145	92.03 137	88.05 165	93.46 135	95.95 149	93.41 139	74.59 211	92.40 141	75.91 161	84.22 140	86.41 115	92.49 132	94.42 147	93.85 154	98.44 164	96.96 163

PMMVS		94.61 78	95.56 68	93.50 99	94.30 122	96.74 126	94.91 117	89.56 112	95.58 86	87.72 104	96.15 36	92.86 75	96.06 75	95.47 127	95.02 120	98.43 166	97.09 158

v148		87.51 181	86.79 190	88.36 157	89.39 183	95.21 177	89.84 191	88.20 130	87.61 189	77.56 148	73.38 190	70.32 198	86.80 184	90.70 195	92.31 182	98.37 167	97.98 133

LTVRE_ROB	87.32 16	87.55 180	88.25 169	86.73 186	90.66 162	95.80 157	93.05 146	84.77 169	83.35 207	60.32 215	83.12 146	67.39 208	93.32 125	94.36 149	94.86 124	98.28 168	98.87 88
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
baseline		94.83 70	95.82 65	93.68 96	94.75 109	97.80 100	96.51 82	88.53 125	97.02 47	89.34 91	92.93 65	92.18 79	94.69 99	95.78 119	96.08 87	98.27 169	98.97 79

TinyColmap		89.42 153	88.58 165	90.40 134	93.80 133	95.45 168	93.96 133	86.54 146	92.24 146	76.49 156	80.83 156	70.44 196	93.37 124	94.45 146	93.30 164	98.26 170	93.37 202

CHOSEN 1792x2688		92.66 114	92.49 124	92.85 108	97.13 66	98.89 59	95.90 99	88.50 126	95.32 89	83.31 123	71.99 197	88.96 102	94.10 111	96.69 84	96.49 77	98.15 171	99.10 54

MS-PatchMatch		91.82 121	92.51 122	91.02 124	95.83 84	96.88 118	95.05 113	84.55 174	93.85 118	82.01 128	82.51 149	91.71 80	90.52 163	95.07 137	93.03 168	98.13 172	94.52 186

FMVSNet5		90.36 141	90.93 148	89.70 143	87.99 202	92.25 207	92.03 168	83.51 178	92.20 147	84.13 118	85.59 131	86.48 113	92.43 133	94.61 141	94.52 138	98.13 172	90.85 208

Anonymous20231206		83.84 201	85.19 200	82.26 202	87.38 206	92.87 204	85.49 204	83.65 177	86.07 199	63.44 210	68.42 205	69.01 203	75.45 210	93.34 163	92.44 180	98.12 174	94.20 190

MIMVSNet1		80.03 206	80.93 207	78.97 207	72.46 220	90.73 211	80.81 213	82.44 185	80.39 212	63.64 208	57.57 215	64.93 215	76.37 208	91.66 189	91.55 188	98.07 175	89.70 210

TAMVS		90.54 140	90.87 150	90.16 137	91.48 155	96.61 130	93.26 143	86.08 151	87.71 187	81.66 132	83.11 147	84.04 133	90.42 164	94.54 143	94.60 132	98.04 176	95.48 182

pmmvs-eth3d		84.33 200	82.94 205	85.96 194	84.16 211	90.94 210	86.55 201	83.79 176	84.25 204	75.85 162	70.64 202	56.43 220	87.44 183	92.20 181	90.41 193	97.97 177	95.68 179

test-mter		90.95 133	93.54 110	87.93 170	90.28 168	96.80 122	91.44 175	82.68 184	92.15 148	74.37 174	89.57 103	88.23 109	90.88 155	96.37 99	94.31 143	97.93 178	97.37 151

test_part1		91.21 130	89.47 158	93.24 105	94.26 123	95.45 168	95.26 110	88.36 127	88.49 180	90.04 77	72.61 194	82.82 141	93.69 120	93.25 166	94.62 131	97.84 179	99.06 62

GG-mvs-BLEND		66.17 213	94.91 82	32.63 219	1.32 227	96.64 129	91.40 176	0.85 225	94.39 111	2.20 228	90.15 99	95.70 63	2.27 224	96.39 96	95.44 109	97.78 180	95.68 179

PatchT		89.13 160	91.71 139	86.11 192	92.92 142	95.59 163	83.64 208	75.09 209	91.87 150	75.19 167	82.63 148	85.06 127	92.05 136	95.21 133	94.56 135	97.76 181	97.08 160

test-LLR		91.62 125	93.56 108	89.35 149	93.31 139	96.57 131	92.02 169	87.06 141	92.34 144	75.05 170	90.20 97	88.64 104	90.93 152	96.19 108	94.07 147	97.75 182	96.90 166

TESTMET0.1,1		91.07 132	93.56 108	88.17 160	90.43 164	96.57 131	92.02 169	82.83 183	92.34 144	75.05 170	90.20 97	88.64 104	90.93 152	96.19 108	94.07 147	97.75 182	96.90 166

IterMVS-SCA-FT		90.24 143	92.48 126	87.63 175	92.85 144	94.30 198	93.79 134	81.47 190	92.66 133	69.95 193	84.66 137	88.38 107	89.99 169	95.39 130	94.34 142	97.74 184	97.63 144

PM-MVS		84.72 199	84.47 203	85.03 195	84.67 210	91.57 209	86.27 202	82.31 186	87.65 188	70.62 188	76.54 172	56.41 221	88.75 177	92.59 175	89.85 196	97.54 185	96.66 171

IterMVS		90.20 144	92.43 128	87.61 176	92.82 146	94.31 197	94.11 130	81.54 188	92.97 129	69.90 194	84.71 136	88.16 110	89.96 170	95.25 132	94.17 145	97.31 186	97.46 148
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
Effi-MVS+-dtu		91.78 122	93.59 107	89.68 145	92.44 149	97.11 115	94.40 127	84.94 168	92.43 139	75.48 163	91.09 90	83.75 136	93.55 122	96.61 87	95.47 108	97.24 187	98.67 97

EPNet_dtu		92.45 117	95.02 80	89.46 146	98.02 54	95.47 167	94.79 120	92.62 69	94.97 100	70.11 192	94.76 54	92.61 78	84.07 202	95.94 113	95.56 105	97.15 188	95.82 177
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
pmmvs6		85.98 194	84.89 202	87.25 182	88.83 197	94.35 196	89.36 193	85.30 164	78.51 215	75.44 164	62.71 214	75.41 172	87.65 180	93.58 159	92.40 181	96.89 189	97.29 154

CVMVSNet		89.77 151	91.66 140	87.56 178	93.21 141	95.45 168	91.94 172	89.22 116	89.62 171	69.34 198	83.99 142	85.90 119	84.81 197	94.30 150	95.28 113	96.85 190	97.09 158

DeepPCF-MVS	95.28 2	97.00 41	98.35 21	95.42 61	97.30 63	98.94 51	94.82 119	96.03 40	98.24 9	92.11 51	95.80 41	98.64 33	95.51 87	98.95 7	98.66 5	96.78 191	99.20 42

pmnet_mix02		86.12 193	87.12 187	84.96 196	89.82 174	94.12 199	84.88 206	86.63 145	91.78 151	65.60 204	80.76 157	76.98 167	86.61 186	87.29 208	84.80 211	96.21 192	94.09 192

CHOSEN 280x420		95.46 61	97.01 46	93.66 97	97.28 64	97.98 98	96.40 86	85.39 161	96.10 69	91.07 59	96.53 34	96.34 56	95.61 84	97.65 56	96.95 68	96.21 192	97.49 147

new-patchmatchnet		78.49 208	78.19 211	78.84 208	84.13 212	90.06 212	77.11 217	80.39 192	79.57 214	59.64 218	66.01 210	55.65 222	75.62 209	84.55 211	80.70 213	96.14 194	90.77 209

EPMVS		90.88 135	92.12 134	89.44 147	94.71 111	97.24 112	93.55 136	76.81 200	95.89 75	81.77 130	91.49 84	86.47 114	93.87 114	90.21 197	90.07 194	95.92 195	93.49 201

SCA		90.92 134	93.04 114	88.45 156	93.72 134	97.33 111	92.77 149	76.08 205	96.02 71	78.26 146	91.96 76	90.86 85	93.99 113	90.98 194	90.04 195	95.88 196	94.06 194

dps		90.11 148	89.37 161	90.98 125	93.89 130	96.21 141	93.49 138	77.61 198	91.95 149	92.74 47	88.85 106	78.77 159	92.37 134	87.71 206	87.71 203	95.80 197	94.38 189

ADS-MVSNet		89.80 150	91.33 144	88.00 168	94.43 120	96.71 127	92.29 161	74.95 210	96.07 70	77.39 149	88.67 109	86.09 117	93.26 126	88.44 203	89.57 197	95.68 198	93.81 198

tpm		87.95 173	89.44 160	86.21 191	92.53 148	94.62 192	91.40 176	76.36 203	91.46 154	69.80 196	87.43 114	75.14 173	91.55 143	89.85 201	90.60 191	95.61 199	96.96 163

EU-MVSNet		85.62 195	87.65 181	83.24 201	88.54 200	92.77 206	87.12 199	85.32 162	86.71 193	64.54 206	78.52 166	75.11 174	78.35 206	92.25 180	92.28 183	95.58 200	95.93 175

CostFormer		90.69 136	90.48 153	90.93 126	94.18 124	96.08 144	94.03 131	78.20 196	93.47 124	89.96 80	90.97 91	80.30 152	93.72 118	87.66 207	88.75 199	95.51 201	96.12 174

PatchmatchNet		90.56 138	92.49 124	88.31 159	93.83 132	96.86 121	92.42 157	76.50 202	95.96 73	78.31 145	91.96 76	89.66 94	93.48 123	90.04 199	89.20 198	95.32 202	93.73 199
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
N_pmnet		84.80 197	85.10 201	84.45 197	89.25 189	92.86 205	84.04 207	86.21 148	88.78 176	66.73 202	72.41 196	74.87 177	85.21 194	88.32 204	86.45 206	95.30 203	92.04 205

RPSCF		94.05 91	94.00 97	94.12 90	96.20 76	96.41 136	96.61 78	91.54 85	95.83 79	89.73 83	96.94 31	92.80 76	95.35 90	91.63 190	90.44 192	95.27 204	93.94 195

MDTV_nov1_ep13_2view		86.30 191	88.27 168	84.01 198	87.71 205	94.67 190	88.08 196	76.78 201	90.59 165	68.66 200	80.46 161	80.12 153	87.58 182	89.95 200	88.20 201	95.25 205	93.90 197

MDTV_nov1_ep13		91.57 126	93.18 112	89.70 143	93.39 137	96.97 116	93.53 137	80.91 191	95.70 81	81.86 129	92.40 71	89.93 92	93.25 127	91.97 187	90.80 190	95.25 205	94.46 188

new_pmnet		81.53 204	82.68 206	80.20 204	83.47 213	89.47 214	82.21 212	78.36 194	87.86 185	60.14 217	67.90 207	69.43 202	82.03 204	89.22 202	87.47 204	94.99 207	87.39 213

MVS-HIRNet		85.36 196	86.89 189	83.57 199	90.13 170	94.51 193	83.57 209	72.61 213	88.27 183	71.22 186	68.97 204	81.81 147	88.91 176	93.08 169	91.94 185	94.97 208	89.64 211

tpmrst		88.86 165	89.62 156	87.97 169	94.33 121	95.98 146	92.62 153	76.36 203	94.62 106	76.94 153	85.98 129	82.80 143	92.80 131	86.90 209	87.15 205	94.77 209	93.93 196

pmmvs3		79.16 207	80.12 209	78.05 209	79.36 215	86.59 216	78.13 216	73.87 212	76.42 217	57.51 220	70.59 203	57.02 219	84.66 198	90.10 198	88.32 200	94.75 210	91.77 207

tpm cat1		88.90 163	87.78 179	90.22 136	93.88 131	95.39 171	93.79 134	78.11 197	92.55 137	89.43 87	81.31 154	79.84 155	91.40 144	84.95 210	86.34 208	94.68 211	94.09 192

CMPMVS	65.18 17	84.76 198	83.10 204	86.69 187	95.29 94	95.05 180	88.37 195	85.51 160	80.27 213	71.31 185	68.37 206	73.85 180	85.25 193	87.72 205	87.75 202	94.38 212	88.70 212
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
MDA-MVSNet-bldmvs		80.11 205	80.24 208	79.94 205	77.01 217	93.21 203	78.86 215	85.94 154	82.71 210	60.86 212	79.71 163	51.77 223	83.71 203	75.60 215	86.37 207	93.28 213	92.35 203

ambc			73.83 213		76.23 218	85.13 217	82.27 211		84.16 205	65.58 205	52.82 217	23.31 228	73.55 211	91.41 192	85.26 210	92.97 214	94.70 185

PMMVS2		64.36 214	65.94 216	62.52 215	67.37 221	77.44 219	64.39 221	69.32 219	61.47 220	34.59 223	46.09 218	41.03 224	48.02 221	74.56 217	78.23 214	91.43 215	82.76 215

DeepMVS_CX						86.86 215	79.50 214	70.43 216	90.73 161	63.66 207	80.36 162	60.83 216	79.68 205	76.23 214		89.46 216	86.53 214

Gipuma		68.35 211	66.71 214	70.27 211	74.16 219	68.78 221	63.93 222	71.77 215	83.34 208	54.57 221	34.37 219	31.88 225	68.69 213	83.30 212	85.53 209	88.48 217	79.78 217
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
test_method		72.96 210	78.68 210	66.28 214	50.17 224	64.90 222	75.45 218	50.90 221	87.89 184	62.54 211	62.98 213	68.34 206	70.45 212	91.90 188	82.41 212	88.19 218	92.35 203

FPMVS		75.84 209	74.59 212	77.29 210	86.92 207	83.89 218	85.01 205	80.05 193	82.91 209	60.61 214	65.25 211	60.41 217	63.86 215	75.60 215	73.60 217	87.29 219	80.47 216

PMVS	63.12 18	67.27 212	66.39 215	68.30 212	77.98 216	60.24 223	59.53 223	76.82 199	66.65 219	60.74 213	54.39 216	59.82 218	51.24 218	73.92 218	70.52 218	83.48 220	79.17 218
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
tmp_tt				66.88 213	86.07 209	73.86 220	68.22 220	33.38 222	96.88 48	80.67 137	88.23 112	78.82 158	49.78 219	82.68 213	77.47 215	83.19 221

E-PMN		50.67 215	47.85 218	53.96 216	64.13 223	50.98 226	38.06 224	69.51 217	51.40 222	24.60 225	29.46 222	24.39 227	56.07 217	48.17 220	59.70 219	71.40 222	70.84 220

EMVS		49.98 216	46.76 219	53.74 217	64.96 222	51.29 225	37.81 225	69.35 218	51.83 221	22.69 226	29.57 221	25.06 226	57.28 216	44.81 221	56.11 220	70.32 223	68.64 221

MVE	50.86 19	49.54 217	51.43 217	47.33 218	44.14 225	59.20 224	36.45 226	60.59 220	41.47 223	31.14 224	29.58 220	17.06 229	48.52 220	62.22 219	74.63 216	63.12 224	75.87 219
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
testmvs		12.09 218	16.94 220	6.42 220	3.15 226	6.08 227	9.51 228	3.84 223	21.46 224	5.31 227	27.49 223	6.76 230	10.89 222	17.06 222	15.01 221	5.84 225	24.75 222

test123		9.58 219	13.53 221	4.97 221	1.31 228	5.47 228	8.32 229	2.95 224	18.14 225	2.03 229	20.82 224	2.34 231	10.60 223	10.00 223	14.16 222	4.60 226	23.77 223

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										63.50 209

9.14												99.28 12

SR-MVS					99.45 10			97.61 16				99.20 16

our_test_3					89.78 175	93.84 201	85.59 203

MTAPA										96.83 11		99.12 21

MTMP										97.18 5		98.83 27

Patchmatch-RL test							34.61 227

mPP-MVS					99.21 25							98.29 39

NP-MVS									95.32 89

Patchmtry						95.96 148	93.36 141	75.99 206		75.19 167

High-resolution multi-view benchmark