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.

LTVRE_ROB	98.82 1	99.76 1	99.75 1	99.77 7	99.87 16	99.71 10	99.77 8	99.76 19	99.52 2	99.80 3	99.79 21	99.91 1	99.56 13	99.83 3	99.75 4	99.86 9	99.75 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
pmmvs6		99.74 2	99.75 1	99.73 11	99.92 5	99.67 15	99.76 10	99.84 11	99.59 1	99.52 24	99.87 11	99.91 1	99.43 27	99.87 1	99.81 2	99.89 6	99.52 10

SixPastTwentyTwo		99.70 3	99.59 4	99.82 2	99.93 3	99.80 1	99.86 2	99.87 6	98.87 12	99.79 5	99.85 14	99.33 66	99.74 5	99.85 2	99.82 1	99.74 24	99.63 5

v7n		99.68 4	99.61 3	99.76 8	99.89 12	99.74 7	99.87 1	99.82 13	99.20 6	99.71 6	99.96 1	99.73 13	99.76 3	99.58 20	99.59 16	99.52 47	99.46 15

anonymousdsp		99.64 5	99.55 6	99.74 10	99.87 16	99.56 26	99.82 3	99.73 23	98.54 17	99.71 6	99.92 4	99.84 7	99.61 9	99.70 9	99.63 9	99.69 33	99.64 3

UniMVSNet_ETH3D		99.61 6	99.59 4	99.63 13	99.96 1	99.70 11	99.53 35	99.86 8	99.28 5	99.48 30	99.44 54	99.86 5	99.01 69	99.78 4	99.76 3	99.90 2	99.33 21

WR-MVS		99.61 6	99.44 8	99.82 2	99.92 5	99.80 1	99.80 4	99.89 1	98.54 17	99.66 13	99.78 22	99.16 87	99.68 7	99.70 9	99.63 9	99.94 1	99.49 13

PEN-MVS		99.54 8	99.30 16	99.83 1	99.92 5	99.76 4	99.80 4	99.88 3	97.60 63	99.71 6	99.59 36	99.52 44	99.75 4	99.64 15	99.51 19	99.90 2	99.46 15

TDRefinement		99.54 8	99.50 7	99.60 17	99.70 68	99.35 46	99.77 8	99.58 51	99.40 4	99.28 49	99.66 26	99.41 55	99.55 15	99.74 8	99.65 8	99.70 30	99.25 26

DTE-MVSNet		99.52 10	99.27 17	99.82 2	99.93 3	99.77 3	99.79 6	99.87 6	97.89 45	99.70 11	99.55 45	99.21 78	99.77 2	99.65 13	99.43 23	99.90 2	99.36 19

PS-CasMVS		99.50 11	99.23 20	99.82 2	99.92 5	99.75 6	99.78 7	99.89 1	97.30 74	99.71 6	99.60 34	99.23 74	99.71 6	99.65 13	99.55 18	99.90 2	99.56 8

WR-MVS_H		99.48 12	99.23 20	99.76 8	99.91 9	99.76 4	99.75 11	99.88 3	97.27 77	99.58 17	99.56 41	99.24 73	99.56 13	99.60 18	99.60 15	99.88 8	99.58 7

pm-mvs1		99.47 13	99.38 9	99.57 21	99.82 28	99.49 30	99.63 23	99.65 39	98.88 11	99.31 43	99.85 14	99.02 106	99.23 46	99.60 18	99.58 17	99.80 15	99.22 33

MIMVSNet1		99.46 14	99.34 10	99.60 17	99.83 23	99.68 14	99.74 14	99.71 27	98.20 27	99.41 35	99.86 13	99.66 25	99.41 30	99.50 24	99.39 26	99.50 52	99.10 44

TransMVSNet (Re)		99.45 15	99.32 13	99.61 15	99.88 14	99.60 21	99.75 11	99.63 43	99.11 7	99.28 49	99.83 18	98.35 141	99.27 43	99.70 9	99.62 13	99.84 10	99.03 52

ACMH	97.81 6	99.44 16	99.33 11	99.56 22	99.81 32	99.42 37	99.73 15	99.58 51	99.02 8	99.10 73	99.41 59	99.69 19	99.60 10	99.45 28	99.26 37	99.55 43	99.05 49
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
CP-MVSNet		99.39 17	99.04 29	99.80 6	99.91 9	99.70 11	99.75 11	99.88 3	96.82 98	99.68 12	99.32 63	98.86 115	99.68 7	99.57 21	99.47 20	99.89 6	99.52 10

COLMAP_ROB	98.29 2	99.37 18	99.25 18	99.51 30	99.74 59	99.12 74	99.56 32	99.39 88	98.96 10	99.17 61	99.44 54	99.63 32	99.58 11	99.48 26	99.27 36	99.60 40	98.81 79
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
DeepC-MVS	97.88 4	99.33 19	99.15 24	99.53 29	99.73 64	99.05 82	99.49 40	99.40 86	98.42 20	99.55 21	99.71 24	99.89 3	99.49 19	99.14 44	98.81 68	99.54 44	99.02 54
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
FC-MVSNet-test		99.32 20	99.33 11	99.31 57	99.87 16	99.65 18	99.63 23	99.75 21	97.76 47	97.29 197	99.87 11	99.63 32	99.52 16	99.66 12	99.63 9	99.77 20	99.12 40

UA-Net		99.30 21	99.22 22	99.39 44	99.94 2	99.66 17	98.91 111	99.86 8	97.74 53	98.74 115	99.00 90	99.60 37	99.17 54	99.50 24	99.39 26	99.70 30	99.64 3

ACMH+	97.53 7	99.29 22	99.20 23	99.40 43	99.81 32	99.22 63	99.59 29	99.50 69	98.64 16	98.29 149	99.21 75	99.69 19	99.57 12	99.53 23	99.33 31	99.66 34	98.81 79

Vis-MVSNet		99.25 23	99.32 13	99.17 67	99.65 79	99.55 28	99.63 23	99.33 104	98.16 28	99.29 46	99.65 30	99.77 10	97.56 145	99.44 30	99.14 42	99.58 41	99.51 12
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
TranMVSNet+NR-MVSNet		99.23 24	98.91 36	99.61 15	99.81 32	99.45 34	99.47 42	99.68 30	97.28 76	99.39 36	99.54 46	99.08 102	99.45 22	99.09 50	98.84 65	99.83 11	99.04 50

CSCG		99.23 24	99.15 24	99.32 56	99.83 23	99.45 34	98.97 103	99.21 125	98.83 13	99.04 83	99.43 56	99.64 30	99.26 44	98.85 76	98.20 104	99.62 38	99.62 6

Gipuma		99.22 26	98.86 40	99.64 12	99.70 68	99.24 57	99.17 85	99.63 43	99.52 2	99.89 1	96.54 176	99.14 91	99.93 1	99.42 32	99.15 41	99.52 47	99.04 50
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
tfpnnormal		99.19 27	98.90 37	99.54 26	99.81 32	99.55 28	99.60 27	99.54 59	98.53 19	99.23 53	98.40 111	98.23 144	99.40 31	99.29 37	99.36 29	99.63 37	98.95 65

Baseline_NR-MVSNet		99.18 28	98.87 38	99.54 26	99.74 59	99.56 26	99.36 57	99.62 48	96.53 119	99.29 46	99.85 14	98.64 133	99.40 31	99.03 61	99.63 9	99.83 11	98.86 74

thisisatest0515		99.16 29	98.94 34	99.41 38	99.75 53	99.43 36	99.36 57	99.63 43	97.68 59	99.35 38	99.31 64	98.90 112	99.09 63	98.95 66	99.20 38	99.27 84	99.11 41

CS-MVS-test		99.16 29	98.78 45	99.60 17	99.80 38	99.72 9	99.69 16	99.73 23	95.88 141	99.51 26	98.53 108	99.54 42	99.21 48	99.24 40	99.43 23	99.66 34	99.15 39

CS-MVS		99.15 31	98.75 47	99.62 14	99.76 49	99.73 8	99.60 27	99.75 21	95.67 148	99.50 27	98.53 108	99.39 60	99.29 40	99.21 42	99.46 22	99.79 18	99.29 24

APDe-MVS		99.15 31	98.95 31	99.39 44	99.77 44	99.28 54	99.52 36	99.54 59	97.22 81	99.06 77	99.20 76	99.64 30	99.05 67	99.14 44	99.02 52	99.39 66	99.17 37
Zhaojie Zeng, Yuesong Wang, Tao Guan: Matching Ambiguity-Resilient Multi-View Stereo via Adaptive Patch Deformation. Pattern Recognition
WB-MVS		99.14 33	99.31 15	98.95 96	99.81 32	99.61 20	98.85 118	99.51 66	99.01 9	97.37 192	99.33 62	99.56 40	98.70 87	99.44 30	99.29 33	99.45 57	98.96 64

FC-MVSNet-train		99.13 34	99.05 28	99.21 62	99.87 16	99.57 25	99.67 18	99.60 50	96.75 104	98.28 150	99.48 50	99.52 44	98.10 123	99.47 27	99.37 28	99.76 22	99.21 34

NR-MVSNet		99.10 35	98.68 57	99.58 20	99.89 12	99.23 60	99.35 61	99.63 43	96.58 112	99.36 37	99.05 84	98.67 131	99.46 20	99.63 16	98.73 78	99.80 15	98.88 73

DVP-MVS++		99.09 36	99.25 18	98.90 102	99.53 108	99.37 44	99.17 85	99.48 74	98.28 25	97.95 170	99.54 46	99.88 4	98.13 122	99.08 51	98.94 56	99.15 97	99.65 2

DVP-MVS		99.09 36	99.07 27	99.12 74	99.55 101	99.40 39	99.36 57	99.44 85	97.75 50	98.23 153	99.23 72	99.80 8	98.97 71	99.08 51	98.96 53	99.19 92	99.25 26
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 (Re)		99.08 38	98.69 55	99.54 26	99.75 53	99.33 49	99.29 69	99.64 42	96.75 104	99.48 30	99.30 66	98.69 126	99.26 44	98.94 68	98.76 74	99.78 19	99.02 54

ACMMPR		99.05 39	98.72 51	99.44 32	99.79 39	99.12 74	99.35 61	99.56 54	97.74 53	99.21 55	97.72 138	99.55 41	99.29 40	98.90 74	98.81 68	99.41 65	99.19 35

DU-MVS		99.04 40	98.59 61	99.56 22	99.74 59	99.23 60	99.29 69	99.63 43	96.58 112	99.55 21	99.05 84	98.68 128	99.36 35	99.03 61	98.60 85	99.77 20	98.97 59

TSAR-MVS + MP.		99.02 41	98.95 31	99.11 77	99.23 158	98.79 118	99.51 37	98.73 166	97.50 67	98.56 126	99.03 87	99.59 38	99.16 56	99.29 37	99.17 40	99.50 52	99.24 30
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
v10		99.01 42	98.66 58	99.41 38	99.52 113	99.39 40	99.57 31	99.66 37	97.59 64	99.32 42	99.88 9	99.23 74	99.50 18	97.77 141	97.98 115	98.92 127	98.78 84

EG-PatchMatch MVS		99.01 42	98.77 46	99.28 61	99.64 82	98.90 111	98.81 124	99.27 115	96.55 116	99.71 6	99.31 64	99.66 25	99.17 54	99.28 39	99.11 44	99.10 99	98.57 100

PVSNet_Blended_VisFu		98.98 44	98.79 43	99.21 62	99.76 49	99.34 47	99.35 61	99.35 100	97.12 87	99.46 32	99.56 41	98.89 113	98.08 127	99.05 55	98.58 87	99.27 84	98.98 58

HFP-MVS		98.97 45	98.70 53	99.29 59	99.67 73	98.98 94	99.13 91	99.53 62	97.76 47	98.90 98	98.07 126	99.50 50	99.14 59	98.64 88	98.78 72	99.37 68	99.18 36

UniMVSNet_NR-MVSNet		98.97 45	98.46 71	99.56 22	99.76 49	99.34 47	99.29 69	99.61 49	96.55 116	99.55 21	99.05 84	97.96 152	99.36 35	98.84 77	98.50 93	99.81 14	98.97 59

casdiffmvs_mvg		98.96 47	98.87 38	99.07 80	99.82 28	99.36 45	99.36 57	99.22 122	98.13 30	97.74 177	99.42 57	99.46 53	98.59 95	98.39 100	98.95 55	99.71 29	98.39 118
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
EC-MVSNet		98.96 47	98.45 74	99.56 22	99.88 14	99.70 11	99.68 17	99.78 16	94.15 182	98.97 87	98.26 118	99.21 78	99.35 37	99.30 36	99.14 42	99.73 25	99.40 18

SED-MVS		98.94 49	98.95 31	98.91 101	99.43 129	99.38 42	99.12 93	99.46 79	97.05 90	98.43 141	99.23 72	99.79 9	97.99 130	99.05 55	98.94 56	99.05 113	99.23 31

ACMMP_NAP		98.94 49	98.72 51	99.21 62	99.67 73	99.08 77	99.26 74	99.39 88	96.84 95	98.88 102	98.22 119	99.68 21	98.82 80	99.06 54	98.90 59	99.25 87	99.25 26

v1144		98.94 49	98.53 66	99.42 36	99.62 86	99.03 88	99.58 30	99.36 97	97.99 36	99.49 29	99.91 8	99.20 81	99.51 17	97.61 146	97.85 122	98.95 122	98.10 141

v8		98.94 49	98.60 59	99.35 53	99.54 105	99.39 40	99.55 33	99.67 34	97.48 68	99.13 69	99.81 19	99.10 98	99.39 33	97.86 136	97.89 120	98.81 136	98.66 93

SteuartSystems-ACMMP		98.94 49	98.52 67	99.43 35	99.79 39	99.13 73	99.33 65	99.55 56	96.17 134	99.04 83	97.53 144	99.65 29	99.46 20	99.04 60	98.76 74	99.44 60	99.35 20
Skip Steuart: Steuart Systems R&D Blog.
v1192		98.91 54	98.48 70	99.41 38	99.61 90	99.03 88	99.64 20	99.25 119	97.91 42	99.58 17	99.92 4	99.07 104	99.45 22	97.55 150	97.68 136	98.93 124	98.23 131

FMVSNet1		98.90 55	99.10 26	98.67 127	99.54 105	99.48 31	99.22 79	99.66 37	98.39 23	97.50 184	99.66 26	99.04 105	96.58 166	99.05 55	99.03 49	99.52 47	99.08 46

ACMM	96.66 11	98.90 55	98.44 76	99.44 32	99.74 59	98.95 100	99.47 42	99.55 56	97.66 61	99.09 74	96.43 178	99.41 55	99.35 37	98.95 66	98.67 81	99.45 57	99.03 52
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
Anonymous20231211		98.89 57	98.79 43	98.99 93	99.82 28	99.41 38	99.18 84	99.31 110	96.92 92	98.54 128	98.58 106	98.84 118	97.46 147	99.45 28	99.29 33	99.65 36	99.08 46

v1921920		98.89 57	98.46 71	99.39 44	99.58 94	99.04 86	99.64 20	99.17 131	97.91 42	99.64 15	99.92 4	98.99 110	99.44 25	97.44 157	97.57 145	98.84 134	98.35 121

GeoE		98.88 59	98.43 79	99.41 38	99.83 23	99.24 57	99.51 37	99.82 13	96.55 116	99.22 54	98.76 98	99.22 77	98.96 72	98.55 91	98.15 106	99.10 99	98.56 103

v144192		98.88 59	98.46 71	99.37 51	99.56 100	99.03 88	99.61 26	99.26 116	97.79 46	99.58 17	99.88 9	99.11 96	99.43 27	97.38 162	97.61 141	98.80 137	98.43 115

SMA-MVS		98.87 61	98.73 50	99.04 86	99.72 65	99.05 82	98.64 135	99.17 131	96.31 129	98.80 109	99.07 82	99.70 18	98.67 89	98.93 71	98.82 66	99.23 90	99.23 31
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
ACMP	96.54 13	98.87 61	98.40 82	99.41 38	99.74 59	98.88 112	99.29 69	99.50 69	96.85 94	98.96 90	97.05 160	99.66 25	99.43 27	98.98 65	98.60 85	99.52 47	98.81 79
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
DCV-MVSNet		98.86 63	98.57 64	99.19 65	99.86 20	99.67 15	99.39 51	99.71 27	97.53 66	98.69 118	95.85 189	98.48 136	97.75 139	99.57 21	99.41 25	99.72 26	99.48 14

v1240		98.86 63	98.41 81	99.38 49	99.59 92	99.05 82	99.65 19	99.14 136	97.68 59	99.66 13	99.93 3	98.72 125	99.45 22	97.38 162	97.72 134	98.79 138	98.35 121

CP-MVS		98.86 63	98.43 79	99.36 52	99.68 71	98.97 98	99.19 82	99.46 79	96.60 110	99.20 56	97.11 159	99.51 48	99.15 58	98.92 72	98.82 66	99.45 57	99.08 46

v2v482		98.85 66	98.40 82	99.38 49	99.65 79	98.98 94	99.55 33	99.39 88	97.92 41	99.35 38	99.85 14	99.14 91	99.39 33	97.50 152	97.78 125	98.98 119	97.60 157

DPE-MVS		98.84 67	98.69 55	99.00 90	99.05 177	99.26 55	99.19 82	99.35 100	95.85 143	98.74 115	99.27 68	99.66 25	98.30 115	98.90 74	98.93 58	99.37 68	99.00 56
Kehua Chen, Zhenlong Yuan, Tianlu Mao, Zhaoqi Wang: Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization. AAAI2025
OPM-MVS		98.84 67	98.59 61	99.12 74	99.52 113	98.50 143	99.13 91	99.22 122	97.76 47	98.76 111	98.70 100	99.61 35	98.90 75	98.67 86	98.37 98	99.19 92	98.57 100
Ray L. Khuboni, Hongjun Xu: Octagram Propagation Matching for Multi-Scale View Stereopsis (OPM-MVS).
test20.03		98.84 67	98.74 49	98.95 96	99.77 44	99.33 49	99.21 81	99.46 79	97.29 75	98.88 102	99.65 30	99.10 98	97.07 157	99.11 47	98.76 74	99.32 77	97.98 145

casdiffmvs		98.84 67	98.75 47	98.94 100	99.75 53	99.21 64	99.33 65	99.04 146	98.04 32	97.46 187	99.72 23	99.72 15	98.60 93	98.30 112	98.37 98	99.48 54	97.92 147
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
LGP-MVS_train		98.84 67	98.33 88	99.44 32	99.78 42	98.98 94	99.39 51	99.55 56	95.41 153	98.90 98	97.51 145	99.68 21	99.44 25	99.03 61	98.81 68	99.57 42	98.91 69

RPSCF		98.84 67	98.81 42	98.89 104	99.37 137	98.95 100	98.51 147	98.85 159	97.73 55	98.33 146	98.97 92	99.14 91	98.95 73	99.18 43	98.68 80	99.31 78	98.99 57

ACMMP		98.82 73	98.33 88	99.39 44	99.77 44	99.14 72	99.37 54	99.54 59	96.47 123	99.03 85	96.26 182	99.52 44	99.28 42	98.92 72	98.80 71	99.37 68	99.16 38
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
V42		98.81 74	98.49 69	99.18 66	99.52 113	98.92 106	99.50 39	99.29 112	97.43 71	98.97 87	99.81 19	99.00 109	99.30 39	97.93 132	98.01 113	98.51 162	98.34 125

LS3D		98.79 75	98.52 67	99.12 74	99.64 82	99.09 76	99.24 77	99.46 79	97.75 50	98.93 96	97.47 147	98.23 144	97.98 131	99.36 33	99.30 32	99.46 55	98.42 116

MP-MVS		98.78 76	98.30 90	99.34 55	99.75 53	98.95 100	99.26 74	99.46 79	95.78 147	99.17 61	96.98 164	99.72 15	99.06 66	98.84 77	98.74 77	99.33 74	99.11 41
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
v148		98.77 77	98.45 74	99.15 70	99.68 71	98.94 104	99.49 40	99.31 110	97.95 38	98.91 97	99.65 30	99.62 34	99.18 51	97.99 130	97.64 140	98.33 167	97.38 162

test1111		98.75 78	98.14 103	99.46 31	99.86 20	99.63 19	99.47 42	99.68 30	98.34 24	98.76 111	99.66 26	90.92 196	99.23 46	99.77 5	99.71 5	99.75 23	98.95 65

ECVR-MVS		98.74 79	98.15 101	99.42 36	99.83 23	99.58 23	99.37 54	99.67 34	98.02 34	98.85 106	99.59 36	91.66 194	99.10 61	99.77 5	99.70 6	99.72 26	98.73 86

SD-MVS		98.73 80	98.54 65	98.95 96	99.14 167	98.76 121	98.46 151	99.14 136	97.71 57	98.56 126	98.06 128	99.61 35	98.85 79	98.56 90	97.74 131	99.54 44	99.32 22
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
MSP-MVS		98.72 81	98.60 59	98.87 106	99.67 73	99.33 49	99.15 88	99.26 116	96.99 91	97.90 173	98.19 121	99.74 12	98.29 116	97.69 144	98.96 53	98.96 120	99.27 25
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
PGM-MVS		98.69 82	98.09 108	99.39 44	99.76 49	99.07 78	99.30 68	99.51 66	94.76 164	99.18 60	96.70 171	99.51 48	99.20 49	98.79 82	98.71 79	99.39 66	99.11 41

pmmvs-eth3d		98.68 83	98.14 103	99.29 59	99.49 118	98.45 146	99.45 47	99.38 93	97.21 82	99.50 27	99.65 30	99.21 78	99.16 56	97.11 169	97.56 146	98.79 138	97.82 151

EU-MVSNet		98.68 83	98.94 34	98.37 147	99.14 167	98.74 123	99.64 20	98.20 191	98.21 26	99.17 61	99.66 26	99.18 84	99.08 64	99.11 47	98.86 61	95.00 204	98.83 76

PMVS	92.51 17	98.66 85	98.86 40	98.43 143	99.26 153	98.98 94	98.60 141	98.59 175	97.73 55	99.45 33	99.38 60	98.54 135	95.24 184	99.62 17	99.61 14	99.42 62	98.17 138
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
DeepC-MVS_fast	97.38 8	98.65 86	98.34 87	99.02 89	99.33 141	98.29 153	98.99 101	98.71 168	97.40 72	99.31 43	98.20 120	99.40 58	98.54 102	98.33 109	98.18 105	99.23 90	98.58 98
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
3Dnovator	98.16 3	98.65 86	98.35 86	99.00 90	99.59 92	98.70 126	98.90 115	99.36 97	97.97 37	99.09 74	96.55 175	99.09 100	97.97 132	98.70 85	98.65 83	99.12 98	98.81 79

TSAR-MVS + ACMM		98.64 88	98.58 63	98.72 121	99.17 165	98.63 132	98.69 130	99.10 143	97.69 58	98.30 148	99.12 80	99.38 61	98.70 87	98.45 95	97.51 148	98.35 166	99.25 26

DELS-MVS		98.63 89	98.70 53	98.55 139	99.24 157	99.04 86	98.96 104	98.52 178	96.83 97	98.38 143	99.58 39	99.68 21	97.06 158	98.74 84	98.44 95	99.10 99	98.59 97
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
QAPM		98.62 90	98.40 82	98.89 104	99.57 99	98.80 117	98.63 136	99.35 100	96.82 98	98.60 122	98.85 97	99.08 102	98.09 125	98.31 110	98.21 102	99.08 105	98.72 87

EPP-MVSNet		98.61 91	98.19 98	99.11 77	99.86 20	99.60 21	99.44 48	99.53 62	97.37 73	96.85 202	98.69 101	93.75 187	99.18 51	99.22 41	99.35 30	99.82 13	99.32 22

3Dnovator+	97.85 5	98.61 91	98.14 103	99.15 70	99.62 86	98.37 150	99.10 94	99.51 66	98.04 32	98.98 86	96.07 186	98.75 124	98.55 100	98.51 93	98.40 96	99.17 94	98.82 77

X-MVS		98.59 93	97.99 114	99.30 58	99.75 53	99.07 78	99.17 85	99.50 69	96.62 108	98.95 92	93.95 205	99.37 62	99.11 60	98.94 68	98.86 61	99.35 72	99.09 45

MVS_111021_HR		98.58 94	98.26 93	98.96 95	99.32 144	98.81 115	98.48 149	98.99 151	96.81 100	99.16 64	98.07 126	99.23 74	98.89 77	98.43 97	98.27 101	98.90 129	98.24 130

MVS_0304		98.57 95	98.36 85	98.82 113	99.72 65	98.94 104	98.92 109	99.14 136	96.76 103	99.33 41	98.30 115	99.73 13	96.74 162	98.05 127	97.79 124	99.08 105	98.97 59

PM-MVS		98.57 95	98.24 95	98.95 96	99.26 153	98.59 135	99.03 98	98.74 165	96.84 95	99.44 34	99.13 79	98.31 143	98.75 85	98.03 128	98.21 102	98.48 163	98.58 98

PHI-MVS		98.57 95	98.20 97	99.00 90	99.48 120	98.91 108	98.68 131	99.17 131	94.97 160	99.27 51	98.33 113	99.33 66	98.05 128	98.82 80	98.62 84	99.34 73	98.38 119

HPM-MVS++		98.56 98	98.08 109	99.11 77	99.53 108	98.61 134	99.02 100	99.32 108	96.29 131	99.06 77	97.23 154	99.50 50	98.77 83	98.15 123	97.90 118	98.96 120	98.90 70

TSAR-MVS + GP.		98.54 99	98.29 92	98.82 113	99.28 151	98.59 135	97.73 190	99.24 121	95.93 140	98.59 123	99.07 82	99.17 85	98.86 78	98.44 96	98.10 108	99.26 86	98.72 87

UGNet		98.52 100	99.00 30	97.96 168	99.58 94	99.26 55	99.27 73	99.40 86	98.07 31	98.28 150	98.76 98	99.71 17	92.24 212	98.94 68	98.85 63	99.00 118	99.43 17
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
Anonymous20231206		98.50 101	98.03 111	99.05 84	99.50 116	99.01 91	99.15 88	99.26 116	96.38 127	99.12 71	99.50 49	99.12 94	98.60 93	97.68 145	97.24 159	98.66 146	97.30 166

CLD-MVS		98.48 102	98.15 101	98.86 109	99.53 108	98.35 151	98.55 144	97.83 200	96.02 139	98.97 87	99.08 81	99.75 11	99.03 68	98.10 126	97.33 155	99.28 82	98.44 114
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
CANet		98.47 103	98.30 90	98.67 127	99.65 79	98.87 113	98.82 123	99.01 149	96.14 135	99.29 46	98.86 95	99.01 107	96.54 167	98.36 104	98.08 110	98.72 142	98.80 83

APD-MVS		98.47 103	97.97 115	99.05 84	99.64 82	98.91 108	98.94 106	99.45 84	94.40 175	98.77 110	97.26 153	99.41 55	98.21 119	98.67 86	98.57 90	99.31 78	98.57 100
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
Vis-MVSNet (Re-imp)		98.46 105	98.23 96	98.73 120	99.81 32	99.29 53	98.79 125	99.50 69	96.20 133	96.03 208	98.29 116	96.98 167	98.54 102	99.11 47	99.08 45	99.70 30	98.62 95

Fast-Effi-MVS+		98.42 106	97.79 121	99.15 70	99.69 70	98.66 130	98.94 106	99.68 30	94.49 169	99.05 79	98.06 128	98.86 115	98.48 105	98.18 120	97.78 125	99.05 113	98.54 106

ETV-MVS		98.41 107	97.76 122	99.17 67	99.58 94	99.01 91	98.91 111	99.50 69	93.33 195	99.31 43	96.82 168	98.42 139	98.17 121	99.13 46	99.08 45	99.54 44	98.56 103

MVS_111021_LR		98.39 108	98.11 106	98.71 123	99.08 174	98.54 141	98.23 172	98.56 177	96.57 114	99.13 69	98.41 110	98.86 115	98.65 91	98.23 118	97.87 121	98.65 148	98.28 127

pmmvs5		98.37 109	97.81 120	99.03 87	99.46 122	98.97 98	99.03 98	98.96 153	95.85 143	99.05 79	99.45 53	98.66 132	98.79 82	96.02 186	97.52 147	98.87 131	98.21 134

OMC-MVS		98.35 110	98.10 107	98.64 133	98.85 184	97.99 172	98.56 143	98.21 189	97.26 79	98.87 104	98.54 107	99.27 72	98.43 107	98.34 107	97.66 137	98.92 127	97.65 156

canonicalmvs		98.34 111	97.92 117	98.83 111	99.45 124	99.21 64	98.37 159	99.53 62	97.06 89	97.74 177	96.95 166	95.05 184	98.36 110	98.77 83	98.85 63	99.51 51	99.53 9

CHOSEN 1792x2688		98.31 112	98.02 112	98.66 129	99.55 101	98.57 138	99.38 53	99.25 119	98.42 20	98.48 136	99.58 39	99.85 6	98.31 114	95.75 189	95.71 184	96.96 191	98.27 129

CPTT-MVS		98.28 113	97.51 135	99.16 69	99.54 105	98.78 119	98.96 104	99.36 97	96.30 130	98.89 101	93.10 209	99.30 69	99.20 49	98.35 106	97.96 116	99.03 116	98.82 77

TinyColmap		98.27 114	97.62 132	99.03 87	99.29 149	97.79 181	98.92 109	98.95 154	97.48 68	99.52 24	98.65 103	97.86 154	98.90 75	98.34 107	97.27 157	98.64 149	95.97 186

diffmvs		98.26 115	98.16 99	98.39 145	99.61 90	98.78 119	98.79 125	98.61 173	97.94 39	97.11 201	99.51 48	99.52 44	97.61 143	96.55 178	96.93 165	98.61 151	97.87 149
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
USDC		98.26 115	97.57 133	99.06 81	99.42 132	97.98 174	98.83 120	98.85 159	97.57 65	99.59 16	99.15 78	98.59 134	98.99 70	97.42 158	96.08 183	98.69 145	96.23 184

SF-MVS		98.25 117	98.16 99	98.35 148	99.43 129	98.42 149	97.05 212	99.09 144	96.42 125	98.13 159	97.73 137	99.20 81	97.22 153	98.36 104	98.38 97	99.16 96	98.62 95

MCST-MVS		98.25 117	97.57 133	99.06 81	99.53 108	98.24 159	98.63 136	99.17 131	95.88 141	98.58 124	96.11 184	99.09 100	99.18 51	97.58 149	97.31 156	99.25 87	98.75 85

IterMVS-LS		98.23 119	97.66 128	98.90 102	99.63 85	99.38 42	99.07 95	99.48 74	97.75 50	98.81 108	99.37 61	94.57 186	97.88 136	96.54 179	97.04 162	98.53 159	98.97 59
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
TAPA-MVS	96.65 12	98.23 119	97.96 116	98.55 139	98.81 186	98.16 163	98.40 156	97.94 198	96.68 106	98.49 134	98.61 104	98.89 113	98.57 98	97.45 155	97.59 143	99.09 104	98.35 121
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
CNVR-MVS		98.22 121	97.76 122	98.76 118	99.33 141	98.26 157	98.48 149	98.88 157	96.22 132	98.47 138	95.79 190	99.33 66	98.35 111	98.37 103	97.99 114	99.03 116	98.38 119

IS_MVSNet		98.20 122	98.00 113	98.44 142	99.82 28	99.48 31	99.25 76	99.56 54	95.58 150	93.93 220	97.56 143	96.52 172	98.27 117	99.08 51	99.20 38	99.80 15	98.56 103

DeepPCF-MVS	96.68 10	98.20 122	98.26 93	98.12 161	97.03 223	98.11 166	98.44 153	97.70 202	96.77 102	98.52 130	98.91 93	99.17 85	98.58 97	98.41 99	98.02 112	98.46 164	98.46 111

MSDG		98.20 122	97.88 119	98.56 137	99.33 141	97.74 182	98.27 169	98.10 192	97.20 84	98.06 163	98.59 105	99.16 87	98.76 84	98.39 100	97.71 135	98.86 133	96.38 181

testgi		98.18 125	98.44 76	97.89 170	99.78 42	99.23 60	98.78 127	99.21 125	97.26 79	97.41 189	97.39 150	99.36 65	92.85 208	98.82 80	98.66 82	99.31 78	98.35 121

Effi-MVS+		98.11 126	97.29 141	99.06 81	99.62 86	98.55 139	98.16 175	99.80 15	94.64 165	99.15 67	96.59 173	97.43 160	98.44 106	97.46 154	97.90 118	99.17 94	98.45 113

FA-MVS(training)		98.08 127	97.68 126	98.56 137	99.14 167	98.69 127	98.41 154	99.83 12	95.85 143	98.57 125	97.95 133	96.92 169	96.85 160	98.51 93	98.09 109	98.54 157	97.74 152

HyFIR lowres test		98.08 127	97.16 150	99.14 73	99.72 65	98.91 108	99.41 49	99.58 51	97.93 40	98.82 107	99.24 70	95.81 178	98.73 86	95.16 200	95.13 193	98.60 153	97.94 146

EIA-MVS		98.03 129	97.20 147	98.99 93	99.66 76	99.24 57	98.53 146	99.52 65	91.56 211	99.25 52	95.34 194	98.78 121	97.72 140	98.38 102	98.58 87	99.28 82	98.54 106

train_agg		97.99 130	97.26 142	98.83 111	99.43 129	98.22 161	98.91 111	99.07 145	94.43 173	97.96 169	96.42 179	99.30 69	98.81 81	97.39 160	96.62 171	98.82 135	98.47 109

MSLP-MVS++		97.99 130	97.64 131	98.40 144	98.91 182	98.47 145	97.12 210	98.78 163	96.49 121	98.48 136	93.57 207	99.12 94	98.51 104	98.31 110	98.58 87	98.58 155	98.95 65

CDPH-MVS		97.99 130	97.23 145	98.87 106	99.58 94	98.29 153	98.83 120	99.20 127	93.76 189	98.11 161	96.11 184	99.16 87	98.23 118	97.80 139	97.22 160	99.29 81	98.28 127

FMVSNet2		97.94 133	98.08 109	97.77 176	98.71 190	99.21 64	98.62 138	99.47 76	96.62 108	96.37 207	99.20 76	97.70 156	94.39 195	97.39 160	97.75 130	99.08 105	98.70 90

PVSNet_BlendedMVS		97.93 134	97.66 128	98.25 154	99.30 146	98.67 128	98.31 164	97.95 196	94.30 179	98.75 113	97.63 140	98.76 122	96.30 174	98.29 113	97.78 125	98.93 124	98.18 136

PVSNet_Blended		97.93 134	97.66 128	98.25 154	99.30 146	98.67 128	98.31 164	97.95 196	94.30 179	98.75 113	97.63 140	98.76 122	96.30 174	98.29 113	97.78 125	98.93 124	98.18 136

OpenMVS	97.26 9	97.88 136	97.17 149	98.70 124	99.50 116	98.55 139	98.34 162	99.11 141	93.92 187	98.90 98	95.04 199	98.23 144	97.38 150	98.11 125	98.12 107	98.95 122	98.23 131

pmmvs4		97.87 137	97.02 154	98.86 109	99.20 159	97.68 185	98.89 116	99.03 147	96.57 114	99.12 71	99.03 87	97.26 164	98.42 108	95.16 200	96.34 175	98.53 159	97.10 173

NCCC		97.84 138	96.96 156	98.87 106	99.39 135	98.27 156	98.46 151	99.02 148	96.78 101	98.73 117	91.12 213	98.91 111	98.57 98	97.83 138	97.49 149	99.04 115	98.33 126

Effi-MVS+-dtu		97.78 139	97.37 139	98.26 152	99.25 155	98.50 143	97.89 184	99.19 130	94.51 167	98.16 157	95.93 187	98.80 120	95.97 177	98.27 117	97.38 152	99.10 99	98.23 131

MDA-MVSNet-bldmvs		97.75 140	97.26 142	98.33 149	99.35 140	98.45 146	99.32 67	97.21 207	97.90 44	99.05 79	99.01 89	96.86 170	99.08 64	99.36 33	92.97 203	95.97 200	96.25 183

CDS-MVSNet		97.75 140	97.68 126	97.83 174	99.08 174	98.20 162	98.68 131	98.61 173	95.63 149	97.80 175	99.24 70	96.93 168	94.09 200	97.96 131	97.82 123	98.71 143	97.99 143
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
CNLPA		97.75 140	97.26 142	98.32 151	98.58 198	97.86 177	97.80 186	98.09 193	96.49 121	98.49 134	96.15 183	98.08 147	98.35 111	98.00 129	97.03 163	98.61 151	97.21 170

PLC	95.63 15	97.73 143	97.01 155	98.57 136	99.10 171	97.80 180	97.72 191	98.77 164	96.34 128	98.38 143	93.46 208	98.06 148	98.66 90	97.90 134	97.65 139	98.77 140	97.90 148
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
MVS_Test		97.69 144	97.15 151	98.33 149	99.27 152	98.43 148	98.25 170	99.29 112	95.00 159	97.39 191	98.86 95	98.00 151	97.14 155	95.38 195	96.22 177	98.62 150	98.15 140

GBi-Net		97.69 144	97.75 124	97.62 177	98.71 190	99.21 64	98.62 138	99.33 104	94.09 183	95.60 210	98.17 123	95.97 175	94.39 195	99.05 55	99.03 49	99.08 105	98.70 90

test1		97.69 144	97.75 124	97.62 177	98.71 190	99.21 64	98.62 138	99.33 104	94.09 183	95.60 210	98.17 123	95.97 175	94.39 195	99.05 55	99.03 49	99.08 105	98.70 90

CANet_DTU		97.65 147	97.50 137	97.82 175	99.19 162	98.08 168	98.41 154	98.67 170	94.40 175	99.16 64	98.32 114	98.69 126	93.96 202	97.87 135	97.61 141	97.51 187	97.56 159

IterMVS-SCA-FT		97.63 148	96.86 158	98.52 141	99.48 120	98.71 125	98.84 119	98.91 155	96.44 124	99.16 64	99.56 41	95.54 180	97.95 133	95.68 192	95.07 196	96.76 192	97.03 176

TSAR-MVS + COLMAP		97.62 149	97.31 140	97.98 166	98.47 204	97.39 189	98.29 166	98.25 188	96.68 106	97.54 183	98.87 94	98.04 150	97.08 156	96.78 173	96.26 176	98.26 170	97.12 172

MS-PatchMatch		97.60 150	97.22 146	98.04 165	98.67 194	97.18 194	97.91 182	98.28 187	95.82 146	98.34 145	97.66 139	98.38 140	97.77 138	97.10 170	97.25 158	97.27 189	97.18 171

PCF-MVS	95.58 16	97.60 150	96.67 159	98.69 125	99.44 127	98.23 160	98.37 159	98.81 161	93.01 199	98.22 154	97.97 132	99.59 38	98.20 120	95.72 191	95.08 194	99.08 105	97.09 175
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
HQP-MVS		97.58 152	96.65 162	98.66 129	99.30 146	97.99 172	97.88 185	98.65 171	94.58 166	98.66 119	94.65 203	99.15 90	98.59 95	96.10 184	95.59 185	98.90 129	98.50 108

DI_MVS_plusplus_trai		97.57 153	96.55 164	98.77 117	99.55 101	98.76 121	99.22 79	99.00 150	97.08 88	97.95 170	97.78 136	91.35 195	98.02 129	96.20 182	96.81 167	98.87 131	97.87 149

AdaColmap		97.57 153	96.57 163	98.74 119	99.25 155	98.01 170	98.36 161	98.98 152	94.44 172	98.47 138	92.44 210	97.91 153	98.62 92	98.19 119	97.74 131	98.73 141	97.28 167

baseline		97.50 155	97.51 135	97.50 181	99.18 163	97.38 190	98.00 178	98.00 195	96.52 120	97.49 185	99.28 67	99.43 54	95.31 183	95.27 197	96.22 177	96.99 190	98.47 109

IterMVS		97.40 156	96.67 159	98.25 154	99.45 124	98.66 130	98.87 117	98.73 166	96.40 126	98.94 95	99.56 41	95.26 182	97.58 144	95.38 195	94.70 198	95.90 201	96.72 179
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
dmvs_re		97.38 157	96.15 172	98.82 113	99.39 135	98.34 152	98.65 134	98.88 157	90.80 218	98.86 105	92.35 211	95.13 183	98.09 125	98.84 77	98.88 60	99.06 112	98.71 89

CVMVSNet		97.38 157	97.39 138	97.37 184	98.58 198	97.72 183	98.70 129	97.42 205	97.21 82	95.95 209	99.46 52	93.31 190	97.38 150	97.60 147	97.78 125	96.18 197	98.66 93

new-patchmatchnet		97.26 159	96.12 173	98.58 135	99.55 101	98.63 132	99.14 90	97.04 209	98.80 14	99.19 58	99.92 4	99.19 83	98.92 74	95.51 194	87.04 212	97.66 184	93.73 202

MIMVSNet		97.24 160	97.15 151	97.36 185	99.03 178	98.52 142	98.55 144	99.73 23	94.94 163	94.94 217	97.98 131	97.37 162	93.66 203	97.60 147	97.34 154	98.23 173	96.29 182

PatchMatch-RL		97.24 160	96.45 167	98.17 158	98.70 193	97.57 188	97.31 205	98.48 181	94.42 174	98.39 142	95.74 191	96.35 174	97.88 136	97.75 142	97.48 150	98.24 172	95.87 187

thisisatest0530		97.20 162	95.95 177	98.66 129	99.46 122	98.84 114	98.29 166	99.20 127	94.51 167	98.25 152	97.42 148	85.03 211	97.68 141	98.43 97	98.56 91	99.08 105	98.89 72

tttt0517		97.18 163	95.92 178	98.65 132	99.49 118	98.92 106	98.29 166	99.20 127	94.37 177	98.17 155	97.37 151	84.72 214	97.68 141	98.55 91	98.56 91	99.10 99	98.95 65

MDTV_nov1_ep13_2view		97.12 164	96.19 171	98.22 157	99.13 170	98.05 169	99.24 77	99.47 76	97.61 62	99.15 67	99.59 36	99.01 107	98.40 109	94.87 203	90.14 206	93.91 207	94.04 201

MAR-MVS		97.12 164	96.28 170	98.11 162	98.94 180	97.22 192	97.65 195	99.38 93	90.93 217	98.15 158	95.17 196	97.13 165	96.48 170	97.71 143	97.40 151	98.06 177	98.40 117
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
Fast-Effi-MVS+-dtu		96.99 166	96.46 166	97.61 179	98.98 179	97.89 175	97.54 199	99.76 19	93.43 193	96.55 206	94.93 200	98.06 148	94.32 198	96.93 171	96.50 173	98.53 159	97.47 160

FPMVS		96.97 167	97.20 147	96.70 201	97.75 215	96.11 206	97.72 191	95.47 213	97.13 86	98.02 165	97.57 142	96.67 171	92.97 207	99.00 64	98.34 100	98.28 169	95.58 189

TAMVS		96.95 168	96.94 157	96.97 196	99.07 176	97.67 187	97.98 180	97.12 208	95.04 158	95.41 213	99.27 68	95.57 179	94.09 200	97.32 164	97.11 161	98.16 175	96.59 180

FMVSNet3		96.85 169	96.67 159	97.06 190	97.56 218	99.01 91	97.99 179	99.33 104	94.09 183	95.60 210	98.17 123	95.97 175	93.26 206	94.76 205	96.22 177	98.59 154	98.46 111

GA-MVS		96.84 170	95.86 180	97.98 166	99.16 166	98.29 153	97.91 182	98.64 172	95.14 156	97.71 179	98.04 130	88.90 199	96.50 169	96.41 181	96.61 172	97.97 181	97.60 157

CHOSEN 280x420		96.80 171	96.30 169	97.39 182	99.09 172	96.52 198	98.76 128	99.29 112	93.88 188	97.65 180	98.34 112	93.66 188	96.29 176	98.28 115	97.73 133	93.27 210	95.70 188

gg-mvs-nofinetune		96.77 172	96.52 165	97.06 190	99.66 76	97.82 179	97.54 199	99.86 8	98.69 15	98.61 121	99.94 2	89.62 197	88.37 220	97.55 150	96.67 169	98.30 168	95.35 190

DPM-MVS		96.73 173	95.70 183	97.95 169	98.93 181	97.26 191	97.39 204	98.44 183	95.47 152	97.62 181	90.71 214	98.47 138	97.03 159	95.02 202	95.27 190	98.26 170	97.67 154

baseline1		96.72 174	95.40 185	98.26 152	99.53 108	98.81 115	98.32 163	98.80 162	94.96 161	96.78 205	96.50 177	84.87 213	96.68 165	97.42 158	97.91 117	99.46 55	97.33 165

N_pmnet		96.68 175	95.70 183	97.84 173	99.42 132	98.00 171	99.35 61	98.21 189	98.40 22	98.13 159	99.42 57	99.30 69	97.44 149	94.00 209	88.79 207	94.47 206	91.96 208

pmnet_mix02		96.61 176	95.32 186	98.11 162	99.41 134	97.68 185	99.05 96	97.59 203	98.16 28	99.05 79	99.48 50	99.11 96	98.32 113	92.36 213	87.67 209	95.26 203	92.80 206

new_pmnet		96.59 177	96.40 168	96.81 198	98.24 211	95.46 215	97.71 193	94.75 216	96.92 92	96.80 204	99.23 72	97.81 155	96.69 163	96.58 177	95.16 192	96.69 193	93.64 203

PMMVS		96.47 178	95.81 181	97.23 186	97.38 220	95.96 210	97.31 205	96.91 210	93.21 196	97.93 172	97.14 157	97.64 158	95.70 179	95.24 198	96.18 180	98.17 174	95.33 191

EPNet		96.44 179	96.08 174	96.86 197	99.32 144	97.15 195	97.69 194	99.32 108	93.67 190	98.11 161	95.64 192	93.44 189	89.07 218	96.86 172	96.83 166	97.67 183	98.97 59
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
thres600view7		96.35 180	94.27 188	98.79 116	99.66 76	99.18 69	98.94 106	99.38 93	94.37 177	97.21 199	87.19 216	84.10 215	98.10 123	98.16 121	99.47 20	99.42 62	97.43 161

EPNet_dtu		96.31 181	95.96 176	96.72 200	99.18 163	95.39 216	97.03 213	99.13 140	93.02 198	99.35 38	97.23 154	97.07 166	90.70 217	95.74 190	95.08 194	94.94 205	98.16 139
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
pmmvs3		96.30 182	95.87 179	96.80 199	97.66 217	96.48 199	97.93 181	93.80 217	93.40 194	98.54 128	98.27 117	97.50 159	97.37 152	97.49 153	93.11 202	95.52 202	94.85 195

PMMVS2		96.29 183	97.05 153	95.40 211	98.32 210	96.16 203	98.18 174	97.46 204	97.20 84	84.51 226	99.60 34	98.68 128	96.37 171	98.59 89	97.38 152	97.58 186	91.76 209

thres200		96.23 184	94.13 189	98.69 125	99.44 127	99.18 69	98.58 142	99.38 93	93.52 192	97.35 193	86.33 221	85.83 209	97.93 134	98.16 121	98.78 72	99.42 62	97.10 173

thres400		96.22 185	94.08 191	98.72 121	99.58 94	99.05 82	98.83 120	99.22 122	94.01 186	97.40 190	86.34 220	84.91 212	97.93 134	97.85 137	99.08 45	99.37 68	97.28 167

tfpn200view9		96.17 186	94.08 191	98.60 134	99.37 137	99.18 69	98.68 131	99.39 88	92.02 205	97.30 195	86.53 218	86.34 206	97.45 148	98.15 123	99.08 45	99.43 61	97.28 167

CMPMVS	74.71 19	96.17 186	96.06 175	96.30 205	97.41 219	94.52 219	94.83 221	95.46 214	91.57 210	97.26 198	94.45 204	98.33 142	94.98 186	98.28 115	97.59 143	97.86 182	97.68 153
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
test2506		96.12 188	93.35 201	99.35 53	99.83 23	99.58 23	99.37 54	99.67 34	98.02 34	98.44 140	97.51 145	60.03 229	99.10 61	99.77 5	99.70 6	99.72 26	98.86 74

IB-MVS	95.85 14	95.87 189	94.88 187	97.02 193	99.09 172	98.25 158	97.16 207	97.38 206	91.97 208	97.77 176	83.61 223	97.29 163	92.03 215	97.16 168	97.66 137	98.66 146	98.20 135
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
test0.0.03 1		95.81 190	95.77 182	95.85 210	99.20 159	98.15 165	97.49 203	98.50 179	92.24 201	92.74 223	96.82 168	92.70 191	88.60 219	97.31 166	97.01 164	98.57 156	96.19 185

thres100view900		95.74 191	93.66 200	98.17 158	99.37 137	98.59 135	98.10 176	98.33 186	92.02 205	97.30 195	86.53 218	86.34 206	96.69 163	96.77 174	98.47 94	99.24 89	96.89 177

ET-MVSNet_ETH3D		95.72 192	93.85 196	97.89 170	97.30 221	98.09 167	98.19 173	98.40 184	94.46 171	98.01 168	96.71 170	77.85 225	96.76 161	96.08 185	96.39 174	98.70 144	97.36 163

baseline2		95.58 193	94.04 193	97.38 183	98.80 187	98.16 163	97.14 208	97.80 201	91.45 212	97.49 185	95.22 195	83.63 216	94.98 186	96.42 180	96.66 170	98.06 177	96.76 178

PatchT		95.49 194	93.29 202	98.06 164	98.65 195	96.20 202	98.91 111	99.73 23	92.00 207	98.50 131	96.67 172	83.25 217	96.34 172	94.40 206	95.50 186	96.21 196	95.04 193

CR-MVSNet		95.38 195	93.01 203	98.16 160	98.63 196	95.85 212	97.64 196	99.78 16	91.27 214	98.50 131	96.84 167	82.16 218	96.34 172	94.40 206	95.50 186	98.05 179	95.04 193

MVSTER		95.38 195	93.99 195	97.01 194	98.83 185	98.95 100	96.62 214	99.14 136	92.17 203	97.44 188	97.29 152	77.88 224	91.63 216	97.45 155	96.18 180	98.41 165	97.99 143

MVS-HIRNet		94.86 197	93.83 197	96.07 206	97.07 222	94.00 220	94.31 222	99.17 131	91.23 216	98.17 155	98.69 101	97.43 160	95.66 180	94.05 208	91.92 204	92.04 217	89.46 217

test-LLR		94.79 198	93.71 198	96.06 207	99.20 159	96.16 203	96.31 216	98.50 179	89.98 219	94.08 218	97.01 161	86.43 204	92.20 213	96.76 175	95.31 188	96.05 198	94.31 198

RPMNet		94.72 199	92.01 208	97.88 172	98.56 201	95.85 212	97.78 187	99.70 29	91.27 214	98.33 146	93.69 206	81.88 219	94.91 189	92.60 211	94.34 200	98.01 180	94.46 197

gm-plane-assit		94.62 200	91.39 210	98.39 145	99.90 11	99.47 33	99.40 50	99.65 39	97.44 70	99.56 20	99.68 25	59.40 230	94.23 199	96.17 183	94.77 197	97.61 185	92.79 207

test-mter		94.62 200	94.02 194	95.32 212	97.72 216	96.75 196	96.23 218	95.67 212	89.83 222	93.23 222	96.99 163	85.94 208	92.66 211	97.32 164	96.11 182	96.44 194	95.22 192

FMVSNet5		94.57 202	92.77 204	96.67 202	97.88 213	98.72 124	97.54 199	98.70 169	88.64 223	95.11 215	86.90 217	81.77 220	93.27 205	97.92 133	98.07 111	97.50 188	97.34 164

SCA		94.53 203	91.95 209	97.55 180	98.58 198	97.86 177	98.49 148	99.68 30	95.11 157	99.07 76	95.87 188	87.24 202	96.53 168	89.77 216	87.08 211	92.96 212	90.69 212

MDTV_nov1_ep13		94.47 204	92.15 206	97.17 187	98.54 203	96.42 200	98.10 176	98.89 156	94.49 169	98.02 165	97.41 149	86.49 203	95.56 181	90.85 214	87.95 208	93.91 207	91.45 211

TESTMET0.1,1		94.44 205	93.71 198	95.30 213	97.84 214	96.16 203	96.31 216	95.32 215	89.98 219	94.08 218	97.01 161	86.43 204	92.20 213	96.76 175	95.31 188	96.05 198	94.31 198

ADS-MVSNet		94.41 206	92.13 207	97.07 189	98.86 183	96.60 197	98.38 158	98.47 182	96.13 137	98.02 165	96.98 164	87.50 201	95.87 178	89.89 215	87.58 210	92.79 214	90.27 214

tpm		93.89 207	91.21 211	97.03 192	98.36 208	96.07 207	97.53 202	99.65 39	92.24 201	98.64 120	97.23 154	74.67 228	94.64 193	92.68 210	90.73 205	93.37 209	94.82 196

PatchmatchNet		93.88 208	91.08 212	97.14 188	98.75 189	96.01 209	98.25 170	99.39 88	94.95 162	98.96 90	96.32 180	85.35 210	95.50 182	88.89 217	85.89 215	91.99 218	90.15 215
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
EPMVS		93.67 209	90.82 213	96.99 195	98.62 197	96.39 201	98.40 156	99.11 141	95.54 151	97.87 174	97.14 157	81.27 222	94.97 188	88.54 219	86.80 213	92.95 213	90.06 216

MVE	82.47 18	93.12 210	94.09 190	91.99 216	90.79 224	82.50 225	93.93 223	96.30 211	96.06 138	88.81 224	98.19 121	96.38 173	97.56 145	97.24 167	95.18 191	84.58 224	93.07 204
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
CostFormer		92.75 211	89.49 215	96.55 203	98.78 188	95.83 214	97.55 198	98.59 175	91.83 209	97.34 194	96.31 181	78.53 223	94.50 194	86.14 220	84.92 216	92.54 215	92.84 205

tpmrst		92.45 212	89.48 216	95.92 209	98.43 206	95.03 217	97.14 208	97.92 199	94.16 181	97.56 182	97.86 135	81.63 221	93.56 204	85.89 221	82.86 217	90.91 222	88.95 219

dps		92.35 213	88.78 218	96.52 204	98.21 212	95.94 211	97.78 187	98.38 185	89.88 221	96.81 203	95.07 198	75.31 227	94.70 192	88.62 218	86.21 214	93.21 211	90.41 213

E-PMN		92.28 214	90.12 214	94.79 214	98.56 201	90.90 222	95.16 220	93.68 218	95.36 154	95.10 216	96.56 174	89.05 198	95.24 184	95.21 199	81.84 219	90.98 220	81.94 221

EMVS		91.84 215	89.39 217	94.70 215	98.44 205	90.84 223	95.27 219	93.53 219	95.18 155	95.26 214	95.62 193	87.59 200	94.77 191	94.87 203	80.72 220	90.95 221	80.88 222

tpm cat1		91.52 216	87.70 219	95.97 208	98.33 209	94.98 218	97.06 211	98.03 194	92.11 204	98.03 164	94.77 202	77.19 226	92.71 209	83.56 222	82.24 218	91.67 219	89.04 218

test_method		77.69 217	85.40 220	68.69 217	42.66 226	55.39 227	82.17 226	52.05 221	92.83 200	84.52 225	94.88 201	95.41 181	65.37 221	92.49 212	79.32 221	85.36 223	87.50 220

GG-mvs-BLEND		65.66 218	92.62 205	34.20 219	1.45 229	93.75 221	85.40 225	1.64 225	91.37 213	17.21 228	87.25 215	94.78 185	3.25 225	95.64 193	93.80 201	96.27 195	91.74 210

testmvs		9.73 219	13.38 221	5.48 221	3.62 227	4.12 228	6.40 229	3.19 224	14.92 224	7.68 230	22.10 224	13.89 232	6.83 223	13.47 223	10.38 223	5.14 227	14.81 223

test123		9.37 220	12.26 222	6.00 220	3.32 228	4.06 229	6.39 230	3.41 223	13.20 225	10.48 229	16.43 225	16.22 231	6.76 224	11.37 224	10.40 222	5.62 226	14.10 224

uanet_test		0.00 221	0.00 223	0.00 222	0.00 230	0.00 230	0.00 231	0.00 226	0.00 226	0.00 231	0.00 226	0.00 233	0.00 226	0.00 225	0.00 224	0.00 228	0.00 225

sosnet-low-res		0.00 221	0.00 223	0.00 222	0.00 230	0.00 230	0.00 231	0.00 226	0.00 226	0.00 231	0.00 226	0.00 233	0.00 226	0.00 225	0.00 224	0.00 228	0.00 225

sosnet		0.00 221	0.00 223	0.00 222	0.00 230	0.00 230	0.00 231	0.00 226	0.00 226	0.00 231	0.00 226	0.00 233	0.00 226	0.00 225	0.00 224	0.00 228	0.00 225

TPM-MVS					98.38 207	97.20 193	96.44 215			97.17 200	95.17 196	98.68 128	92.69 210			98.11 176	97.67 154
Ray Leroy Khuboni and Hongjun Xu: Textureless Resilient Propagation Matching in Multiple View Stereosis (TPM-MVS). SATNAC 2025
RE-MVS-def										99.88 2

9.14												98.83 119

SR-MVS					99.62 86			99.47 76				99.40 58

Anonymous202405211			98.44 76		99.79 39	99.32 52	99.05 96	99.34 103	96.59 111		97.95 133	97.68 157	97.16 154	99.36 33	99.28 35	99.61 39	98.90 70

our_test_3					99.29 149	97.72 183	98.98 102

ambc			97.89 118		99.45 124	97.88 176	97.78 187		97.27 77	99.80 3	98.99 91	98.48 136	98.55 100	97.80 139	96.68 168	98.54 157	98.10 141

MTAPA										99.19 58		99.68 21

MTMP										99.20 56		99.54 42

Patchmatch-RL test							32.47 228

tmp_tt				65.28 218	82.24 225	71.50 226	70.81 227	23.21 222	96.14 135	81.70 227	85.98 222	92.44 192	49.84 222	95.81 188	94.36 199	83.86 225

XVS					99.77 44	99.07 78	99.46 45			98.95 92		99.37 62				99.33 74

X-MVStestdata					99.77 44	99.07 78	99.46 45			98.95 92		99.37 62				99.33 74

mPP-MVS					99.75 53							99.49 52

NP-MVS									93.07 197

Patchmtry						96.05 208	97.64 196	99.78 16		98.50 131

DeepMVS_CX						87.86 224	92.27 224	61.98 220	93.64 191	93.62 221	91.17 212	91.67 193	94.90 190	95.99 187		92.48 216	94.18 200

High-resolution multi-view benchmark