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	99.39 1	99.90 1	99.87 1	99.93 1	99.97 2	99.82 8	99.91 3	99.92 38	99.75 4	99.93 6	99.89 31	100.00 1	99.87 2	99.93 3	99.82 10	99.96 3	99.90 2
Andreas Kuhn, Heiko HirschmÃ¼ller, Daniel Scharstein, Helmut Mayer: A TV Prior for High-Quality Scalable Multi-View Stereo Reconstruction. International Journal of Computer Vision 2016
v7n		99.89 2	99.86 3	99.93 1	99.97 2	99.83 4	99.93 1	99.96 12	99.77 3	99.89 18	99.99 1	99.86 83	99.84 5	99.89 11	99.81 11	99.97 1	99.88 7

SixPastTwentyTwo		99.89 2	99.85 5	99.93 1	99.97 2	99.88 1	99.92 2	99.97 1	99.66 16	99.94 5	99.94 11	99.74 116	99.81 7	99.97 1	99.89 1	99.96 3	99.89 5

pmmvs6		99.88 4	99.87 1	99.89 9	99.97 2	99.76 22	99.89 5	99.96 12	99.82 2	99.90 16	99.92 16	99.95 27	99.68 33	99.93 3	99.88 3	99.95 7	99.86 13

anonymousdsp		99.87 5	99.86 3	99.88 13	99.95 10	99.75 28	99.90 4	99.96 12	99.69 8	99.83 53	99.96 4	99.99 3	99.74 22	99.95 2	99.83 7	99.91 25	99.88 7

FC-MVSNet-test		99.84 6	99.80 6	99.89 9	99.96 7	99.83 4	99.84 16	99.95 23	99.37 57	99.77 70	99.95 6	99.96 15	99.85 3	99.93 3	99.83 7	99.95 7	99.72 42

WB-MVS		99.82 7	99.76 9	99.89 9	99.94 23	99.82 8	99.79 30	99.93 27	99.67 11	99.97 2	99.83 49	99.78 113	99.79 12	99.72 39	99.70 22	99.95 7	99.78 29

UniMVSNet_ETH3D		99.81 8	99.79 7	99.85 19	99.98 1	99.76 22	99.73 49	99.96 12	99.68 10	99.87 31	99.59 93	99.91 64	99.58 53	99.90 10	99.85 6	99.96 3	99.81 21

TDRefinement		99.81 8	99.76 9	99.86 16	99.83 95	99.53 67	99.89 5	99.91 44	99.73 5	99.88 24	99.83 49	99.96 15	99.76 17	99.91 9	99.81 11	99.86 42	99.59 73

WR-MVS		99.79 10	99.68 14	99.91 5	99.95 10	99.83 4	99.87 9	99.96 12	99.39 55	99.93 6	99.87 36	99.29 159	99.77 15	99.83 22	99.72 20	99.97 1	99.82 18

MIMVSNet1		99.79 10	99.75 11	99.84 22	99.89 46	99.83 4	99.84 16	99.89 54	99.31 62	99.93 6	99.92 16	99.97 10	99.68 33	99.89 11	99.64 28	99.82 56	99.66 56

pm-mvs1		99.77 12	99.69 13	99.86 16	99.94 23	99.68 37	99.84 16	99.93 27	99.59 26	99.87 31	99.92 16	99.21 162	99.65 39	99.88 15	99.77 16	99.93 21	99.78 29

PEN-MVS		99.77 12	99.65 19	99.91 5	99.95 10	99.80 16	99.86 10	99.97 1	99.08 91	99.89 18	99.69 76	99.68 125	99.84 5	99.81 27	99.64 28	99.95 7	99.81 21

EU-MVSNet		99.76 14	99.74 12	99.78 43	99.82 102	99.81 13	99.88 7	99.87 60	99.31 62	99.75 78	99.91 24	99.76 115	99.78 13	99.84 21	99.74 19	99.56 148	99.81 21

Vis-MVSNet		99.76 14	99.78 8	99.75 53	99.92 31	99.77 21	99.83 19	99.85 72	99.43 49	99.85 44	99.84 45	100.00 1	99.13 126	99.83 22	99.66 25	99.90 29	99.90 2
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
CS-MVS		99.75 16	99.66 18	99.85 19	99.87 58	99.86 2	99.83 19	99.91 44	98.84 128	99.92 10	99.57 95	99.85 89	99.60 48	99.82 25	99.79 13	99.94 16	99.87 11

SPE-MVS-test		99.75 16	99.67 15	99.84 22	99.91 35	99.85 3	99.85 13	99.92 38	98.75 138	99.89 18	99.64 83	99.95 27	99.55 56	99.89 11	99.79 13	99.92 22	99.83 16

DTE-MVSNet		99.75 16	99.61 27	99.92 4	99.95 10	99.81 13	99.86 10	99.96 12	99.18 80	99.92 10	99.66 79	99.45 144	99.85 3	99.80 28	99.56 34	99.96 3	99.79 28

tfpnnormal		99.74 19	99.63 24	99.86 16	99.93 28	99.75 28	99.80 29	99.89 54	99.31 62	99.88 24	99.43 117	99.66 128	99.77 15	99.80 28	99.71 21	99.92 22	99.76 33

DeepC-MVS	99.05 5	99.74 19	99.64 20	99.84 22	99.90 40	99.39 106	99.79 30	99.81 101	99.69 8	99.90 16	99.87 36	99.98 6	99.81 7	99.62 55	99.32 63	99.83 53	99.65 59
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
thisisatest0515		99.73 21	99.67 15	99.81 32	99.93 28	99.74 30	99.68 59	99.91 44	99.59 26	99.88 24	99.73 64	99.81 100	99.55 56	99.59 56	99.53 39	99.89 32	99.70 50

PS-CasMVS		99.73 21	99.59 33	99.90 8	99.95 10	99.80 16	99.85 13	99.97 1	98.95 112	99.86 37	99.73 64	99.36 151	99.81 7	99.83 22	99.67 24	99.95 7	99.83 16

WR-MVS_H		99.73 21	99.61 27	99.88 13	99.95 10	99.82 8	99.83 19	99.96 12	99.01 103	99.84 48	99.71 73	99.41 150	99.74 22	99.77 33	99.70 22	99.95 7	99.82 18

TransMVSNet (Re)		99.72 24	99.59 33	99.88 13	99.95 10	99.76 22	99.88 7	99.94 24	99.58 28	99.92 10	99.90 28	98.55 179	99.65 39	99.89 11	99.76 17	99.95 7	99.70 50

ACMH	99.11 4	99.72 24	99.63 24	99.84 22	99.87 58	99.59 52	99.83 19	99.88 59	99.46 46	99.87 31	99.66 79	99.95 27	99.76 17	99.73 38	99.47 48	99.84 48	99.52 106
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
FC-MVSNet-train		99.70 26	99.67 15	99.74 59	99.94 23	99.71 33	99.82 25	99.91 44	99.14 88	99.53 143	99.70 74	99.88 75	99.33 94	99.88 15	99.61 33	99.94 16	99.77 31

EC-MVSNet		99.70 26	99.57 36	99.85 19	99.95 10	99.81 13	99.85 13	99.93 27	98.39 176	99.76 73	99.48 113	99.94 38	99.70 31	99.85 19	99.66 25	99.91 25	99.87 11

COLMAP_ROB	99.18 2	99.70 26	99.60 31	99.81 32	99.84 88	99.37 113	99.76 37	99.84 81	99.54 36	99.82 56	99.64 83	99.95 27	99.75 19	99.79 30	99.56 34	99.83 53	99.37 141
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
ACMH+	98.94 6	99.69 29	99.59 33	99.81 32	99.88 53	99.41 103	99.75 41	99.86 65	99.43 49	99.80 60	99.54 99	99.97 10	99.73 25	99.82 25	99.52 41	99.85 45	99.43 127

test20.03		99.68 30	99.60 31	99.76 49	99.91 35	99.70 36	99.68 59	99.87 60	99.05 99	99.88 24	99.92 16	99.88 75	99.50 72	99.77 33	99.42 56	99.75 79	99.49 109

CP-MVSNet		99.68 30	99.51 46	99.89 9	99.95 10	99.76 22	99.83 19	99.96 12	98.83 132	99.84 48	99.65 82	99.09 165	99.80 10	99.78 31	99.62 32	99.95 7	99.82 18

casdiffmvs_mvg		99.67 32	99.61 27	99.74 59	99.94 23	99.60 48	99.62 73	99.77 124	99.54 36	99.67 115	99.82 52	99.80 106	99.52 66	99.40 81	99.51 42	99.91 25	99.59 73
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
viewdifsd2359ckpt11		99.66 33	99.64 20	99.68 71	99.90 40	99.67 39	99.56 86	99.72 143	99.67 11	99.69 104	99.87 36	99.93 46	99.53 60	99.51 62	99.23 67	99.69 103	99.60 69

viewmsd2359difaftdt		99.66 33	99.64 20	99.68 71	99.90 40	99.67 39	99.56 86	99.72 143	99.67 11	99.69 104	99.87 36	99.93 46	99.53 60	99.51 62	99.23 67	99.69 103	99.60 69

PVSNet_Blended_VisFu		99.66 33	99.64 20	99.67 74	99.91 35	99.71 33	99.61 74	99.79 112	99.41 51	99.91 14	99.85 43	99.61 131	99.00 137	99.67 46	99.42 56	99.81 59	99.81 21

v10		99.65 36	99.51 46	99.81 32	99.83 95	99.61 47	99.75 41	99.94 24	99.56 32	99.76 73	99.94 11	99.60 133	99.73 25	99.11 141	99.01 110	99.85 45	99.74 37

CHOSEN 1792x2688		99.65 36	99.55 40	99.77 48	99.93 28	99.60 48	99.79 30	99.92 38	99.73 5	99.74 83	99.93 14	99.98 6	99.80 10	98.83 183	99.01 110	99.45 166	99.76 33

UA-Net		99.64 38	99.62 26	99.66 76	99.97 2	99.82 8	99.14 169	99.96 12	98.95 112	99.52 149	99.38 127	99.86 83	99.55 56	99.72 39	99.66 25	99.80 64	99.94 1

viewmacassd2359aftdt		99.63 39	99.56 39	99.71 64	99.89 46	99.56 60	99.55 90	99.77 124	99.65 17	99.72 93	99.84 45	99.99 3	99.53 60	99.25 112	99.09 96	99.81 59	99.57 88

GeoE		99.63 39	99.51 46	99.78 43	99.91 35	99.57 56	99.78 33	99.97 1	99.23 71	99.72 93	99.72 69	99.80 106	99.50 72	99.45 77	99.10 94	99.79 67	99.71 48

Baseline_NR-MVSNet		99.62 41	99.48 51	99.78 43	99.85 82	99.76 22	99.59 79	99.82 93	98.84 128	99.88 24	99.91 24	99.04 166	99.61 46	99.46 70	99.78 15	99.94 16	99.60 69

pmmvs-eth3d		99.61 42	99.48 51	99.75 53	99.87 58	99.30 128	99.75 41	99.89 54	99.23 71	99.85 44	99.88 35	99.97 10	99.49 77	99.46 70	99.01 110	99.68 106	99.52 106

v1144		99.61 42	99.43 62	99.82 27	99.88 53	99.41 103	99.76 37	99.86 65	99.64 20	99.84 48	99.95 6	99.49 142	99.74 22	99.00 153	98.93 122	99.84 48	99.58 82

v8		99.61 42	99.45 59	99.79 42	99.80 108	99.59 52	99.73 49	99.93 27	99.48 44	99.77 70	99.90 28	99.48 143	99.67 36	99.11 141	98.89 127	99.84 48	99.73 39

casdiffmvs		99.61 42	99.55 40	99.68 71	99.89 46	99.53 67	99.64 67	99.68 157	99.51 40	99.62 125	99.90 28	99.96 15	99.37 88	99.28 106	99.25 66	99.88 34	99.44 124
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
CSCG		99.61 42	99.52 45	99.71 64	99.89 46	99.62 45	99.52 99	99.76 134	99.61 24	99.69 104	99.73 64	99.96 15	99.57 54	99.27 109	98.62 158	99.81 59	99.85 15

v1192		99.60 47	99.41 66	99.82 27	99.89 46	99.43 95	99.81 27	99.84 81	99.63 22	99.85 44	99.95 6	99.35 154	99.72 27	99.01 150	98.90 126	99.82 56	99.58 82

APDe-MVS		99.60 47	99.48 51	99.73 62	99.85 82	99.51 79	99.75 41	99.85 72	99.17 81	99.81 59	99.56 97	99.94 38	99.44 84	99.42 79	99.22 69	99.67 108	99.54 98
Zhaojie Zeng, Yuesong Wang, Tao Guan: Matching Ambiguity-Resilient Multi-View Stereo via Adaptive Patch Deformation. Pattern Recognition
FE-MVSNET		99.59 49	99.41 66	99.80 37	99.80 108	99.53 67	99.83 19	99.87 60	99.06 95	99.88 24	99.47 114	99.94 38	99.71 30	99.58 58	99.06 100	99.73 90	99.26 155

v1921920		99.59 49	99.40 70	99.82 27	99.88 53	99.45 89	99.81 27	99.83 86	99.65 17	99.86 37	99.95 6	99.29 159	99.75 19	98.98 156	98.86 131	99.78 69	99.59 73

TranMVSNet+NR-MVSNet		99.59 49	99.42 65	99.80 37	99.87 58	99.55 61	99.64 67	99.86 65	99.05 99	99.88 24	99.72 69	99.33 157	99.64 43	99.47 69	99.14 80	99.91 25	99.67 55

EG-PatchMatch MVS		99.59 49	99.49 50	99.70 68	99.82 102	99.26 135	99.39 129	99.83 86	98.99 106	99.93 6	99.54 99	99.92 56	99.51 68	99.78 31	99.50 43	99.73 90	99.41 131

pmmvs5		99.58 53	99.47 54	99.70 68	99.84 88	99.50 80	99.58 83	99.80 109	98.98 109	99.73 90	99.92 16	99.81 100	99.49 77	99.28 106	99.05 104	99.77 73	99.73 39

v144192		99.58 53	99.39 71	99.80 37	99.87 58	99.44 91	99.77 34	99.84 81	99.64 20	99.86 37	99.93 14	99.35 154	99.72 27	98.92 162	98.82 135	99.74 85	99.66 56

v148		99.58 53	99.43 62	99.76 49	99.87 58	99.40 105	99.76 37	99.85 72	99.48 44	99.83 53	99.82 52	99.83 95	99.51 68	99.20 124	98.82 135	99.75 79	99.45 121

v1240		99.58 53	99.38 75	99.82 27	99.89 46	99.49 82	99.82 25	99.83 86	99.63 22	99.86 37	99.96 4	98.92 172	99.75 19	99.15 135	98.96 119	99.76 75	99.56 90

V42		99.57 57	99.41 66	99.75 53	99.84 88	99.37 113	99.73 49	99.83 86	99.41 51	99.75 78	99.89 31	99.42 148	99.60 48	99.15 135	98.96 119	99.76 75	99.65 59

TSAR-MVS + MP.		99.56 58	99.54 43	99.58 93	99.69 154	99.14 157	99.73 49	99.45 197	99.50 42	99.35 180	99.60 91	99.93 46	99.50 72	99.56 59	99.37 60	99.77 73	99.64 62
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
v2v482		99.56 58	99.35 77	99.81 32	99.87 58	99.35 118	99.75 41	99.85 72	99.56 32	99.87 31	99.95 6	99.44 146	99.66 37	98.91 165	98.76 142	99.86 42	99.45 121

Gipuma		99.55 60	99.23 96	99.91 5	99.87 58	99.52 75	99.86 10	99.93 27	99.87 1	99.96 3	96.72 221	99.55 138	99.97 1	99.77 33	99.46 50	99.87 40	99.74 37
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
viewmanbaseed2359cas		99.53 61	99.46 57	99.61 89	99.85 82	99.49 82	99.37 132	99.69 150	99.54 36	99.68 113	99.73 64	99.96 15	99.32 97	99.14 138	98.86 131	99.76 75	99.52 106

DVP-MVS		99.53 61	99.51 46	99.55 105	99.82 102	99.58 54	99.54 95	99.78 117	99.28 68	99.21 190	99.70 74	99.97 10	99.32 97	99.32 94	99.14 80	99.64 121	99.58 82
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
diffmvs_AUTHOR		99.52 63	99.47 54	99.57 98	99.90 40	99.47 86	99.45 118	99.70 147	99.70 7	99.57 138	99.92 16	99.95 27	99.20 111	98.88 170	98.92 123	99.63 124	99.48 112

NR-MVSNet		99.52 63	99.29 85	99.80 37	99.96 7	99.38 109	99.55 90	99.81 101	98.86 125	99.87 31	99.51 109	98.81 174	99.72 27	99.86 18	99.04 106	99.89 32	99.54 98

viewcassd2359sk11		99.51 65	99.45 59	99.57 98	99.84 88	99.50 80	99.37 132	99.67 162	99.58 28	99.72 93	99.79 58	99.92 56	99.08 129	99.07 146	98.81 138	99.73 90	99.48 112

ACMMPR		99.51 65	99.32 80	99.72 63	99.87 58	99.33 121	99.61 74	99.85 72	99.19 78	99.73 90	98.73 175	99.95 27	99.61 46	99.35 88	99.14 80	99.66 111	99.58 82

UniMVSNet (Re)		99.50 67	99.29 85	99.75 53	99.86 74	99.47 86	99.51 102	99.82 93	98.90 120	99.89 18	99.64 83	99.00 167	99.55 56	99.32 94	99.08 98	99.90 29	99.59 73

FMVSNet1		99.50 67	99.57 36	99.42 128	99.67 163	99.65 42	99.60 78	99.91 44	99.40 53	99.39 173	99.83 49	99.27 161	98.14 178	99.68 43	99.50 43	99.81 59	99.68 52

HyFIR lowres test		99.50 67	99.26 90	99.80 37	99.95 10	99.62 45	99.76 37	99.97 1	99.67 11	99.56 139	99.94 11	98.40 182	99.78 13	98.84 182	98.59 162	99.76 75	99.72 42

PM-MVS		99.49 70	99.43 62	99.57 98	99.76 131	99.34 120	99.53 96	99.77 124	98.93 116	99.75 78	99.46 115	99.83 95	99.11 128	99.72 39	99.29 65	99.49 161	99.46 120

Anonymous20231206		99.48 71	99.31 82	99.69 70	99.79 113	99.57 56	99.63 71	99.79 112	98.88 122	99.91 14	99.72 69	99.93 46	99.59 50	99.24 113	98.63 157	99.43 170	99.18 159

DU-MVS		99.48 71	99.26 90	99.75 53	99.85 82	99.38 109	99.50 106	99.81 101	98.86 125	99.89 18	99.51 109	98.98 168	99.59 50	99.46 70	98.97 117	99.87 40	99.63 63

RPSCF		99.48 71	99.45 59	99.52 113	99.73 147	99.33 121	99.13 170	99.77 124	99.33 60	99.47 160	99.39 126	99.92 56	99.36 89	99.63 52	99.13 88	99.63 124	99.41 131

ACMMP_NAP		99.47 74	99.33 78	99.63 84	99.85 82	99.28 133	99.56 86	99.83 86	98.75 138	99.48 157	99.03 162	99.95 27	99.47 83	99.48 66	99.19 72	99.57 144	99.59 73

Anonymous20231211		99.47 74	99.39 71	99.57 98	99.89 46	99.60 48	99.50 106	99.69 150	98.91 119	99.62 125	99.17 148	99.35 154	98.86 151	99.63 52	99.46 50	99.84 48	99.62 66

SteuartSystems-ACMMP		99.47 74	99.22 99	99.76 49	99.88 53	99.36 115	99.65 66	99.84 81	98.47 163	99.80 60	98.68 178	99.96 15	99.68 33	99.37 85	99.06 100	99.72 96	99.66 56
Skip Steuart: Steuart Systems R&D Blog.
ACMM	98.37 12	99.47 74	99.23 96	99.74 59	99.86 74	99.19 151	99.68 59	99.86 65	99.16 85	99.71 101	98.52 188	99.95 27	99.62 45	99.35 88	99.02 108	99.74 85	99.42 130
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
DVP-MVS++		99.46 78	99.57 36	99.33 149	99.75 135	99.57 56	99.44 120	99.81 101	99.38 56	98.56 222	99.81 56	99.99 3	98.79 156	99.33 92	99.13 88	99.62 131	99.81 21

HFP-MVS		99.46 78	99.30 83	99.65 78	99.82 102	99.25 139	99.50 106	99.82 93	99.23 71	99.58 135	98.86 166	99.94 38	99.56 55	99.14 138	99.12 92	99.63 124	99.56 90

LGP-MVS_train		99.46 78	99.18 109	99.78 43	99.87 58	99.25 139	99.71 56	99.87 60	98.02 195	99.79 64	98.90 165	99.96 15	99.66 37	99.49 65	99.17 76	99.79 67	99.49 109

viewdifsd2359ckpt13		99.45 81	99.39 71	99.53 108	99.83 95	99.44 91	99.17 164	99.66 166	99.51 40	99.66 118	99.75 61	99.92 56	99.14 122	99.01 150	98.62 158	99.72 96	99.47 117

SED-MVS		99.45 81	99.46 57	99.42 128	99.77 126	99.57 56	99.42 123	99.80 109	99.06 95	99.38 174	99.66 79	99.96 15	98.65 164	99.31 96	99.14 80	99.53 153	99.55 95

ETV-MVS		99.45 81	99.32 80	99.60 90	99.79 113	99.60 48	99.40 128	99.78 117	97.88 201	99.83 53	99.33 130	99.70 123	98.97 140	99.74 36	99.43 54	99.84 48	99.58 82

ACMP	98.32 13	99.44 84	99.18 109	99.75 53	99.83 95	99.18 152	99.64 67	99.83 86	98.81 134	99.79 64	98.42 197	99.96 15	99.64 43	99.46 70	98.98 116	99.74 85	99.44 124
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
DCV-MVSNet		99.43 85	99.23 96	99.67 74	99.92 31	99.76 22	99.64 67	99.93 27	99.06 95	99.68 113	97.77 208	98.97 169	98.97 140	99.72 39	99.54 38	99.88 34	99.81 21

SMA-MVS		99.43 85	99.41 66	99.45 124	99.82 102	99.31 126	99.02 184	99.59 176	99.06 95	99.34 183	99.53 105	99.96 15	99.38 87	99.29 101	99.13 88	99.53 153	99.59 73
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
testgi		99.43 85	99.47 54	99.38 137	99.90 40	99.67 39	99.30 148	99.73 142	98.64 151	99.53 143	99.52 107	99.90 67	98.08 181	99.65 50	99.40 59	99.75 79	99.55 95

DELS-MVS		99.42 88	99.53 44	99.29 153	99.52 191	99.43 95	99.42 123	99.28 212	99.16 85	99.72 93	99.82 52	99.97 10	98.17 175	99.56 59	99.16 77	99.65 113	99.59 73
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
3Dnovator	99.16 3	99.42 88	99.22 99	99.65 78	99.78 118	99.13 161	99.50 106	99.85 72	99.40 53	99.80 60	98.59 184	99.79 110	99.30 102	99.20 124	99.06 100	99.71 100	99.35 144

viewmambaseed2359dif		99.41 90	99.27 89	99.58 93	99.83 95	99.42 99	99.56 86	99.68 157	99.27 69	99.58 135	99.80 57	99.85 89	99.14 122	98.70 191	98.41 172	99.67 108	99.47 117

DPE-MVS		99.41 90	99.36 76	99.47 120	99.66 164	99.48 84	99.46 117	99.75 139	98.65 147	99.41 170	99.67 77	99.95 27	98.82 152	99.21 121	99.14 80	99.72 96	99.40 136
Kehua Chen, Zhenlong Yuan, Tianlu Mao, Zhaoqi Wang: Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization. AAAI2025
UniMVSNet_NR-MVSNet		99.41 90	99.12 121	99.76 49	99.86 74	99.48 84	99.50 106	99.81 101	98.84 128	99.89 18	99.45 116	98.32 185	99.59 50	99.22 117	98.89 127	99.90 29	99.63 63

CP-MVS		99.41 90	99.20 105	99.65 78	99.80 108	99.23 146	99.44 120	99.75 139	98.60 156	99.74 83	98.66 179	99.93 46	99.48 80	99.33 92	99.16 77	99.73 90	99.48 112

QAPM		99.41 90	99.21 104	99.64 83	99.78 118	99.16 154	99.51 102	99.85 72	99.20 75	99.72 93	99.43 117	99.81 100	99.25 107	98.87 172	98.71 149	99.71 100	99.30 149

UGNet		99.40 95	99.61 27	99.16 173	99.88 53	99.64 43	99.61 74	99.77 124	99.31 62	99.63 124	99.33 130	99.93 46	96.46 215	99.63 52	99.53 39	99.63 124	99.89 5
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
Vis-MVSNet (Re-imp)		99.40 95	99.28 87	99.55 105	99.92 31	99.68 37	99.31 143	99.87 60	98.69 144	99.16 192	99.08 157	98.64 178	99.20 111	99.65 50	99.46 50	99.83 53	99.72 42

OPM-MVS		99.39 97	99.22 99	99.59 91	99.76 131	98.82 185	99.51 102	99.79 112	99.17 81	99.53 143	99.31 135	99.95 27	99.35 90	99.22 117	98.79 141	99.60 136	99.27 152
Ray L. Khuboni, Hongjun Xu: Octagram Propagation Matching for Multi-Scale View Stereopsis (OPM-MVS).
Fast-Effi-MVS+		99.39 97	99.18 109	99.63 84	99.86 74	99.28 133	99.45 118	99.91 44	98.47 163	99.61 128	99.50 111	99.57 135	99.17 113	99.24 113	98.66 154	99.78 69	99.59 73

LS3D		99.39 97	99.28 87	99.52 113	99.77 126	99.39 106	99.55 90	99.82 93	98.93 116	99.64 122	98.52 188	99.67 127	98.58 168	99.74 36	99.63 30	99.75 79	99.06 175

diffmvs		99.38 100	99.33 78	99.45 124	99.87 58	99.39 106	99.28 152	99.58 180	99.55 34	99.50 153	99.85 43	99.85 89	98.94 146	98.58 196	98.68 152	99.51 158	99.39 138
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
CANet		99.36 101	99.39 71	99.34 148	99.80 108	99.35 118	99.41 127	99.47 195	99.20 75	99.74 83	99.54 99	99.68 125	98.05 183	99.23 115	98.97 117	99.57 144	99.73 39

ACMMP		99.36 101	99.06 129	99.71 64	99.86 74	99.36 115	99.63 71	99.85 72	98.33 178	99.72 93	97.73 210	99.94 38	99.53 60	99.37 85	99.13 88	99.65 113	99.56 90
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
SD-MVS		99.35 103	99.26 90	99.46 122	99.66 164	99.15 156	98.92 195	99.67 162	99.55 34	99.35 180	98.83 168	99.91 64	99.35 90	99.19 127	98.53 164	99.78 69	99.68 52
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
MP-MVS		99.35 103	99.09 127	99.65 78	99.84 88	99.22 147	99.59 79	99.78 117	98.13 187	99.67 115	98.44 193	99.93 46	99.43 86	99.31 96	99.09 96	99.60 136	99.49 109
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
pmmvs4		99.34 105	99.15 116	99.57 98	99.77 126	98.90 178	99.51 102	99.77 124	99.07 93	99.73 90	99.72 69	99.84 93	99.07 131	98.85 177	98.39 174	99.55 151	99.27 152

EPP-MVSNet		99.34 105	99.10 125	99.62 88	99.94 23	99.74 30	99.66 65	99.80 109	99.07 93	98.93 204	99.61 88	96.13 201	99.49 77	99.67 46	99.63 30	99.92 22	99.86 13

TSAR-MVS + GP.		99.33 107	99.17 113	99.51 115	99.71 152	99.00 173	98.84 203	99.71 146	98.23 184	99.74 83	99.53 105	99.90 67	99.35 90	99.38 84	98.85 133	99.72 96	99.31 147

PHI-MVS		99.33 107	99.19 107	99.49 118	99.69 154	99.25 139	99.27 153	99.59 176	98.44 167	99.78 68	99.15 149	99.92 56	98.95 145	99.39 82	99.04 106	99.64 121	99.18 159

MSP-MVS		99.32 109	99.26 90	99.38 137	99.76 131	99.54 64	99.42 123	99.72 143	98.92 118	98.84 212	98.96 164	99.96 15	98.91 147	98.72 190	99.14 80	99.63 124	99.58 82
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		99.32 109	98.99 138	99.71 64	99.86 74	99.31 126	99.59 79	99.86 65	97.51 210	99.75 78	98.23 200	99.94 38	99.53 60	99.29 101	99.08 98	99.65 113	99.54 98

DeepC-MVS_fast	98.69 9	99.32 109	99.13 119	99.53 108	99.63 170	98.78 188	99.53 96	99.33 210	99.08 91	99.77 70	99.18 147	99.89 69	99.29 103	99.00 153	98.70 150	99.65 113	99.30 149
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
MSDG		99.32 109	99.09 127	99.58 93	99.75 135	98.74 192	99.36 135	99.54 184	99.14 88	99.72 93	99.24 139	99.89 69	99.51 68	99.30 98	98.76 142	99.62 131	98.54 194

TSAR-MVS + ACMM		99.31 113	99.26 90	99.37 143	99.66 164	98.97 176	99.20 161	99.56 182	99.33 60	99.19 191	99.54 99	99.91 64	99.32 97	99.12 140	98.34 177	99.29 184	99.65 59

3Dnovator+	98.92 7	99.31 113	99.03 133	99.63 84	99.77 126	98.90 178	99.52 99	99.81 101	99.37 57	99.72 93	98.03 205	99.73 119	99.32 97	98.99 155	98.81 138	99.67 108	99.36 142

X-MVS		99.30 115	98.99 138	99.66 76	99.85 82	99.30 128	99.49 113	99.82 93	98.32 179	99.69 104	97.31 219	99.93 46	99.50 72	99.37 85	99.16 77	99.60 136	99.53 101

MVS_111021_HR		99.30 115	99.14 117	99.48 119	99.58 187	99.25 139	99.27 153	99.61 171	98.74 140	99.66 118	99.02 163	99.84 93	99.33 94	99.20 124	98.76 142	99.44 167	99.18 159

TAPA-MVS	98.54 10	99.30 115	99.24 95	99.36 147	99.44 206	98.77 190	99.00 186	99.41 201	99.23 71	99.60 130	99.50 111	99.86 83	99.15 120	99.29 101	98.95 121	99.56 148	99.08 171
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
CLD-MVS		99.30 115	99.01 137	99.63 84	99.75 135	98.89 181	99.35 138	99.60 173	98.53 161	99.86 37	99.57 95	99.94 38	99.52 66	98.96 157	98.10 190	99.70 102	99.08 171
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
USDC		99.29 119	98.98 140	99.65 78	99.72 149	98.87 183	99.47 115	99.66 166	99.35 59	99.87 31	99.58 94	99.87 82	99.51 68	98.85 177	97.93 196	99.65 113	98.38 198

PMVS	94.32 17	99.27 120	99.55 40	98.94 190	99.60 179	99.43 95	99.39 129	99.54 184	98.99 106	99.69 104	99.60 91	99.81 100	95.68 220	99.88 15	99.83 7	99.73 90	99.31 147
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
FA-MVS(training)		99.26 121	99.12 121	99.44 126	99.60 179	99.26 135	99.24 158	99.97 1	98.84 128	99.76 73	99.43 117	98.74 175	98.47 171	99.39 82	99.10 94	99.57 144	99.07 174

MVS_111021_LR		99.25 122	99.13 119	99.39 133	99.50 199	99.14 157	99.23 159	99.50 192	98.67 145	99.61 128	99.12 153	99.81 100	99.16 116	99.28 106	98.67 153	99.35 180	99.21 158

ECVR-MVS		99.24 123	98.74 163	99.82 27	99.95 10	99.78 18	99.67 63	99.93 27	99.45 47	99.80 60	99.86 41	92.58 219	99.65 39	99.93 3	99.88 3	99.94 16	99.71 48

baseline		99.24 123	99.30 83	99.17 172	99.78 118	99.14 157	99.10 174	99.69 150	98.97 110	99.49 155	99.84 45	99.88 75	97.99 188	98.85 177	98.73 147	98.98 199	99.72 42

EIA-MVS		99.23 125	99.03 133	99.47 120	99.83 95	99.64 43	99.16 166	99.81 101	97.11 217	99.65 121	98.44 193	99.78 113	98.61 167	99.46 70	99.22 69	99.75 79	99.59 73

HPM-MVS++		99.23 125	98.98 140	99.53 108	99.75 135	99.02 171	99.44 120	99.77 124	98.65 147	99.52 149	98.72 176	99.92 56	99.33 94	98.77 188	98.40 173	99.40 174	99.36 142

PMMVS2		99.23 125	99.22 99	99.24 160	99.80 108	99.14 157	99.50 106	99.82 93	99.12 90	98.41 228	99.91 24	99.98 6	98.51 169	99.48 66	98.76 142	99.38 176	98.14 206

MVS_0304		99.22 128	99.22 99	99.23 161	99.87 58	99.58 54	99.70 57	99.59 176	99.58 28	98.98 200	99.40 124	97.31 198	97.53 197	99.41 80	99.43 54	99.69 103	99.81 21

test1111		99.21 129	98.67 169	99.84 22	99.96 7	99.82 8	99.72 53	99.94 24	99.54 36	99.78 68	99.89 31	91.89 222	99.69 32	99.93 3	99.89 1	99.95 7	99.75 35

CPTT-MVS		99.21 129	98.89 150	99.58 93	99.72 149	99.12 164	99.30 148	99.76 134	98.62 152	99.66 118	97.51 215	99.89 69	99.48 80	99.01 150	98.64 156	99.58 143	99.40 136

TinyColmap		99.21 129	98.89 150	99.59 91	99.61 175	98.61 200	99.47 115	99.67 162	99.02 102	99.82 56	99.15 149	99.74 116	99.35 90	99.17 133	98.33 178	99.63 124	98.22 204

Effi-MVS+		99.20 132	98.93 145	99.50 117	99.79 113	99.26 135	98.82 206	99.96 12	98.37 177	99.60 130	99.12 153	98.36 183	99.05 134	98.93 160	98.82 135	99.78 69	99.68 52

PVSNet_BlendedMVS		99.20 132	99.17 113	99.23 161	99.69 154	99.33 121	99.04 179	99.13 215	98.41 172	99.79 64	99.33 130	99.36 151	98.10 179	99.29 101	98.87 129	99.65 113	99.56 90

PVSNet_Blended		99.20 132	99.17 113	99.23 161	99.69 154	99.33 121	99.04 179	99.13 215	98.41 172	99.79 64	99.33 130	99.36 151	98.10 179	99.29 101	98.87 129	99.65 113	99.56 90

MCST-MVS		99.17 135	98.82 158	99.57 98	99.75 135	98.70 196	99.25 157	99.69 150	98.62 152	99.59 132	98.54 186	99.79 110	99.53 60	98.48 200	98.15 186	99.64 121	99.43 127

APD-MVS		99.17 135	98.92 146	99.46 122	99.78 118	99.24 144	99.34 139	99.78 117	97.79 204	99.48 157	98.25 199	99.88 75	98.77 157	99.18 130	98.92 123	99.63 124	99.18 159
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
OpenMVS	98.82 8	99.17 135	98.85 154	99.53 108	99.75 135	99.06 169	99.36 135	99.82 93	98.28 181	99.76 73	98.47 190	99.61 131	98.91 147	98.80 185	98.70 150	99.60 136	99.04 179

IterMVS-LS		99.16 138	98.82 158	99.57 98	99.87 58	99.71 33	99.58 83	99.92 38	99.24 70	99.71 101	99.73 64	95.79 202	98.91 147	98.82 184	98.66 154	99.43 170	99.77 31
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
DeepPCF-MVS	98.38 11	99.16 138	99.20 105	99.12 177	99.20 223	98.71 195	98.85 202	99.06 218	99.17 81	98.96 203	99.61 88	99.86 83	99.29 103	99.17 133	98.72 148	99.36 178	99.15 167

IterMVS-SCA-FT		99.15 140	98.96 142	99.38 137	99.87 58	99.54 64	99.53 96	99.79 112	98.94 114	99.82 56	99.92 16	97.65 192	98.82 152	98.95 159	98.26 180	98.45 209	99.47 117

CDS-MVSNet		99.15 140	99.10 125	99.21 168	99.59 184	99.22 147	99.48 114	99.47 195	98.89 121	99.41 170	99.84 45	98.11 188	97.76 191	99.26 111	99.01 110	99.57 144	99.38 139
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
IS_MVSNet		99.15 140	99.12 121	99.19 170	99.92 31	99.73 32	99.55 90	99.86 65	98.45 166	96.91 234	98.74 174	98.33 184	99.02 136	99.54 61	99.47 48	99.88 34	99.61 68

dmvs_re		99.14 143	98.76 161	99.58 93	99.75 135	99.38 109	99.30 148	99.68 157	96.94 222	99.74 83	97.70 211	99.20 163	99.29 103	99.22 117	99.35 61	99.73 90	99.55 95

MDA-MVSNet-bldmvs		99.11 144	99.11 124	99.12 177	99.91 35	99.38 109	99.77 34	98.72 222	99.31 62	99.85 44	99.43 117	98.26 186	99.48 80	99.85 19	98.47 167	96.99 220	99.08 171

OMC-MVS		99.11 144	98.95 143	99.29 153	99.37 212	98.57 202	99.19 162	99.20 214	98.87 124	99.58 135	99.13 151	99.88 75	99.00 137	99.19 127	98.46 168	99.43 170	98.57 193

MVS_Test		99.09 146	98.92 146	99.29 153	99.61 175	99.07 168	99.04 179	99.81 101	98.58 158	99.37 177	99.74 62	98.87 173	98.41 173	98.61 195	98.01 194	99.50 160	99.57 88

CNVR-MVS		99.08 147	98.83 155	99.37 143	99.61 175	98.74 192	99.15 167	99.54 184	98.59 157	99.37 177	98.15 202	99.88 75	99.08 129	98.91 165	98.46 168	99.48 162	99.06 175

IterMVS		99.08 147	98.90 149	99.29 153	99.87 58	99.53 67	99.52 99	99.77 124	98.94 114	99.75 78	99.91 24	97.52 196	98.72 161	98.86 175	98.14 187	98.09 212	99.43 127
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
FMVSNet2		99.07 149	99.19 107	98.93 192	99.02 228	99.53 67	99.31 143	99.84 81	98.86 125	98.88 207	99.64 83	98.44 181	96.92 209	99.35 88	99.00 114	99.61 133	99.53 101

CVMVSNet		99.06 150	98.88 153	99.28 157	99.52 191	99.53 67	99.42 123	99.69 150	98.74 140	98.27 230	99.89 31	95.48 207	99.44 84	99.46 70	99.33 62	99.32 183	99.75 35

CDPH-MVS		99.05 151	98.63 170	99.54 107	99.75 135	98.78 188	99.59 79	99.68 157	97.79 204	99.37 177	98.20 201	99.86 83	99.14 122	98.58 196	98.01 194	99.68 106	99.16 165

TAMVS		99.05 151	99.02 136	99.08 182	99.69 154	99.22 147	99.33 140	99.32 211	99.16 85	98.97 202	99.87 36	97.36 197	97.76 191	99.21 121	99.00 114	99.44 167	99.33 145

CANet_DTU		99.03 153	99.18 109	98.87 195	99.58 187	99.03 170	99.18 163	99.41 201	98.65 147	99.74 83	99.55 98	99.71 120	96.13 218	99.19 127	98.92 123	99.17 193	99.18 159

Effi-MVS+-dtu		99.01 154	99.05 130	98.98 186	99.60 179	99.13 161	99.03 183	99.61 171	98.52 162	99.01 197	98.53 187	99.83 95	96.95 208	99.48 66	98.59 162	99.66 111	99.25 157

sasdasda		99.00 155	98.68 167	99.37 143	99.68 160	99.42 99	98.94 193	99.89 54	99.00 104	98.99 198	98.43 195	95.69 203	98.96 143	99.18 130	99.18 73	99.74 85	99.88 7

canonicalmvs		99.00 155	98.68 167	99.37 143	99.68 160	99.42 99	98.94 193	99.89 54	99.00 104	98.99 198	98.43 195	95.69 203	98.96 143	99.18 130	99.18 73	99.74 85	99.88 7

MIMVSNet		99.00 155	99.03 133	98.97 189	99.32 218	99.32 125	99.39 129	99.91 44	98.41 172	98.76 215	99.24 139	99.17 164	97.13 202	99.30 98	98.80 140	99.29 184	99.01 180

CHOSEN 280x420		98.99 158	98.91 148	99.07 183	99.77 126	99.26 135	99.55 90	99.92 38	98.62 152	98.67 219	99.62 87	97.20 199	98.44 172	99.50 64	99.18 73	98.08 213	98.99 183

MGCFI-Net		98.98 159	98.69 166	99.33 149	99.68 160	99.42 99	98.95 191	99.90 53	99.04 101	98.88 207	98.45 192	95.64 205	98.81 154	99.15 135	99.21 71	99.75 79	99.90 2

SF-MVS		98.96 160	98.95 143	98.98 186	99.64 169	98.89 181	98.00 232	99.58 180	98.42 170	99.08 196	98.63 181	99.83 95	98.04 185	99.02 149	98.76 142	99.52 155	99.13 168

GBi-Net		98.96 160	99.05 130	98.85 196	99.02 228	99.53 67	99.31 143	99.78 117	98.13 187	98.48 224	99.43 117	97.58 193	96.92 209	99.68 43	99.50 43	99.61 133	99.53 101

test1		98.96 160	99.05 130	98.85 196	99.02 228	99.53 67	99.31 143	99.78 117	98.13 187	98.48 224	99.43 117	97.58 193	96.92 209	99.68 43	99.50 43	99.61 133	99.53 101

PCF-MVS	97.86 15	98.95 163	98.53 175	99.44 126	99.70 153	98.80 187	98.96 188	99.69 150	98.65 147	99.59 132	99.33 130	99.94 38	99.12 127	98.01 210	97.11 207	99.59 142	97.83 210
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
MS-PatchMatch		98.94 164	98.71 165	99.21 168	99.52 191	98.22 218	98.97 187	99.53 189	98.76 136	99.50 153	98.59 184	99.56 137	98.68 162	98.63 194	98.45 170	99.05 196	98.73 190

AdaColmap		98.93 165	98.53 175	99.39 133	99.52 191	98.65 199	99.11 173	99.59 176	98.08 191	99.44 163	97.46 217	99.45 144	99.24 108	98.92 162	98.44 171	99.44 167	98.73 190

MSLP-MVS++		98.92 166	98.73 164	99.14 174	99.44 206	99.00 173	98.36 222	99.35 207	98.82 133	99.38 174	96.06 223	99.79 110	99.07 131	98.88 170	99.05 104	99.27 186	99.53 101

new_pmnet		98.91 167	98.89 150	98.94 190	99.51 197	98.27 214	99.15 167	98.66 223	99.17 81	99.48 157	99.79 58	99.80 106	98.49 170	99.23 115	98.20 184	98.34 210	97.74 214

train_agg		98.89 168	98.48 180	99.38 137	99.69 154	98.76 191	99.31 143	99.60 173	97.71 206	98.98 200	97.89 206	99.89 69	99.29 103	98.32 201	97.59 203	99.42 173	99.16 165

NCCC		98.88 169	98.42 181	99.42 128	99.62 171	98.81 186	99.10 174	99.54 184	98.76 136	99.53 143	95.97 224	99.80 106	99.16 116	98.49 199	98.06 193	99.55 151	99.05 177

PLC	97.83 16	98.88 169	98.52 177	99.30 152	99.45 204	98.60 201	98.65 212	99.49 193	98.66 146	99.59 132	96.33 222	99.59 134	99.17 113	98.87 172	98.53 164	99.46 164	99.05 177
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
pmmvs3		98.85 171	98.60 171	99.13 175	99.66 164	98.72 194	99.37 132	99.06 218	98.44 167	99.76 73	99.74 62	99.55 138	99.15 120	99.04 147	96.00 215	97.80 214	98.72 192

Fast-Effi-MVS+-dtu		98.82 172	98.80 160	98.84 198	99.51 197	98.90 178	98.96 188	99.91 44	98.29 180	99.11 195	98.47 190	99.63 130	96.03 219	99.21 121	98.12 188	99.52 155	99.01 180

CNLPA		98.82 172	98.52 177	99.18 171	99.21 222	98.50 206	98.73 210	99.34 209	98.73 142	99.56 139	97.55 214	99.42 148	99.06 133	98.93 160	98.10 190	99.21 192	98.38 198

PatchMatch-RL		98.80 174	98.52 177	99.12 177	99.38 211	98.70 196	98.56 215	99.55 183	97.81 203	99.34 183	97.57 213	99.31 158	98.67 163	99.27 109	98.62 158	99.22 191	98.35 200

thisisatest0530		98.78 175	98.26 184	99.39 133	99.78 118	99.43 95	99.07 176	99.64 169	98.44 167	99.42 168	99.22 141	92.68 218	98.63 165	99.30 98	99.14 80	99.80 64	99.60 69

tttt0517		98.77 176	98.25 186	99.38 137	99.79 113	99.46 88	99.07 176	99.64 169	98.40 175	99.38 174	99.21 143	92.54 220	98.63 165	99.34 91	99.14 80	99.80 64	99.62 66

DI_MVS_pp		98.74 177	98.08 194	99.51 115	99.79 113	99.29 132	99.61 74	99.60 173	99.20 75	99.46 161	99.09 156	92.93 212	98.97 140	98.27 204	98.35 176	99.65 113	99.45 121

TSAR-MVS + COLMAP		98.74 177	98.58 173	98.93 192	99.29 219	98.23 215	99.04 179	99.24 213	98.79 135	98.80 214	99.37 128	99.71 120	98.06 182	98.02 209	97.46 205	99.16 194	98.48 196

MDTV_nov1_ep13_2view		98.73 179	98.31 183	99.22 165	99.75 135	99.24 144	99.75 41	99.93 27	99.31 62	99.84 48	99.86 41	99.81 100	99.31 101	97.40 218	94.77 217	96.73 222	97.81 211

PMMVS		98.71 180	98.55 174	98.90 194	99.28 220	98.45 208	98.53 218	99.45 197	97.67 208	99.15 194	98.76 172	99.54 140	97.79 190	98.77 188	98.23 182	99.16 194	98.46 197

HQP-MVS		98.70 181	98.19 190	99.28 157	99.61 175	98.52 204	98.71 211	99.35 207	97.97 198	99.53 143	97.38 218	99.85 89	99.14 122	97.53 214	96.85 211	99.36 178	99.26 155

N_pmnet		98.64 182	98.23 189	99.11 180	99.78 118	99.25 139	99.75 41	99.39 205	99.65 17	99.70 103	99.78 60	99.89 69	98.81 154	97.60 213	94.28 218	97.24 219	97.15 218

CMPMVS	76.62 19	98.64 182	98.60 171	98.68 203	99.33 216	97.07 231	98.11 230	98.50 224	97.69 207	99.26 186	98.35 198	99.66 128	97.62 194	99.43 78	99.02 108	99.24 189	99.01 180
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
FMVSNet3		98.63 184	98.75 162	98.49 209	98.10 234	99.44 91	99.02 184	99.78 117	98.13 187	98.48 224	99.43 117	97.58 193	96.16 217	98.85 177	98.39 174	99.40 174	99.41 131

GA-MVS		98.59 185	98.15 191	99.09 181	99.59 184	99.13 161	98.84 203	99.52 191	98.61 155	99.35 180	99.67 77	93.03 211	97.73 193	98.90 169	98.26 180	99.51 158	99.48 112

MAR-MVS		98.54 186	98.15 191	98.98 186	99.37 212	98.09 221	98.56 215	99.65 168	96.11 232	99.27 185	97.16 220	99.50 141	98.03 186	98.87 172	98.23 182	99.01 197	99.13 168
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
new-patchmatchnet		98.49 187	97.60 196	99.53 108	99.90 40	99.55 61	99.77 34	99.48 194	99.67 11	99.86 37	99.98 3	99.98 6	99.50 72	96.90 220	91.52 224	98.67 206	95.62 224

FPMVS		98.48 188	98.83 155	98.07 219	99.09 226	97.98 224	99.07 176	98.04 230	98.99 106	99.22 189	98.85 167	99.43 147	93.79 228	99.66 48	99.11 93	99.24 189	97.76 212

MVS-HIRNet		98.45 189	98.25 186	98.69 202	99.12 224	97.81 230	98.55 217	99.85 72	98.58 158	99.67 115	99.61 88	99.86 83	97.46 198	97.95 211	96.37 213	97.49 216	97.56 215

test0.0.03 1		98.41 190	98.41 182	98.40 213	99.62 171	99.16 154	98.87 200	99.41 201	97.15 215	96.60 236	99.31 135	97.00 200	96.55 214	98.91 165	98.51 166	99.37 177	98.82 187

gg-mvs-nofinetune		98.40 191	98.26 184	98.57 207	99.83 95	98.86 184	98.77 209	99.97 1	99.57 31	99.99 1	99.99 1	93.81 209	93.50 229	98.91 165	98.20 184	99.33 182	98.52 195

baseline1		98.39 192	97.59 197	99.31 151	99.78 118	99.45 89	99.13 170	99.53 189	98.06 193	98.87 209	98.63 181	90.04 226	98.76 158	98.85 177	98.84 134	99.81 59	99.28 151

pmnet_mix02		98.28 193	97.48 199	99.22 165	99.78 118	99.12 164	99.68 59	99.39 205	99.49 43	99.86 37	99.82 52	99.89 69	99.23 109	95.54 223	92.36 221	97.38 217	96.14 222

PatchT		98.11 194	97.12 205	99.26 159	99.65 168	98.34 212	99.57 85	99.97 1	97.48 211	99.43 165	99.04 161	90.84 224	98.15 176	98.04 207	97.78 197	98.82 203	98.30 201

DPM-MVS		98.10 195	97.32 203	99.01 185	99.52 191	97.92 225	98.47 220	99.45 197	98.25 182	98.91 205	93.99 228	99.69 124	98.73 160	96.29 222	96.32 214	99.00 198	98.77 188

EPNet_dtu		98.09 196	98.25 186	97.91 221	99.58 187	98.02 223	98.19 227	99.67 162	97.94 199	99.74 83	99.07 159	98.71 177	93.40 230	97.50 215	97.09 208	96.89 221	99.44 124
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
EPNet		98.06 197	98.11 193	98.00 220	99.60 179	98.99 175	98.38 221	99.68 157	98.18 186	98.85 211	97.89 206	95.60 206	92.72 231	98.30 202	98.10 190	98.76 204	99.72 42
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
CR-MVSNet		97.91 198	96.80 208	99.22 165	99.60 179	98.23 215	98.91 196	99.97 1	96.89 225	99.43 165	99.10 155	89.24 229	98.15 176	98.04 207	97.78 197	99.26 187	98.30 201

thres200		97.87 199	96.56 210	99.39 133	99.76 131	99.52 75	99.13 170	99.76 134	96.88 227	98.66 220	92.87 232	88.77 232	99.16 116	99.11 141	99.42 56	99.88 34	99.33 145

baseline2		97.87 199	97.18 204	98.67 204	99.34 215	99.17 153	98.48 219	98.82 221	97.08 218	98.83 213	98.75 173	89.47 228	97.03 207	98.67 193	98.27 179	99.52 155	98.83 186

thres600view7		97.86 201	96.53 213	99.41 131	99.84 88	99.52 75	99.36 135	99.76 134	97.32 213	98.38 229	93.24 229	87.25 234	99.23 109	99.11 141	99.75 18	99.88 34	99.48 112

tfpn200view9		97.85 202	96.54 211	99.38 137	99.74 145	99.52 75	99.17 164	99.76 134	96.10 233	98.70 217	92.99 230	89.10 230	99.00 137	99.11 141	99.56 34	99.88 34	99.41 131

thres400		97.82 203	96.47 214	99.40 132	99.81 107	99.44 91	99.29 151	99.69 150	97.15 215	98.57 221	92.82 233	87.96 233	99.16 116	98.96 157	99.55 37	99.86 42	99.41 131

IB-MVS	98.10 14	97.76 204	97.40 202	98.18 215	99.62 171	99.11 166	98.24 225	98.35 226	96.56 229	99.44 163	91.28 234	98.96 171	93.84 227	98.09 206	98.62 158	99.56 148	99.18 159
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
test-LLR		97.74 205	97.46 200	98.08 217	99.62 171	98.37 210	98.26 223	99.41 201	97.03 219	97.38 232	99.54 99	92.89 213	95.12 224	98.78 186	97.68 201	98.65 207	97.90 208

RPMNet		97.70 206	96.54 211	99.06 184	99.57 190	98.23 215	98.95 191	99.97 1	96.89 225	99.49 155	99.13 151	89.63 227	97.09 204	96.68 221	97.02 209	99.26 187	98.19 205

thres100view900		97.69 207	96.37 215	99.23 161	99.74 145	99.21 150	98.81 207	99.43 200	96.10 233	98.70 217	92.99 230	89.10 230	98.88 150	98.58 196	99.31 64	99.82 56	99.27 152

FMVSNet5		97.69 207	96.98 206	98.53 208	98.53 232	99.36 115	98.90 199	99.54 184	96.38 230	98.44 227	95.38 226	90.08 225	97.05 206	99.46 70	99.06 100	98.73 205	99.12 170

MVE	91.08 18	97.68 209	97.65 195	97.71 227	98.46 233	91.62 237	97.92 233	98.86 220	98.73 142	97.99 231	98.64 180	99.96 15	99.17 113	99.59 56	97.75 199	93.87 236	97.27 216
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
test-mter		97.65 210	97.57 198	97.75 225	98.90 231	98.56 203	98.15 228	98.45 225	96.92 224	96.84 235	99.52 107	92.53 221	95.24 223	99.04 147	98.12 188	98.90 201	98.29 203

TESTMET0.1,1		97.62 211	97.46 200	97.81 223	99.07 227	98.37 210	98.26 223	98.35 226	97.03 219	97.38 232	99.54 99	92.89 213	95.12 224	98.78 186	97.68 201	98.65 207	97.90 208

test2506		97.57 212	95.67 221	99.78 43	99.95 10	99.78 18	99.67 63	99.93 27	99.45 47	99.55 142	99.20 144	71.73 241	99.65 39	99.93 3	99.88 3	99.94 16	99.72 42

MVSTER		97.55 213	96.75 209	98.48 210	99.46 203	99.54 64	98.24 225	99.77 124	97.56 209	99.41 170	99.31 135	84.86 236	94.66 226	98.86 175	97.75 199	99.34 181	99.38 139

ET-MVSNet_ETH3D		97.44 214	96.29 216	98.78 199	97.93 235	98.95 177	98.91 196	99.09 217	98.00 196	99.24 187	98.83 168	84.62 237	98.02 187	97.43 217	97.38 206	99.48 162	98.84 185

MDTV_nov1_ep13		97.41 215	96.26 217	98.76 200	99.47 201	98.43 209	99.26 156	99.82 93	98.06 193	99.23 188	99.22 141	92.86 215	98.05 183	95.33 225	93.66 220	96.73 222	96.26 221

ADS-MVSNet		97.29 216	96.17 218	98.59 206	99.59 184	98.70 196	99.32 141	99.86 65	98.47 163	99.56 139	99.08 157	98.16 187	97.34 200	92.92 227	91.17 225	95.91 225	94.72 227

SCA		97.25 217	96.05 219	98.64 205	99.36 214	99.02 171	99.27 153	99.96 12	98.25 182	99.69 104	98.71 177	94.66 208	97.95 189	93.95 226	92.35 222	95.64 226	95.40 226

gm-plane-assit		96.82 218	94.84 226	99.13 175	99.95 10	99.78 18	99.69 58	99.92 38	99.19 78	99.84 48	99.92 16	72.93 240	96.44 216	98.21 205	97.01 210	98.92 200	96.87 220

PatchmatchNet		96.81 219	95.41 223	98.43 212	99.43 208	98.30 213	99.23 159	99.93 27	98.19 185	99.64 122	98.81 171	93.50 210	97.43 199	92.89 228	90.78 227	94.94 231	95.41 225
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
EPMVS		96.76 220	95.30 225	98.46 211	99.42 209	98.47 207	99.32 141	99.91 44	98.42 170	99.51 151	99.07 159	92.81 216	97.12 203	92.39 229	91.71 223	95.51 227	94.20 229

E-PMN		96.72 221	95.78 220	97.81 223	99.45 204	95.46 234	98.14 229	98.33 228	97.99 197	98.73 216	98.09 203	98.97 169	97.54 196	97.45 216	91.09 226	94.70 233	91.40 232

tpm		96.56 222	94.68 227	98.74 201	99.12 224	97.90 226	98.79 208	99.93 27	96.79 228	99.69 104	99.19 146	81.48 239	97.56 195	95.46 224	93.97 219	97.37 218	97.99 207

EMVS		96.47 223	95.38 224	97.74 226	99.42 209	95.37 235	98.07 231	98.27 229	97.85 202	98.90 206	97.48 216	98.73 176	97.20 201	97.21 219	90.39 228	94.59 235	90.65 233

tpmrst		96.18 224	94.47 228	98.18 215	99.52 191	97.89 227	98.96 188	99.79 112	98.07 192	99.16 192	99.30 138	92.69 217	96.69 212	90.76 231	88.85 231	94.96 230	93.69 230

CostFormer		95.61 225	93.35 231	98.24 214	99.48 200	98.03 222	98.65 212	99.83 86	96.93 223	99.42 168	98.83 168	83.65 238	97.08 205	90.39 232	89.54 230	94.94 231	96.11 223

dps		95.59 226	93.46 230	98.08 217	99.33 216	98.22 218	98.87 200	99.70 147	96.17 231	98.87 209	97.75 209	86.85 235	96.60 213	91.24 230	89.62 229	95.10 229	94.34 228

tpm cat1		95.52 227	93.49 229	97.88 222	99.28 220	97.87 228	98.65 212	99.77 124	97.27 214	99.46 161	98.04 204	90.99 223	95.46 221	88.57 233	88.14 232	94.64 234	93.54 231

test_method		91.96 228	95.51 222	87.82 229	70.84 237	82.79 238	92.13 237	87.74 232	98.88 122	95.40 237	99.20 144	98.04 189	85.65 233	97.71 212	94.95 216	95.13 228	97.00 219

GG-mvs-BLEND		70.44 229	96.91 207	39.57 230	3.32 240	96.51 232	91.01 238	4.05 236	97.03 219	33.20 239	94.67 227	97.75 191	7.59 236	98.28 203	96.85 211	98.24 211	97.26 217

testmvs		22.33 230	29.66 232	13.79 231	8.97 238	10.35 239	15.53 241	8.09 235	32.51 235	19.87 240	45.18 235	30.56 243	17.05 235	29.96 234	24.74 233	13.21 237	34.30 234

test123		21.52 231	28.47 233	13.42 232	7.29 239	10.12 240	15.70 240	8.31 234	31.54 236	19.34 241	36.33 236	37.40 242	17.14 234	27.45 235	23.17 234	12.73 238	33.30 235

uanet_test		0.00 232	0.00 234	0.00 233	0.00 241	0.00 241	0.00 242	0.00 237	0.00 237	0.00 242	0.00 237	0.00 244	0.00 237	0.00 236	0.00 235	0.00 239	0.00 236

sosnet-low-res		0.00 232	0.00 234	0.00 233	0.00 241	0.00 241	0.00 242	0.00 237	0.00 237	0.00 242	0.00 237	0.00 244	0.00 237	0.00 236	0.00 235	0.00 239	0.00 236

sosnet		0.00 232	0.00 234	0.00 233	0.00 241	0.00 241	0.00 242	0.00 237	0.00 237	0.00 242	0.00 237	0.00 244	0.00 237	0.00 236	0.00 235	0.00 239	0.00 236

TPM-MVS					99.47 201	97.86 229	97.79 234			98.49 223	97.62 212	99.83 95	95.33 222			98.90 201	98.77 188
Ray Leroy Khuboni and Hongjun Xu: Textureless Resilient Propagation Matching in Multiple View Stereosis (TPM-MVS). SATNAC 2025
RE-MVS-def										99.96 3

9.14												99.57 135

SR-MVS					99.73 147			99.74 141				99.88 75

Anonymous202405211			99.14 117		99.87 58	99.55 61	99.50 106	99.70 147	98.55 160		98.61 183	98.46 180	98.76 158	99.66 48	99.50 43	99.85 45	99.63 63

our_test_3					99.75 135	99.11 166	99.74 48

ambc			98.83 155		99.72 149	98.52 204	98.84 203		98.96 111	99.92 10	99.34 129	99.74 116	99.04 135	98.68 192	97.57 204	99.46 164	98.99 183

MTAPA										99.62 125		99.95 27

MTMP										99.53 143		99.92 56

Patchmatch-RL test							65.75 239

tmp_tt				88.14 228	96.68 236	91.91 236	93.70 236	61.38 233	99.61 24	90.51 238	99.40 124	99.71 120	90.32 232	99.22 117	99.44 53	96.25 224

XVS					99.86 74	99.30 128	99.72 53			99.69 104		99.93 46				99.60 136

X-MVStestdata					99.86 74	99.30 128	99.72 53			99.69 104		99.93 46				99.60 136

mPP-MVS					99.84 88							99.92 56

NP-MVS									97.37 212

Patchmtry						98.19 220	98.91 196	99.97 1		99.43 165

DeepMVS_CX						96.39 233	97.15 235	88.89 231	97.94 199	99.51 151	95.71 225	97.88 190	98.19 174	98.92 162		97.73 215	97.75 213

High-resolution multi-view benchmark