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	97.71 1	99.33 1	99.47 1	99.16 7	99.16 41	99.11 14	99.39 12	99.16 11	99.26 2	99.22 5	99.51 18	99.75 4	98.54 15	99.71 1	99.47 3	99.52 12	99.46 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
SixPastTwentyTwo		99.25 2	99.20 3	99.32 1	99.53 14	99.32 8	99.64 2	99.19 10	98.05 11	99.19 6	99.74 4	98.96 50	99.03 2	99.69 2	99.58 1	99.32 25	99.06 6

WR-MVS		99.22 3	99.15 5	99.30 2	99.54 10	99.62 1	99.63 4	99.45 1	97.75 15	98.47 22	99.71 5	99.05 40	98.88 4	99.54 5	99.49 2	99.81 1	98.87 9

PS-CasMVS		99.08 4	98.90 11	99.28 3	99.65 3	99.56 4	99.59 6	99.39 3	96.36 36	98.83 14	99.46 21	99.09 33	98.62 10	99.51 7	99.36 8	99.63 3	98.97 7

PEN-MVS		99.08 4	98.95 8	99.23 5	99.65 3	99.59 2	99.64 2	99.34 6	96.68 28	98.65 17	99.43 23	99.33 16	98.47 17	99.50 8	99.32 9	99.60 5	98.79 11

v7n		99.03 6	99.03 7	99.02 9	99.09 52	99.11 14	99.57 9	98.82 19	98.21 9	99.25 3	99.84 2	99.59 6	98.76 6	99.23 19	98.83 32	98.63 71	98.40 33

DTE-MVSNet		99.03 6	98.88 12	99.21 6	99.66 2	99.59 2	99.62 5	99.34 6	96.92 24	98.52 19	99.36 29	98.98 46	98.57 13	99.49 9	99.23 12	99.56 9	98.55 25

TDRefinement		99.00 8	99.13 6	98.86 10	98.99 62	99.05 19	99.58 7	98.29 49	98.96 4	97.96 37	99.40 26	98.67 76	98.87 5	99.60 3	99.46 4	99.46 18	98.74 14

WR-MVS_H		98.97 9	98.82 14	99.14 8	99.56 8	99.56 4	99.54 11	99.42 2	96.07 41	98.37 24	99.34 31	99.09 33	98.43 18	99.45 10	99.41 5	99.53 10	98.86 10

UniMVSNet_ETH3D		98.93 10	99.20 3	98.63 22	99.54 10	99.33 7	98.73 63	99.37 4	98.87 5	97.86 39	99.27 35	99.78 2	96.59 85	99.52 6	99.40 6	99.67 2	98.21 41

CP-MVSNet		98.91 11	98.61 19	99.25 4	99.63 5	99.50 6	99.55 10	99.36 5	95.53 67	98.77 16	99.11 42	98.64 79	98.57 13	99.42 11	99.28 11	99.61 4	98.78 12

anonymousdsp		98.85 12	98.88 12	98.83 11	98.69 82	98.20 79	99.68 1	97.35 123	97.09 23	98.98 10	99.86 1	99.43 10	98.94 3	99.28 14	99.19 13	99.33 23	99.08 5

pmmvs6		98.77 13	99.35 2	98.09 43	98.32 101	98.92 25	98.57 70	99.03 12	99.36 1	96.86 83	99.77 3	99.86 1	96.20 100	99.56 4	99.39 7	99.59 6	98.61 22

ACMH	95.26 7	98.75 14	98.93 9	98.54 25	98.86 67	99.01 21	99.58 7	98.10 68	98.67 6	97.30 61	99.18 39	99.42 11	98.40 19	99.19 21	98.86 30	98.99 48	98.19 42
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
COLMAP_ROB	96.84 2	98.75 14	98.82 14	98.66 20	99.14 45	98.79 39	99.30 17	97.67 95	98.33 8	97.82 41	99.20 38	99.18 31	98.76 6	99.27 17	98.96 22	99.29 27	98.03 46
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
UA-Net		98.66 16	98.60 22	98.73 15	99.83 1	99.28 9	98.56 72	99.24 8	96.04 42	97.12 70	98.44 79	98.95 51	98.17 28	99.15 24	99.00 21	99.48 17	99.33 3

DeepC-MVS	96.08 5	98.58 17	98.49 24	98.68 18	99.37 26	98.52 64	99.01 35	98.17 63	97.17 22	98.25 27	99.56 15	99.62 5	98.29 22	98.40 63	98.09 70	98.97 50	98.08 45
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
TranMVSNet+NR-MVSNet		98.45 18	98.22 31	98.72 17	99.32 31	99.06 17	98.99 36	98.89 14	95.52 68	97.53 50	99.42 25	98.83 63	98.01 34	98.55 55	98.34 57	99.57 8	97.80 57

CSCG		98.45 18	98.61 19	98.26 37	99.11 49	99.06 17	98.17 92	97.49 108	97.93 13	97.37 58	98.88 56	99.29 19	98.10 29	98.40 63	97.51 89	99.32 25	99.16 4

DVP-MVS++		98.44 20	98.92 10	97.88 63	99.17 39	99.00 22	98.89 46	98.26 51	97.54 18	96.05 117	99.35 30	99.76 3	96.34 95	98.79 37	98.65 41	98.56 77	99.35 2

Gipuma		98.43 21	98.15 34	98.76 14	99.00 61	98.29 76	97.91 107	98.06 70	99.02 3	99.50 1	96.33 129	98.67 76	99.22 1	99.02 27	98.02 75	98.88 62	97.66 65
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
ACMH+	94.90 8	98.40 22	98.71 17	98.04 53	98.93 64	98.84 32	99.30 17	97.86 87	97.78 14	94.19 174	98.77 66	99.39 13	98.61 11	99.33 13	99.07 14	99.33 23	97.81 56

ACMMPR		98.31 23	98.07 39	98.60 23	99.58 6	98.83 33	99.09 27	98.48 31	96.25 38	97.03 74	96.81 117	99.09 33	98.39 20	98.55 55	98.45 49	99.01 45	98.53 28

APDe-MVS		98.29 24	98.42 26	98.14 40	99.45 21	98.90 26	99.18 23	98.30 47	95.96 48	95.13 148	98.79 63	99.25 26	97.92 38	98.80 35	98.71 36	98.85 64	98.54 26
Zhaojie Zeng, Yuesong Wang, Tao Guan: Matching Ambiguity-Resilient Multi-View Stereo via Adaptive Patch Deformation. Pattern Recognition
DVP-MVS		98.27 25	98.61 19	97.87 64	99.17 39	99.03 20	99.07 29	98.17 63	96.75 27	94.35 169	98.92 52	99.58 7	97.86 41	98.67 46	98.70 37	98.63 71	98.63 20
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
TransMVSNet (Re)		98.23 26	98.72 16	97.66 77	98.22 110	98.73 50	98.66 66	98.03 75	98.60 7	96.40 102	99.60 12	98.24 100	95.26 122	99.19 21	99.05 17	99.36 20	97.64 66

DU-MVS		98.23 26	97.74 55	98.81 12	99.23 33	98.77 41	98.76 57	98.88 15	94.10 115	98.50 20	98.87 58	98.32 97	97.99 35	98.40 63	98.08 73	99.49 16	97.64 66

UniMVSNet (Re)		98.23 26	97.85 48	98.67 19	99.15 42	98.87 28	98.74 60	98.84 17	94.27 113	97.94 38	99.01 45	98.39 93	97.82 42	98.35 68	98.29 62	99.51 15	97.78 58

MIMVSNet1		98.22 29	98.51 23	97.87 64	99.40 25	98.82 37	99.31 16	98.53 28	97.39 19	96.59 93	99.31 33	99.23 28	94.76 132	98.93 32	98.67 39	98.63 71	97.25 90

HFP-MVS		98.17 30	98.02 40	98.35 35	99.36 27	98.62 56	98.79 56	98.46 34	96.24 39	96.53 95	97.13 113	98.98 46	98.02 33	98.20 71	98.42 51	98.95 54	98.54 26

Baseline_NR-MVSNet		98.17 30	97.90 45	98.48 29	99.23 33	98.59 57	98.83 53	98.73 24	93.97 120	96.95 77	99.66 7	98.23 102	97.90 39	98.40 63	99.06 16	99.25 29	97.42 82

TSAR-MVS + MP.		98.15 32	98.23 30	98.06 51	98.47 92	98.16 85	99.23 20	96.87 139	95.58 62	96.72 86	98.41 80	99.06 37	98.05 32	98.99 29	98.90 26	99.00 46	98.51 29
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
pm-mvs1		98.14 33	98.66 18	97.53 86	97.93 132	98.49 66	98.14 94	98.19 59	97.95 12	96.17 113	99.63 10	98.85 59	95.41 120	98.91 33	98.89 27	99.34 22	97.86 55

SMA-MVS		98.13 34	98.22 31	98.02 56	99.44 23	98.73 50	98.24 89	97.87 86	95.22 75	96.76 85	98.66 72	99.35 15	97.03 70	98.53 58	98.39 53	98.80 66	98.69 16
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
ACMMP_NAP		98.12 35	98.08 38	98.18 39	99.34 28	98.74 49	98.97 38	98.00 77	95.13 79	96.90 78	97.54 102	99.27 23	97.18 64	98.72 42	98.45 49	98.68 70	98.69 16

UniMVSNet_NR-MVSNet		98.12 35	97.56 63	98.78 13	99.13 47	98.89 27	98.76 57	98.78 20	93.81 123	98.50 20	98.81 62	97.64 123	97.99 35	98.18 74	97.92 77	99.53 10	97.64 66

ACMM	94.29 11	98.12 35	97.71 56	98.59 24	99.51 16	98.58 59	99.24 19	98.25 52	96.22 40	96.90 78	95.01 153	98.89 56	98.52 16	98.66 48	98.32 60	99.13 36	98.28 39
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
SteuartSystems-ACMMP		98.06 38	97.78 53	98.39 33	99.54 10	98.79 39	98.94 42	98.42 36	93.98 119	95.85 124	96.66 123	99.25 26	98.61 11	98.71 44	98.38 54	98.97 50	98.67 19
Skip Steuart: Steuart Systems R&D Blog.
SED-MVS		98.05 39	98.46 25	97.57 82	99.01 58	98.99 23	98.82 55	98.24 53	95.76 57	94.70 159	98.96 47	99.49 9	96.19 101	98.74 38	98.65 41	98.46 85	98.63 20

OPM-MVS		98.01 40	98.01 41	98.00 58	99.11 49	98.12 88	98.68 64	97.72 93	96.65 30	96.68 90	98.40 81	99.28 22	97.44 55	98.20 71	97.82 84	98.40 91	97.58 71
Ray L. Khuboni, Hongjun Xu: Octagram Propagation Matching for Multi-Scale View Stereopsis (OPM-MVS).
Vis-MVSNet		98.01 40	98.42 26	97.54 85	96.89 181	98.82 37	99.14 24	97.59 98	96.30 37	97.04 73	99.26 36	98.83 63	96.01 106	98.73 40	98.21 64	98.58 76	98.75 13
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
CS-MVS		98.00 42	97.38 68	98.73 15	98.72 77	99.15 11	99.12 26	98.76 21	91.58 153	98.15 31	96.70 121	98.72 75	98.20 24	98.64 51	98.92 24	99.43 19	97.97 49

NR-MVSNet		98.00 42	97.88 46	98.13 41	98.33 99	98.77 41	98.83 53	98.88 15	94.10 115	97.46 55	98.87 58	98.58 84	95.78 109	99.13 25	98.16 68	99.52 12	97.53 74

CP-MVS		98.00 42	97.57 62	98.50 26	99.47 20	98.56 61	98.91 44	98.38 42	94.71 94	97.01 75	95.20 149	99.06 37	98.20 24	98.61 52	98.46 46	99.02 43	98.40 33

DPE-MVS		97.99 45	98.12 35	97.84 67	98.65 86	98.86 29	98.86 50	98.05 73	94.18 114	95.49 141	98.90 54	99.33 16	97.11 66	98.53 58	98.65 41	98.86 63	98.39 35
Kehua Chen, Zhenlong Yuan, Tianlu Mao, Zhaoqi Wang: Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization. AAAI2025
ACMMP		97.99 45	97.60 61	98.45 31	99.53 14	98.83 33	99.13 25	98.30 47	94.57 100	96.39 106	95.32 147	98.95 51	98.37 21	98.61 52	98.47 45	99.00 46	98.45 30
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
MP-MVS		97.98 47	97.53 64	98.50 26	99.56 8	98.58 59	98.97 38	98.39 41	93.49 126	97.14 67	96.08 136	99.23 28	98.06 31	98.50 60	98.38 54	98.90 58	98.44 31
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
EG-PatchMatch MVS		97.98 47	97.92 43	98.04 53	98.84 70	98.04 96	97.90 108	96.83 142	95.07 81	98.79 15	99.07 43	99.37 14	97.88 40	98.74 38	98.16 68	98.01 113	96.96 98

ACMP	94.03 12	97.97 49	97.61 60	98.39 33	99.43 24	98.51 65	98.97 38	98.06 70	94.63 98	96.10 115	96.12 135	99.20 30	98.63 9	98.68 45	98.20 67	99.14 33	97.93 52
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
CS-MVS-test		97.96 50	97.38 68	98.64 21	98.57 88	99.13 12	99.36 13	98.66 25	91.67 152	98.17 30	96.91 116	98.84 61	97.99 35	98.80 35	98.88 28	99.08 41	97.43 81

LGP-MVS_train		97.96 50	97.53 64	98.45 31	99.45 21	98.64 55	99.09 27	98.27 50	92.99 138	96.04 118	96.57 124	99.29 19	98.66 8	98.73 40	98.42 51	99.19 31	98.09 44

LS3D		97.93 52	97.80 50	98.08 47	99.20 36	98.77 41	98.89 46	97.92 82	96.59 31	96.99 76	96.71 120	97.14 135	96.39 94	99.04 26	98.96 22	99.10 40	97.39 83

SD-MVS		97.84 53	97.78 53	97.90 61	98.33 99	98.06 93	97.95 104	97.80 92	96.03 46	96.72 86	97.57 100	99.18 31	97.50 53	97.88 77	97.08 102	99.11 38	98.68 18
Zhenlong Yuan, Jiakai Cao, Zhaoxin Li, Hao Jiang and Zhaoqi Wang: SD-MVS: Segmentation-driven Deformation Multi-View Stereo with Spherical Refinement and EM optimization. AAAI2024
RPSCF		97.83 54	98.27 28	97.31 97	98.23 108	98.06 93	97.44 133	95.79 172	96.90 25	95.81 126	98.76 67	98.61 83	97.70 47	98.90 34	98.36 56	98.90 58	98.29 36

thisisatest0515		97.82 55	97.67 57	97.99 59	98.49 91	98.07 92	98.48 77	98.06 70	95.35 73	97.74 43	98.83 61	97.61 124	96.74 77	97.53 95	98.30 61	98.43 90	98.01 48

PGM-MVS		97.82 55	97.25 74	98.48 29	99.54 10	98.75 48	99.02 31	98.35 45	92.41 142	96.84 84	95.39 146	98.99 45	98.24 23	98.43 62	98.34 57	98.90 58	98.41 32

PMVS	90.51 17	97.77 57	97.98 42	97.53 86	98.68 83	98.14 87	97.67 118	97.03 134	96.43 32	98.38 23	98.72 69	97.03 137	94.44 137	99.37 12	99.30 10	98.98 49	96.86 105
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
MSP-MVS		97.67 58	97.88 46	97.43 92	99.34 28	98.99 23	98.87 49	98.12 66	95.63 59	94.16 175	97.45 103	99.50 8	96.44 93	96.35 133	98.70 37	97.65 129	98.57 24
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
tfpnnormal		97.66 59	97.79 51	97.52 88	98.32 101	98.53 63	98.45 80	97.69 94	97.59 17	96.12 114	97.79 95	96.70 142	95.69 113	98.35 68	98.34 57	98.85 64	97.22 93

FC-MVSNet-train		97.65 60	98.16 33	97.05 109	98.85 68	98.85 30	99.34 14	98.08 69	94.50 105	94.41 166	99.21 37	98.80 67	92.66 164	98.98 30	98.85 31	98.96 52	97.94 51

v10		97.64 61	97.26 73	98.08 47	98.07 121	98.56 61	98.86 50	98.18 61	94.48 106	98.24 28	99.56 15	98.98 46	97.72 46	96.05 143	96.26 130	97.42 138	96.93 99

EC-MVSNet		97.63 62	96.88 96	98.50 26	98.74 76	99.16 10	99.33 15	98.83 18	88.77 182	96.62 92	96.48 126	97.75 116	98.19 26	99.00 28	98.76 34	99.29 27	98.27 40

X-MVS		97.60 63	97.00 91	98.29 36	99.50 17	98.76 44	98.90 45	98.37 43	94.67 97	96.40 102	91.47 198	98.78 69	97.60 52	98.55 55	98.50 44	98.96 52	98.29 36

3Dnovator+	96.20 4	97.58 64	97.14 82	98.10 42	98.98 63	97.85 108	98.60 69	98.33 46	96.41 34	97.23 65	94.66 162	97.26 132	96.91 74	97.91 76	97.87 80	98.53 80	98.03 46

DCV-MVSNet		97.56 65	97.63 59	97.47 90	98.41 96	99.12 13	98.63 67	98.57 26	95.71 58	95.60 138	93.79 177	98.01 111	94.25 139	99.16 23	98.88 28	99.35 21	98.74 14

HPM-MVS++		97.56 65	97.11 86	98.09 43	99.18 38	97.95 103	98.57 70	98.20 57	94.08 117	97.25 64	95.96 140	98.81 66	97.13 65	97.51 96	97.30 99	98.21 101	98.15 43

FC-MVSNet-test		97.54 67	98.26 29	96.70 126	98.87 66	97.79 115	98.49 76	98.56 27	96.04 42	90.39 204	99.65 8	98.67 76	95.15 124	99.23 19	99.07 14	98.73 69	97.39 83

TSAR-MVS + ACMM		97.54 67	97.79 51	97.26 98	98.23 108	98.10 91	97.71 116	97.88 85	95.97 47	95.57 140	98.71 70	98.57 85	97.36 58	97.74 84	96.81 111	96.83 164	98.59 23

DeepC-MVS_fast	95.38 6	97.53 69	97.30 72	97.79 72	98.83 71	97.64 118	98.18 90	97.14 130	95.57 63	97.83 40	97.10 114	98.80 67	96.53 90	97.41 99	97.32 97	98.24 100	97.26 89
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
v1192		97.52 70	97.03 90	98.09 43	98.31 104	98.01 98	98.96 41	97.25 126	95.22 75	98.89 12	99.64 9	98.83 63	97.68 48	95.63 150	95.91 140	97.47 134	95.97 135

v1144		97.51 71	97.05 89	98.04 53	98.26 106	97.98 100	98.88 48	97.42 116	95.38 72	98.56 18	99.59 14	99.01 44	97.65 49	95.77 147	96.06 137	97.47 134	95.56 147

v8		97.51 71	97.16 80	97.91 60	97.99 128	98.48 67	98.76 57	98.17 63	94.54 104	97.69 45	99.48 20	98.76 72	97.63 51	96.10 142	96.14 132	97.20 148	96.64 113

v1921920		97.50 73	97.00 91	98.07 49	98.20 112	97.94 106	99.03 30	97.06 132	95.29 74	99.01 9	99.62 11	98.73 74	97.74 45	95.52 153	95.78 145	97.39 140	96.12 131

Anonymous20231211		97.49 74	97.91 44	97.00 112	98.31 104	98.72 52	98.27 87	97.84 89	94.76 93	94.77 158	98.14 88	98.38 95	93.60 149	98.96 31	98.66 40	99.22 30	97.77 60

v144192		97.49 74	96.99 93	98.07 49	98.11 120	97.95 103	99.02 31	97.21 127	94.90 89	98.88 13	99.53 17	98.89 56	97.75 44	95.59 151	95.90 141	97.43 137	96.16 129

test1111		97.48 76	97.20 77	97.81 71	98.78 74	98.85 30	98.68 64	98.40 37	96.68 28	94.84 156	99.13 41	90.32 190	97.01 71	99.27 17	99.05 17	99.19 31	97.10 95

GeoE		97.48 76	96.84 100	98.22 38	99.01 58	98.39 70	98.85 52	98.76 21	92.37 143	97.53 50	97.58 99	98.23 102	97.11 66	97.57 94	96.98 105	98.10 109	96.78 108

APD-MVS		97.47 78	97.16 80	97.84 67	99.32 31	98.39 70	98.47 79	98.21 56	92.08 148	95.23 145	96.68 122	98.90 54	96.99 72	98.20 71	98.21 64	98.80 66	97.67 64
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
PVSNet_Blended_VisFu		97.44 79	97.14 82	97.79 72	99.15 42	98.44 68	98.32 85	97.66 96	93.74 125	97.73 44	98.79 63	96.93 140	95.64 118	97.69 86	96.91 108	98.25 99	97.50 77

PHI-MVS		97.44 79	97.17 79	97.74 75	98.14 117	98.41 69	98.03 100	97.50 106	92.07 149	98.01 36	97.33 108	98.62 82	96.02 105	98.34 70	98.21 64	98.76 68	97.24 92

v1240		97.43 81	96.87 99	98.09 43	98.25 107	97.92 107	99.02 31	97.06 132	94.77 92	99.09 8	99.68 6	98.51 88	97.78 43	95.25 158	95.81 143	97.32 144	96.13 130

ECVR-MVS		97.40 82	97.11 86	97.73 76	98.66 84	98.83 33	98.50 74	98.40 37	96.04 42	95.00 154	98.95 49	91.07 187	96.70 79	99.28 14	99.04 19	99.14 33	96.58 115

FMVSNet1		97.40 82	98.09 36	96.60 131	97.80 146	98.76 44	98.26 88	98.50 30	96.79 26	93.13 192	99.28 34	98.64 79	92.90 161	97.67 88	97.86 81	99.02 43	97.64 66

casdiffmvs_mvg		97.34 84	97.65 58	96.97 113	97.74 149	98.33 74	98.45 80	97.18 128	95.81 53	93.92 179	99.04 44	99.05 40	95.48 119	97.00 116	97.71 87	99.07 42	96.63 114
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
v2v482		97.33 85	96.84 100	97.90 61	98.19 113	97.83 109	98.74 60	97.44 114	95.42 71	98.23 29	99.46 21	98.84 61	97.46 54	95.51 154	96.10 135	97.36 142	94.72 157

EPP-MVSNet		97.29 86	96.88 96	97.76 74	98.70 79	99.10 16	98.92 43	98.36 44	95.12 80	93.36 190	97.39 105	91.00 188	97.65 49	98.72 42	98.91 25	99.58 7	97.92 53

MVS_111021_HR		97.27 87	97.11 86	97.46 91	98.46 93	97.82 112	97.50 129	96.86 140	94.97 85	97.13 69	96.99 115	98.39 93	96.82 76	97.65 91	97.38 92	98.02 112	96.56 118

SF-MVS		97.26 88	97.43 66	97.05 109	98.80 73	97.83 109	96.02 182	97.44 114	94.98 84	95.74 130	97.16 111	98.45 92	95.72 111	97.85 78	97.97 76	98.60 74	97.78 58

TSAR-MVS + GP.		97.26 88	97.33 71	97.18 103	98.21 111	98.06 93	96.38 173	97.66 96	93.92 122	95.23 145	98.48 77	98.33 96	97.41 56	97.63 92	97.35 93	98.18 103	97.57 72

OMC-MVS		97.23 90	97.21 76	97.25 101	97.85 137	97.52 127	97.92 106	95.77 173	95.83 52	97.09 72	97.86 93	98.52 87	96.62 83	97.51 96	96.65 117	98.26 97	96.57 116

3Dnovator	96.31 3	97.22 91	97.19 78	97.25 101	98.14 117	97.95 103	98.03 100	96.77 145	96.42 33	97.14 67	95.11 150	97.59 125	95.14 126	97.79 82	97.72 85	98.26 97	97.76 62

MVS_0304		97.18 92	96.84 100	97.58 81	99.15 42	98.19 80	98.11 95	97.81 91	92.36 144	98.06 34	97.43 104	99.06 37	94.24 140	96.80 121	96.54 121	98.12 107	97.52 75

canonicalmvs		97.11 93	96.88 96	97.38 93	98.34 98	98.72 52	97.52 128	97.94 80	95.60 60	95.01 153	94.58 163	94.50 166	96.59 85	97.84 79	98.03 74	98.90 58	98.91 8

V42		97.10 94	96.97 94	97.26 98	97.64 153	97.60 120	98.45 80	95.99 161	94.44 107	97.35 59	99.40 26	98.63 81	97.34 60	96.33 136	96.38 127	96.82 166	96.00 133

CPTT-MVS		97.08 95	96.25 114	98.05 52	99.21 35	98.30 75	98.54 73	97.98 78	94.28 111	95.89 123	89.57 207	98.54 86	98.18 27	97.82 81	97.32 97	98.54 78	97.91 54

DeepPCF-MVS	94.55 10	97.05 96	97.13 85	96.95 115	96.06 196	97.12 144	98.01 102	95.44 179	95.18 77	97.50 52	97.86 93	98.08 107	97.31 62	97.23 104	97.00 104	97.36 142	97.45 79

QAPM		97.04 97	97.14 82	96.93 117	97.78 148	98.02 97	97.36 138	96.72 146	94.68 96	96.23 108	97.21 110	97.68 121	95.70 112	97.37 100	97.24 101	97.78 122	97.77 60

CNVR-MVS		97.03 98	96.77 105	97.34 94	98.89 65	97.67 117	97.64 121	97.17 129	94.40 109	95.70 134	94.02 172	98.76 72	96.49 92	97.78 83	97.29 100	98.12 107	97.47 78

casdiffmvs		97.00 99	97.36 70	96.59 132	97.65 152	97.98 100	98.06 97	96.81 143	95.78 55	92.77 198	99.40 26	99.26 25	95.65 117	96.70 125	96.39 126	98.59 75	95.99 134
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
v148		96.99 100	96.70 107	97.34 94	97.89 135	97.23 136	98.33 84	96.96 135	95.57 63	97.12 70	98.99 46	99.40 12	97.23 63	96.22 139	95.45 150	96.50 172	94.02 169

DELS-MVS		96.90 101	97.24 75	96.50 137	97.85 137	98.18 81	97.88 111	95.92 165	93.48 127	95.34 143	98.86 60	98.94 53	94.03 143	97.33 102	97.04 103	98.00 114	96.85 106
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
MVS_111021_LR		96.86 102	96.72 106	97.03 111	97.80 146	97.06 147	97.04 152	95.51 178	94.55 101	97.47 53	97.35 107	97.68 121	96.66 81	97.11 109	96.73 113	97.69 126	96.57 116

PM-MVS		96.85 103	96.62 109	97.11 105	97.13 176	96.51 160	98.29 86	94.65 196	94.84 90	98.12 32	98.59 73	97.20 133	97.41 56	96.24 138	96.41 125	97.09 153	96.56 118

pmmvs-eth3d		96.84 104	96.22 116	97.56 83	97.63 155	96.38 167	98.74 60	96.91 138	94.63 98	98.26 26	99.43 23	98.28 98	96.58 87	94.52 168	95.54 148	97.24 146	94.75 156

CANet		96.81 105	96.50 110	97.17 104	99.10 51	97.96 102	97.86 112	97.51 104	91.30 156	97.75 42	97.64 97	97.89 114	93.39 153	96.98 117	96.73 113	97.40 139	96.99 97

Fast-Effi-MVS+		96.80 106	95.92 127	97.84 67	98.57 88	97.46 130	98.06 97	98.24 53	89.64 177	97.57 49	96.45 127	97.35 130	96.73 78	97.22 105	96.64 118	97.86 119	96.65 112

MCST-MVS		96.79 107	96.08 120	97.62 79	98.78 74	97.52 127	98.01 102	97.32 124	93.20 130	95.84 125	93.97 174	98.12 105	97.34 60	96.34 134	95.88 142	98.45 86	97.51 76

UGNet		96.79 107	97.82 49	95.58 160	97.57 158	98.39 70	98.48 77	97.84 89	95.85 51	94.68 160	97.91 92	99.07 36	87.12 204	97.71 85	97.51 89	97.80 120	98.29 36
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
TAPA-MVS	93.96 13	96.79 107	96.70 107	96.90 119	97.64 153	97.58 121	97.54 127	94.50 198	95.14 78	96.64 91	96.76 119	97.90 113	96.63 82	95.98 144	96.14 132	98.45 86	97.39 83
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
CLD-MVS		96.73 110	96.92 95	96.51 136	98.70 79	97.57 123	97.64 121	92.07 205	93.10 136	96.31 107	98.29 83	99.02 43	95.99 107	97.20 106	96.47 123	98.37 93	96.81 107
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
train_agg		96.68 111	95.93 126	97.56 83	99.08 53	97.16 140	98.44 83	97.37 120	91.12 160	95.18 147	95.43 145	98.48 90	97.36 58	96.48 130	95.52 149	97.95 117	97.34 87

CDPH-MVS		96.68 111	95.99 123	97.48 89	99.13 47	97.64 118	98.08 96	97.46 110	90.56 166	95.13 148	94.87 158	98.27 99	96.56 88	97.09 110	96.45 124	98.54 78	97.08 96

MSLP-MVS++		96.66 113	96.46 113	96.89 120	98.02 123	97.71 116	95.57 189	96.96 135	94.36 110	96.19 112	91.37 199	98.24 100	97.07 68	97.69 86	97.89 78	97.52 132	97.95 50

TinyColmap		96.64 114	96.07 121	97.32 96	97.84 142	96.40 164	97.63 123	96.25 156	95.86 50	98.98 10	97.94 91	96.34 149	96.17 102	97.30 103	95.38 153	97.04 155	93.24 176

IS_MVSNet		96.62 115	96.48 112	96.78 124	98.46 93	98.68 54	98.61 68	98.24 53	92.23 145	89.63 209	95.90 141	94.40 167	96.23 97	98.65 49	98.77 33	99.52 12	96.76 109

NCCC		96.56 116	95.68 129	97.59 80	99.04 57	97.54 126	97.67 118	97.56 102	94.84 90	96.10 115	87.91 210	98.09 106	96.98 73	97.20 106	96.80 112	98.21 101	97.38 86

WB-MVS		96.54 117	98.09 36	94.73 177	96.68 188	98.34 73	94.77 208	97.39 117	98.12 10	89.72 208	98.95 49	99.32 18	93.33 154	98.67 46	97.88 79	96.47 174	95.38 150

ETV-MVS		96.54 117	95.27 137	98.02 56	99.07 55	97.48 129	98.16 93	98.19 59	87.33 197	97.58 48	92.67 186	95.93 155	96.22 98	98.49 61	98.46 46	98.91 57	96.50 121

Effi-MVS+		96.46 119	95.28 136	97.85 66	98.64 87	97.16 140	97.15 150	98.75 23	90.27 170	98.03 35	93.93 175	96.21 150	96.55 89	96.34 134	96.69 116	97.97 116	96.33 124

IterMVS-LS		96.35 120	95.85 128	96.93 117	97.53 159	98.00 99	97.37 136	97.97 79	95.49 70	96.71 89	98.94 51	93.23 174	94.82 131	93.15 187	95.05 156	97.17 150	97.12 94
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
USDC		96.30 121	95.64 131	97.07 107	97.62 156	96.35 169	97.17 149	95.71 174	95.52 68	99.17 7	98.11 89	97.46 127	95.67 114	95.44 156	93.60 176	97.09 153	92.99 180

Vis-MVSNet (Re-imp)		96.29 122	96.50 110	96.05 145	97.96 131	97.83 109	97.30 141	97.86 87	93.14 132	88.90 212	96.80 118	95.28 159	95.15 124	98.37 67	98.25 63	99.12 37	95.84 137

MSDG		96.27 123	96.17 119	96.38 141	97.85 137	96.27 171	96.55 170	94.41 199	94.55 101	95.62 137	97.56 101	97.80 115	96.22 98	97.17 108	96.27 129	97.67 128	93.60 173

CNLPA		96.24 124	95.97 124	96.57 134	97.48 165	97.10 146	96.75 163	94.95 190	94.92 88	96.20 111	94.81 159	96.61 144	96.25 96	96.94 118	95.64 146	97.79 121	95.74 143

EIA-MVS		96.23 125	94.85 149	97.84 67	99.08 53	98.21 78	97.69 117	98.03 75	85.68 207	98.09 33	91.75 197	97.07 136	95.66 116	97.58 93	97.72 85	98.47 84	95.91 136

PLC	92.55 15	96.10 126	95.36 133	96.96 114	98.13 119	96.88 151	96.49 171	96.67 150	94.07 118	95.71 133	91.14 200	96.09 152	96.84 75	96.70 125	96.58 120	97.92 118	96.03 132
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
test20.03		96.08 127	96.80 103	95.25 169	99.19 37	97.58 121	97.24 146	97.56 102	94.95 87	91.91 199	98.58 74	98.03 109	87.88 200	97.43 98	96.94 107	97.69 126	94.05 168

FA-MVS(training)		96.07 128	95.59 132	96.63 129	98.00 127	97.44 131	97.36 138	98.53 28	92.21 146	95.97 120	96.18 133	94.22 170	92.98 158	96.79 122	96.70 115	96.95 160	95.56 147

TSAR-MVS + COLMAP		96.05 129	95.94 125	96.18 144	97.46 166	96.41 163	97.26 145	95.83 169	94.69 95	95.30 144	98.31 82	96.52 145	94.71 133	95.48 155	94.87 158	96.54 171	95.33 151

EU-MVSNet		96.03 130	96.23 115	95.80 154	95.48 209	94.18 190	98.99 36	91.51 207	97.22 21	97.66 46	99.15 40	98.51 88	98.08 30	95.92 145	92.88 183	93.09 196	95.72 144

PCF-MVS	92.69 14	95.98 131	95.05 144	97.06 108	98.43 95	97.56 124	97.76 114	96.65 151	89.95 175	95.70 134	96.18 133	98.48 90	95.74 110	93.64 179	93.35 180	98.09 111	96.18 128
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
HQP-MVS		95.97 132	95.01 146	97.08 106	98.72 77	97.19 138	97.07 151	96.69 149	91.49 154	95.77 129	92.19 192	97.93 112	96.15 103	94.66 165	94.16 167	98.10 109	97.45 79

Effi-MVS+-dtu		95.94 133	95.08 143	96.94 116	98.54 90	97.38 132	96.66 167	97.89 84	88.68 183	95.92 121	92.90 185	97.28 131	94.18 142	96.68 127	96.13 134	98.45 86	96.51 120

diffmvs		95.86 134	96.21 117	95.44 163	97.25 174	96.85 154	96.99 155	95.23 184	94.96 86	92.82 197	98.89 55	98.85 59	93.52 151	94.21 174	94.25 166	96.84 163	95.49 149
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
AdaColmap		95.85 135	94.65 152	97.26 98	98.70 79	97.20 137	97.33 140	97.30 125	91.28 158	95.90 122	88.16 209	96.17 151	96.60 84	97.34 101	96.82 110	97.71 123	95.60 146

FMVSNet2		95.77 136	96.20 118	95.27 167	96.77 184	98.18 81	97.28 142	97.90 83	93.12 133	91.37 201	98.25 85	96.05 153	90.04 184	94.96 163	95.94 139	98.28 94	96.90 100

OpenMVS	94.63 9	95.75 137	95.04 145	96.58 133	97.85 137	97.55 125	96.71 165	96.07 158	90.15 173	96.47 97	90.77 205	95.95 154	94.41 138	97.01 115	96.95 106	98.00 114	96.90 100

pmmvs5		95.70 138	95.22 138	96.26 142	96.55 191	97.24 135	97.50 129	94.99 189	90.95 162	96.87 80	98.47 78	97.40 128	94.45 136	92.86 188	94.98 157	97.23 147	94.64 159

Anonymous20231206		95.69 139	95.68 129	95.70 156	98.32 101	96.95 149	97.37 136	96.65 151	93.33 128	93.61 184	98.70 71	98.03 109	91.04 173	95.07 161	94.59 165	97.20 148	93.09 179

MAR-MVS		95.51 140	94.49 156	96.71 125	97.92 133	96.40 164	96.72 164	98.04 74	86.74 201	96.72 86	92.52 189	95.14 161	94.02 144	96.81 120	96.54 121	96.85 161	97.25 90
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
DI_MVS_plusplus_trai		95.48 141	94.51 154	96.61 130	97.13 176	97.30 134	98.05 99	96.79 144	93.75 124	95.08 151	96.38 128	89.76 192	94.95 127	93.97 178	94.82 162	97.64 130	95.63 145

MDA-MVSNet-bldmvs		95.45 142	95.20 139	95.74 155	94.24 214	96.38 167	97.93 105	94.80 191	95.56 66	96.87 80	98.29 83	95.24 160	96.50 91	98.65 49	90.38 195	94.09 190	91.93 184

PVSNet_BlendedMVS		95.44 143	95.09 141	95.86 152	97.31 171	97.13 142	96.31 176	95.01 187	88.55 186	96.23 108	94.55 166	97.75 116	92.56 166	96.42 131	95.44 151	97.71 123	95.81 138

PVSNet_Blended		95.44 143	95.09 141	95.86 152	97.31 171	97.13 142	96.31 176	95.01 187	88.55 186	96.23 108	94.55 166	97.75 116	92.56 166	96.42 131	95.44 151	97.71 123	95.81 138

pmmvs4		95.37 145	94.25 157	96.67 128	97.01 179	95.28 184	97.60 124	96.07 158	93.11 134	97.29 62	98.09 90	94.23 169	95.21 123	91.56 199	93.91 173	96.82 166	93.59 174

MVS_Test		95.34 146	94.88 148	95.89 151	96.93 180	96.84 155	96.66 167	97.08 131	90.06 174	94.02 176	97.61 98	96.64 143	93.59 150	92.73 191	94.02 171	97.03 156	96.24 125

GBi-Net		95.21 147	95.35 134	95.04 172	96.77 184	98.18 81	97.28 142	97.58 99	88.43 188	90.28 205	96.01 137	92.43 178	90.04 184	97.67 88	97.86 81	98.28 94	96.90 100

test1		95.21 147	95.35 134	95.04 172	96.77 184	98.18 81	97.28 142	97.58 99	88.43 188	90.28 205	96.01 137	92.43 178	90.04 184	97.67 88	97.86 81	98.28 94	96.90 100

IterMVS-SCA-FT		95.16 149	93.95 161	96.56 135	97.89 135	96.69 157	96.94 157	96.05 160	93.06 137	97.35 59	98.79 63	91.45 183	95.93 108	92.78 189	91.00 193	95.22 186	93.91 171

HyFIR lowres test		95.05 150	93.54 166	96.81 123	97.81 145	96.88 151	98.18 90	97.46 110	94.28 111	94.98 155	96.57 124	92.89 177	96.15 103	90.90 204	91.87 189	96.28 178	91.35 185

CHOSEN 1792x2688		94.98 151	94.69 151	95.31 165	97.27 173	95.58 180	97.90 108	95.56 177	95.03 82	93.77 183	95.65 143	99.29 19	95.30 121	91.51 200	91.28 192	92.05 204	94.50 161

CANet_DTU		94.96 152	94.62 153	95.35 164	98.03 122	96.11 173	96.92 159	95.60 176	88.59 185	97.27 63	95.27 148	96.50 146	88.77 196	95.53 152	95.59 147	95.54 184	94.78 155

CDS-MVSNet		94.91 153	95.17 140	94.60 181	97.85 137	96.21 172	96.90 161	96.39 154	90.81 163	93.40 188	97.24 109	94.54 165	85.78 210	96.25 137	96.15 131	97.26 145	95.01 154
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
DPM-MVS		94.86 154	93.90 163	95.99 147	98.19 113	96.52 159	96.29 178	95.95 162	93.11 134	94.61 162	88.17 208	96.44 147	93.77 148	93.33 182	93.54 178	97.11 152	96.22 126

MS-PatchMatch		94.84 155	94.76 150	94.94 175	96.38 192	94.69 189	95.90 184	94.03 201	92.49 141	93.81 181	95.79 142	96.38 148	94.54 134	94.70 164	94.85 159	94.97 188	94.43 163

thisisatest0530		94.81 156	93.06 172	96.85 122	98.01 124	97.18 139	96.93 158	97.36 121	89.73 176	95.80 127	94.98 154	77.88 213	94.89 128	96.73 124	97.35 93	98.13 106	97.54 73

tttt0517		94.81 156	93.04 173	96.88 121	98.15 116	97.37 133	96.99 155	97.36 121	89.51 178	95.74 130	94.89 156	77.53 215	94.89 128	96.94 118	97.35 93	98.17 104	97.70 63

testgi		94.81 156	96.05 122	93.35 192	99.06 56	96.87 153	97.57 126	96.70 148	95.77 56	88.60 214	93.19 183	98.87 58	81.21 218	97.03 114	96.64 118	96.97 159	93.99 170

PatchMatch-RL		94.79 159	93.75 165	96.00 146	96.80 183	95.00 186	95.47 194	95.25 183	90.68 165	95.80 127	92.97 184	93.64 172	95.67 114	96.13 141	95.81 143	96.99 158	92.01 183

FPMVS		94.70 160	94.99 147	94.37 183	95.84 202	93.20 195	96.00 183	91.93 206	95.03 82	94.64 161	94.68 160	93.29 173	90.95 174	98.07 75	97.34 96	96.85 161	93.29 175

new-patchmatchnet		94.48 161	94.02 159	95.02 174	97.51 163	95.00 186	95.68 188	94.26 200	97.32 20	95.73 132	99.60 12	98.22 104	91.30 169	94.13 175	84.41 205	95.65 183	89.45 196

IterMVS		94.48 161	93.46 168	95.66 157	97.52 160	96.43 161	97.20 147	94.73 194	92.91 140	96.44 98	98.75 68	91.10 185	94.53 135	92.10 195	90.10 197	93.51 193	92.84 182
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
MDTV_nov1_ep13_2view		94.39 163	93.34 169	95.63 158	97.23 175	95.33 183	97.76 114	96.84 141	94.55 101	97.47 53	98.96 47	97.70 119	93.88 145	92.27 193	86.81 203	90.56 206	87.73 204

Fast-Effi-MVS+-dtu		94.34 164	93.26 171	95.62 159	97.82 143	95.97 176	95.86 185	99.01 13	86.88 199	93.39 189	90.83 203	95.46 158	90.61 178	94.46 170	94.68 163	97.01 157	94.51 160

thres600view7		94.34 164	92.31 182	96.70 126	98.19 113	98.12 88	97.85 113	97.45 112	91.49 154	93.98 178	84.27 213	82.02 204	94.24 140	97.04 111	98.76 34	98.49 82	94.47 162

EPNet		94.33 166	93.52 167	95.27 167	98.81 72	94.71 188	96.77 162	98.20 57	88.12 191	96.53 95	92.53 188	91.19 184	85.25 214	95.22 159	95.26 154	96.09 181	97.63 70
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
test2506		94.29 167	91.43 190	97.64 78	98.66 84	98.83 33	98.50 74	98.40 37	96.04 42	94.45 165	94.88 157	55.05 229	96.70 79	99.28 14	99.04 19	99.14 33	96.87 104

GA-MVS		94.18 168	92.98 174	95.58 160	97.36 168	96.42 162	96.21 179	95.86 166	90.29 169	95.08 151	96.19 132	85.37 196	92.82 162	94.01 177	94.14 168	96.16 180	94.41 164

gg-mvs-nofinetune		94.13 169	93.93 162	94.37 183	97.99 128	95.86 177	95.45 197	99.22 9	97.61 16	95.10 150	99.50 19	84.50 197	81.73 217	95.31 157	94.12 169	96.71 169	90.59 189

baseline		94.07 170	94.50 155	93.57 190	96.34 193	93.40 194	95.56 192	92.39 204	92.07 149	94.00 177	98.24 86	97.51 126	89.19 190	91.75 197	92.72 184	93.96 192	95.79 140

FMVSNet3		94.06 171	93.85 164	94.31 186	95.46 210	97.80 114	96.34 174	97.58 99	88.43 188	90.28 205	96.01 137	92.43 178	88.67 197	91.82 196	93.96 172	97.53 131	96.50 121

thres400		94.04 172	91.94 185	96.50 137	97.98 130	97.82 112	97.66 120	96.96 135	90.96 161	94.20 172	83.24 215	82.82 202	93.80 146	96.50 129	98.09 70	98.38 92	94.15 166

dmvs_re		94.02 173	92.39 180	95.91 150	97.71 150	95.43 182	97.00 154	95.94 163	82.49 216	94.61 162	83.69 214	93.01 176	92.71 163	97.83 80	97.56 88	97.50 133	96.73 110

CVMVSNet		94.01 174	94.25 157	93.73 189	94.36 213	92.44 198	97.45 132	88.56 210	95.59 61	93.06 195	98.88 56	90.03 191	94.84 130	94.08 176	93.45 179	94.09 190	95.31 152

thres200		93.98 175	91.90 186	96.40 140	97.66 151	98.12 88	97.20 147	97.45 112	90.16 172	93.82 180	83.08 216	83.74 200	93.80 146	97.04 111	97.48 91	98.49 82	93.70 172

baseline1		93.89 176	92.82 176	95.14 171	97.62 156	96.97 148	96.12 180	96.36 155	91.30 156	91.53 200	94.68 160	80.72 206	90.80 176	95.71 148	96.29 128	98.44 89	94.09 167

tfpn200view9		93.80 177	91.75 187	96.20 143	97.52 160	98.15 86	97.48 131	97.47 109	87.65 193	93.56 186	83.03 217	84.12 198	92.62 165	97.04 111	98.09 70	98.52 81	94.17 165

MIMVSNet		93.68 178	93.96 160	93.35 192	97.82 143	96.08 174	96.34 174	98.46 34	91.28 158	86.67 219	94.95 155	94.87 163	84.39 215	94.53 166	94.65 164	96.45 175	91.34 186

pmnet_mix02		93.59 179	92.65 177	94.69 179	96.76 187	94.16 191	97.03 153	93.00 203	95.79 54	96.03 119	98.91 53	97.69 120	92.99 157	90.03 207	84.10 207	92.35 202	87.89 203

EPNet_dtu		93.45 180	92.51 179	94.55 182	98.39 97	91.67 207	95.46 195	97.50 106	86.56 202	97.38 57	93.52 178	94.20 171	85.82 209	93.31 184	92.53 185	92.72 198	95.76 142
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
IB-MVS	92.44 16	93.33 181	92.15 184	94.70 178	97.42 167	96.39 166	95.57 189	94.67 195	86.40 205	93.59 185	78.28 221	95.76 157	89.59 189	95.88 146	95.98 138	97.39 140	96.34 123
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
ET-MVSNet_ETH3D		93.18 182	90.80 193	95.95 148	96.05 197	96.07 175	96.92 159	96.51 153	89.34 179	95.63 136	94.08 171	72.31 224	93.13 155	94.33 172	94.83 160	97.44 136	94.65 158

thres100view900		92.93 183	90.89 192	95.31 165	97.52 160	96.82 156	96.41 172	95.08 185	87.65 193	93.56 186	83.03 217	84.12 198	91.12 172	94.53 166	96.91 108	98.17 104	93.21 177

N_pmnet		92.46 184	92.38 181	92.55 198	97.91 134	93.47 193	97.42 134	94.01 202	96.40 35	88.48 215	98.50 76	98.07 108	88.14 199	91.04 203	84.30 206	89.35 211	84.85 210

TAMVS		92.46 184	93.34 169	91.44 206	97.03 178	93.84 192	94.68 209	90.60 208	90.44 168	85.31 220	97.14 112	93.03 175	85.78 210	94.34 171	93.67 175	95.22 186	90.93 188

CMPMVS	71.81 19	92.34 186	92.85 175	91.75 204	92.70 218	90.43 212	88.84 221	88.56 210	85.87 206	94.35 169	90.98 201	95.89 156	91.14 171	96.14 140	94.83 160	94.93 189	95.78 141
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
baseline2		92.06 187	89.82 196	94.68 180	97.32 169	95.72 178	94.97 205	95.08 185	84.75 210	94.34 171	90.68 206	77.75 214	90.13 183	93.38 180	93.58 177	96.25 179	92.90 181

MVSTER		91.97 188	90.31 194	93.91 187	96.81 182	96.91 150	94.22 210	95.64 175	84.98 208	92.98 196	93.42 179	72.56 222	86.64 208	95.11 160	93.89 174	97.16 151	95.31 152

CR-MVSNet		91.94 189	88.50 199	95.94 149	96.14 195	92.08 202	95.23 200	98.47 32	84.30 212	96.44 98	94.58 163	75.57 216	92.92 159	90.22 205	92.22 186	96.43 176	90.56 190

gm-plane-assit		91.85 190	87.91 201	96.44 139	99.14 45	98.25 77	99.02 31	97.38 119	95.57 63	98.31 25	99.34 31	51.00 230	88.93 193	93.16 186	91.57 190	95.85 182	86.50 207

PMMVS		91.67 191	91.47 189	91.91 203	89.43 223	88.61 218	94.99 204	85.67 215	87.50 195	93.80 182	94.42 169	94.88 162	90.71 177	92.26 194	92.96 182	96.83 164	89.65 194

CHOSEN 280x420		91.55 192	90.27 195	93.05 195	94.61 212	88.01 219	96.56 169	94.62 197	88.04 192	94.20 172	92.66 187	86.60 194	90.82 175	95.06 162	91.89 188	87.49 216	89.61 195

PatchT		91.40 193	88.54 198	94.74 176	91.48 222	92.18 201	97.42 134	97.51 104	84.96 209	96.44 98	94.16 170	75.47 217	92.92 159	90.22 205	92.22 186	92.66 201	90.56 190

pmmvs3		91.20 194	91.40 191	90.96 208	91.71 221	91.08 208	95.41 198	81.34 219	87.36 196	94.57 164	95.02 152	94.30 168	90.42 179	94.28 173	89.26 199	92.30 203	88.49 201

test0.0.03 1		91.17 195	91.50 188	90.80 209	98.01 124	95.46 181	94.22 210	95.80 170	86.55 203	81.75 222	90.83 203	87.93 193	78.48 219	94.51 169	94.11 170	96.50 172	91.08 187

SCA		91.15 196	87.65 203	95.23 170	96.15 194	95.68 179	96.68 166	98.18 61	90.46 167	97.21 66	92.44 190	80.17 208	93.51 152	86.04 214	83.58 210	89.68 210	85.21 209

new_pmnet		90.85 197	92.26 183	89.21 212	93.68 217	89.05 217	93.20 218	84.16 218	92.99 138	84.25 221	97.72 96	94.60 164	86.80 207	93.20 185	91.30 191	93.21 194	86.94 206

RPMNet		90.52 198	86.27 212	95.48 162	95.95 200	92.08 202	95.55 193	98.12 66	84.30 212	95.60 138	87.49 211	72.78 221	91.24 170	87.93 209	89.34 198	96.41 177	89.98 193

MDTV_nov1_ep13		90.30 199	87.32 207	93.78 188	96.00 199	92.97 196	95.46 195	95.39 180	88.61 184	95.41 142	94.45 168	80.39 207	89.87 187	86.58 212	83.54 211	90.56 206	84.71 211

PatchmatchNet		89.98 200	86.23 213	94.36 185	96.56 190	91.90 206	96.07 181	96.72 146	90.18 171	96.87 80	93.36 182	78.06 212	91.46 168	84.71 218	81.40 215	88.45 213	83.97 215
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
ADS-MVSNet		89.89 201	87.70 202	92.43 200	95.52 207	90.91 210	95.57 189	95.33 181	93.19 131	91.21 202	93.41 180	82.12 203	89.05 191	86.21 213	83.77 209	87.92 214	84.31 212

tpm		89.84 202	86.81 209	93.36 191	96.60 189	91.92 205	95.02 203	97.39 117	86.79 200	96.54 94	95.03 151	69.70 225	87.66 201	88.79 208	86.19 204	86.95 218	89.27 197

test-LLR		89.77 203	87.47 205	92.45 199	98.01 124	89.77 214	93.25 216	95.80 170	81.56 218	89.19 210	92.08 193	79.59 209	85.77 212	91.47 201	89.04 201	92.69 199	88.75 198

FMVSNet5		89.65 204	87.60 204	92.04 202	95.63 206	96.61 158	94.82 207	94.75 192	80.11 222	87.72 217	77.73 222	73.81 220	83.81 216	95.64 149	96.08 136	95.49 185	93.21 177

EPMVS		89.28 205	86.28 211	92.79 197	96.01 198	92.00 204	95.83 186	95.85 168	90.78 164	91.00 203	94.58 163	74.65 218	88.93 193	85.00 216	82.88 213	89.09 212	84.09 214

test-mter		89.16 206	88.14 200	90.37 210	94.79 211	91.05 209	93.60 215	85.26 216	81.65 217	88.32 216	92.22 191	79.35 211	87.03 205	92.28 192	90.12 196	93.19 195	90.29 192

CostFormer		89.06 207	85.65 214	93.03 196	95.88 201	92.40 199	95.30 199	95.86 166	86.49 204	93.12 194	93.40 181	74.18 219	88.25 198	82.99 219	81.46 214	89.77 209	88.66 200

MVS-HIRNet		88.72 208	86.49 210	91.33 207	91.81 220	85.66 220	87.02 223	96.25 156	81.48 220	94.82 157	96.31 131	92.14 181	90.32 181	87.60 210	83.82 208	87.74 215	78.42 219

TESTMET0.1,1		88.60 209	87.47 205	89.93 211	94.23 215	89.77 214	93.25 216	84.47 217	81.56 218	89.19 210	92.08 193	79.59 209	85.77 212	91.47 201	89.04 201	92.69 199	88.75 198

dps		88.36 210	84.32 217	93.07 194	93.86 216	92.29 200	94.89 206	95.93 164	83.50 214	93.13 192	91.87 195	67.79 227	90.32 181	85.99 215	83.22 212	90.28 208	85.56 208

tpmrst		87.60 211	84.13 218	91.66 205	95.65 205	89.73 216	93.77 213	94.74 193	88.85 181	93.35 191	95.60 144	72.37 223	87.40 202	81.24 220	78.19 217	85.02 221	82.90 218

tpm cat1		87.19 212	82.78 219	92.33 201	95.66 204	90.61 211	94.19 212	95.27 182	86.97 198	94.38 167	90.91 202	69.40 226	87.21 203	79.57 222	77.82 218	87.25 217	84.18 213

E-PMN		86.94 213	85.10 215	89.09 214	95.77 203	83.54 223	89.89 220	86.55 212	92.18 147	87.34 218	94.02 172	83.42 201	89.63 188	93.32 183	77.11 219	85.33 219	72.09 220

EMVS		86.63 214	84.48 216	89.15 213	95.51 208	83.66 222	90.19 219	86.14 214	91.78 151	88.68 213	93.83 176	81.97 205	89.05 191	92.76 190	76.09 220	85.31 220	71.28 221

PMMVS2		86.47 215	92.62 178	79.29 216	92.01 219	85.63 221	93.74 214	86.37 213	93.95 121	54.18 227	98.19 87	97.39 129	58.46 220	96.57 128	93.07 181	90.99 205	83.55 217

MVE	72.99 18	85.37 216	89.43 197	80.63 215	74.43 224	71.94 225	88.25 222	89.81 209	93.27 129	67.32 225	96.32 130	91.83 182	90.40 180	93.36 181	90.79 194	73.55 224	88.49 201
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
test_method		61.30 217	70.45 220	50.62 217	22.69 226	30.92 227	68.31 226	25.76 221	80.56 221	68.71 223	82.80 219	91.08 186	44.64 221	80.50 221	56.70 221	73.64 223	70.58 222

GG-mvs-BLEND		61.03 218	87.02 208	30.71 219	0.74 229	90.01 213	78.90 225	0.74 225	84.56 211	9.46 228	79.17 220	90.69 189	1.37 225	91.74 198	89.13 200	93.04 197	83.83 216

testmvs		4.99 219	6.88 221	2.78 221	1.73 227	2.04 229	3.10 229	1.71 223	7.27 224	3.92 230	12.18 224	6.71 231	3.31 224	6.94 223	5.51 223	2.94 226	7.51 223

test123		4.41 220	5.71 222	2.88 220	1.28 228	2.21 228	3.09 230	1.65 224	6.35 225	4.98 229	8.53 225	3.88 232	3.46 223	5.79 224	5.71 222	2.85 227	7.50 224

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					97.49 164	96.32 170	95.05 202			94.36 168	91.82 196	96.92 141	88.89 195			96.67 170	96.22 126
Ray Leroy Khuboni and Hongjun Xu: Textureless Resilient Propagation Matching in Multiple View Stereosis (TPM-MVS). SATNAC 2025
RE-MVS-def										99.38 2

9.14												96.98 139

SR-MVS					99.33 30			98.40 37				98.90 54

Anonymous202405211			97.39 67		98.85 68	98.59 57	97.89 110	97.93 81	94.41 108		97.37 106	96.99 138	93.09 156	98.61 52	98.46 46	99.11 38	97.27 88

our_test_3					97.32 169	95.13 185	97.59 125

ambc			96.78 104		99.01 58	97.11 145	95.73 187		95.91 49	99.25 3	98.56 75	97.17 134	97.04 69	96.76 123	95.22 155	96.72 168	96.73 110

MTAPA										97.43 56		99.27 23

MTMP										97.63 47		99.03 42

Patchmatch-RL test							17.42 228

tmp_tt				45.72 218	60.00 225	38.74 226	45.50 227	12.18 222	79.58 223	68.42 224	67.62 223	65.04 228	22.12 222	84.83 217	78.72 216	66.08 225

XVS					99.48 18	98.76 44	99.22 21			96.40 102		98.78 69				98.94 55

X-MVStestdata					99.48 18	98.76 44	99.22 21			96.40 102		98.78 69				98.94 55

mPP-MVS					99.58 6							98.98 46

NP-MVS									89.27 180

Patchmtry						92.70 197	95.23 200	98.47 32		96.44 98

DeepMVS_CX						72.99 224	80.14 224	37.34 220	83.46 215	60.13 226	84.40 212	85.48 195	86.93 206	87.22 211		79.61 222	87.32 205

High-resolution multi-view benchmark