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.

HFP-MVS		98.48 12	98.62 14	98.32 13	99.39 19	99.33 24	99.27 13	97.42 21	98.27 10	95.25 26	98.34 13	98.83 28	99.08 1	98.26 37	98.08 29	99.48 31	99.26 37

ACMMPR		98.40 14	98.49 16	98.28 15	99.41 15	99.40 16	99.36 4	97.35 24	98.30 9	95.02 28	97.79 20	98.39 40	99.04 2	98.26 37	98.10 27	99.50 29	99.22 43

SD-MVS		98.52 10	98.77 11	98.23 17	98.15 53	99.26 29	98.79 30	97.59 18	98.52 4	96.25 18	97.99 18	99.75 7	99.01 3	98.27 36	97.97 35	99.59 7	99.63 2
Zhenlong Yuan, Jiakai Cao, Zhaoxin Li, Hao Jiang and Zhaoqi Wang: SD-MVS: Segmentation-driven Deformation Multi-View Stereo with Spherical Refinement and EM optimization. AAAI2024
CPTT-MVS		97.78 29	97.54 39	98.05 23	98.91 37	99.05 39	99.00 24	96.96 36	97.14 46	95.92 20	95.50 49	98.78 30	98.99 4	97.20 73	96.07 114	98.54 191	99.04 69

DVP-MVS		98.86 6	98.97 5	98.75 4	99.43 14	99.63 1	99.25 15	97.81 3	98.62 3	97.69 4	97.59 23	99.90 2	98.93 5	98.99 4	98.42 13	99.37 65	99.62 4
Zhenlong Yuan, Jinguo Luo, Fei Shen, Zhaoxin Li, Cong Liu, Tianlu Mao, Zhaoqi Wang: DVP-MVS: Synergize Depth-Edge and Visibility Prior for Multi-View Stereo. AAAI2025
MSLP-MVS++		98.04 25	97.93 35	98.18 18	99.10 29	99.09 38	98.34 40	96.99 35	97.54 33	96.60 15	94.82 56	98.45 38	98.89 6	97.46 66	98.77 4	99.17 114	99.37 22

TSAR-MVS + MP.		98.49 11	98.78 10	98.15 21	98.14 54	99.17 35	99.34 8	97.18 32	98.44 6	95.72 22	97.84 19	99.28 14	98.87 7	99.05 1	98.05 30	99.66 2	99.60 7
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
aaEdge-Enhanced		98.97 2	99.00 4	98.94 2	99.53 5	99.47 12	99.35 6	97.66 10	98.36 7	98.80 2	99.17 1	99.76 6	98.86 8	98.57 17	98.32 19	99.42 53	99.33 26

APDe-MVS		98.87 5	98.96 6	98.77 3	99.58 2	99.53 7	99.44 1	97.81 3	98.22 13	97.33 7	98.70 8	99.33 12	98.86 8	98.96 6	98.40 15	99.63 5	99.57 9
Zhaojie Zeng, Yuesong Wang, Tao Guan: Matching Ambiguity-Resilient Multi-View Stereo via Adaptive Patch Deformation. Pattern Recognition
SMA-MVS		98.66 9	98.89 9	98.39 11	99.60 1	99.41 15	99.00 24	97.63 15	97.78 21	95.83 21	98.33 14	99.83 4	98.85 10	98.93 8	98.56 7	99.41 56	99.40 21
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
CNVR-MVS		98.47 13	98.46 19	98.48 9	99.40 16	99.05 39	99.02 22	97.54 19	97.73 22	96.65 14	97.20 32	99.13 22	98.85 10	98.91 9	98.10 27	99.41 56	99.08 60

MED-MVS		99.04 1	99.09 2	98.97 1	99.54 4	99.50 9	99.29 12	97.86 1	98.88 1	98.85 1	99.17 1	99.73 8	98.82 12	98.80 13	98.22 21	99.50 29	99.33 26

PGM-MVS		97.81 28	98.11 31	97.46 31	99.55 3	99.34 23	99.32 11	94.51 48	96.21 73	93.07 40	98.05 17	97.95 45	98.82 12	98.22 40	97.89 42	99.48 31	99.09 59

CP-MVS		98.32 19	98.34 25	98.29 14	99.34 22	99.30 25	99.15 17	97.35 24	97.49 35	95.58 24	97.72 21	98.62 36	98.82 12	98.29 32	97.67 50	99.51 27	99.28 32

MCST-MVS		98.20 20	98.36 22	98.01 24	99.40 16	99.05 39	99.00 24	97.62 16	97.59 32	93.70 37	97.42 30	99.30 13	98.77 15	98.39 30	97.48 55	99.59 7	99.31 31

AdaColmap		97.53 33	96.93 51	98.24 16	99.21 25	98.77 68	98.47 37	97.34 26	96.68 59	96.52 16	95.11 54	96.12 61	98.72 16	97.19 75	96.24 108	99.17 114	98.39 145

ACMMP_NAP		98.20 20	98.49 16	97.85 27	99.50 6	99.40 16	99.26 14	97.64 14	97.47 37	92.62 50	97.59 23	99.09 24	98.71 17	98.82 12	97.86 43	99.40 59	99.19 47

DeepC-MVS_fast	96.13 1	98.13 22	98.27 28	97.97 26	99.16 28	99.03 45	99.05 21	97.24 29	98.22 13	94.17 35	95.82 44	98.07 42	98.69 18	98.83 11	98.80 2	99.52 22	99.10 57
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
DVP-MVS++		98.92 3	99.18 1	98.61 6	99.47 7	99.61 2	99.39 3	97.82 2	98.80 2	96.86 11	98.90 4	99.92 1	98.67 19	99.02 2	98.20 23	99.43 50	99.82 1

SED-MVS		98.90 4	99.07 3	98.69 5	99.38 20	99.61 2	99.33 10	97.80 5	98.25 11	97.60 5	98.87 6	99.89 3	98.67 19	99.02 2	98.26 20	99.36 67	99.61 6

MSP-MVS		98.73 8	98.93 7	98.50 8	99.44 13	99.57 4	99.36 4	97.65 12	98.14 15	96.51 17	98.49 10	99.65 10	98.67 19	98.60 15	98.42 13	99.40 59	99.63 2
Zhenlong Yuan, Cong Liu, Fei Shen, Zhaoxin Li, Jingguo luo, Tianlu Mao and Zhaoqi Wang: MSP-MVS: Multi-granularity Segmentation Prior Guided Multi-View Stereo. AAAI2025
HPM-MVS++		98.34 18	98.47 18	98.18 18	99.46 10	99.15 36	99.10 19	97.69 9	97.67 28	94.93 29	97.62 22	99.70 9	98.60 22	98.45 24	97.46 56	99.31 75	99.26 37

DPE-MVS		98.75 7	98.91 8	98.57 7	99.21 25	99.54 6	99.42 2	97.78 7	97.49 35	96.84 12	98.94 3	99.82 5	98.59 23	98.90 10	98.22 21	99.56 17	99.48 17
Kehua Chen, Zhenlong Yuan, Tianlu Mao, Zhaoqi Wang: Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization. AAAI2025
NCCC		98.10 23	98.05 33	98.17 20	99.38 20	99.05 39	99.00 24	97.53 20	98.04 17	95.12 27	94.80 57	99.18 20	98.58 24	98.49 21	97.78 47	99.39 62	98.98 77

MP-MVS		98.09 24	98.30 27	97.84 28	99.34 22	99.19 34	99.23 16	97.40 22	97.09 48	93.03 43	97.58 25	98.85 27	98.57 25	98.44 26	97.69 49	99.48 31	99.23 41
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
X-MVS		97.84 27	98.19 30	97.42 32	99.40 16	99.35 20	99.06 20	97.25 28	97.38 39	90.85 73	96.06 41	98.72 32	98.53 26	98.41 28	98.15 26	99.46 36	99.28 32

APD-MVS		98.36 17	98.32 26	98.41 10	99.47 7	99.26 29	99.12 18	97.77 8	96.73 57	96.12 19	97.27 31	98.88 26	98.46 27	98.47 22	98.39 16	99.52 22	99.22 43
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
MVSMamba_PlusPlus		96.66 52	97.63 37	95.52 60	94.94 109	99.02 47	97.77 53	92.59 70	97.73 22	89.99 97	95.56 48	94.81 68	98.43 28	98.58 16	98.53 8	99.40 59	99.16 52

EC-MVSNet		96.49 53	97.63 37	95.16 67	94.75 118	98.69 74	97.39 60	88.97 152	96.34 69	92.02 56	96.04 42	96.46 55	98.21 29	98.41 28	97.96 36	99.61 6	99.55 10

train_agg		97.65 32	98.06 32	97.18 35	98.94 34	98.91 59	98.98 28	97.07 34	96.71 58	90.66 80	97.43 29	99.08 25	98.20 30	97.96 49	97.14 68	99.22 96	99.19 47

SPE-MVS-test		97.00 42	97.85 36	96.00 53	97.77 59	99.56 5	96.35 96	91.95 80	97.54 33	92.20 53	96.14 40	96.00 64	98.19 31	98.46 23	97.78 47	99.57 14	99.45 19

DeepC-MVS	94.87 4	96.76 51	96.50 57	97.05 37	98.21 52	99.28 27	98.67 31	97.38 23	97.31 40	90.36 89	89.19 111	93.58 76	98.19 31	98.31 31	98.50 9	99.51 27	99.36 23
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
CS-MVS		96.87 46	97.41 43	96.24 48	97.42 64	99.48 11	97.30 61	91.83 88	97.17 44	93.02 44	94.80 57	94.45 71	98.16 33	98.61 14	97.85 44	99.69 1	99.50 13

SteuartSystems-ACMMP		98.38 16	98.71 13	97.99 25	99.34 22	99.46 13	99.34 8	97.33 27	97.31 40	94.25 33	98.06 16	99.17 21	98.13 34	98.98 5	98.46 11	99.55 18	99.54 11
Skip Steuart: Steuart Systems R&D Blog.
SF-MVS		98.39 15	98.45 20	98.33 12	99.45 11	99.05 39	98.27 41	97.65 12	97.73 22	97.02 10	98.18 15	99.25 17	98.11 35	98.15 42	97.62 51	99.45 40	99.19 47

CSCG		97.44 35	97.18 47	97.75 29	99.47 7	99.52 8	98.55 35	95.41 43	97.69 27	95.72 22	94.29 60	95.53 66	98.10 36	96.20 123	97.38 60	99.24 87	99.62 4

3Dnovator+	93.91 7	97.23 38	97.22 44	97.24 34	98.89 39	98.85 64	98.26 42	93.25 60	97.99 18	95.56 25	90.01 105	98.03 44	98.05 37	97.91 50	98.43 12	99.44 47	99.35 24

TSAR-MVS + GP.		97.45 34	98.36 22	96.39 44	95.56 90	98.93 56	97.74 54	93.31 57	97.61 31	94.24 34	98.44 12	99.19 19	98.03 38	97.60 60	97.41 58	99.44 47	99.33 26

PLC	94.95 3	97.37 36	96.77 54	98.07 22	98.97 33	98.21 117	97.94 50	96.85 38	97.66 29	97.58 6	93.33 66	96.84 52	98.01 39	97.13 77	96.20 110	99.09 128	98.01 161
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
3Dnovator	93.79 8	97.08 40	97.20 45	96.95 40	99.09 30	99.03 45	98.20 43	93.33 56	97.99 18	93.82 36	90.61 99	96.80 53	97.82 40	97.90 51	98.78 3	99.47 35	99.26 37

LS3D		95.46 62	95.14 80	95.84 55	97.91 58	98.90 61	98.58 34	97.79 6	97.07 49	83.65 160	88.71 116	88.64 108	97.82 40	97.49 64	97.42 57	99.26 85	97.72 173

CNLPA		96.90 45	96.28 60	97.64 30	98.56 45	98.63 82	96.85 73	96.60 39	97.73 22	97.08 9	89.78 107	96.28 59	97.80 42	96.73 92	96.63 94	98.94 152	98.14 157

ACMMP		97.37 36	97.48 41	97.25 33	98.88 40	99.28 27	98.47 37	96.86 37	97.04 50	92.15 54	97.57 26	96.05 63	97.67 43	97.27 71	95.99 119	99.46 36	99.14 56
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
sasdasda		95.25 68	95.45 72	95.00 71	95.27 98	98.72 71	96.89 70	89.82 137	96.51 63	90.84 76	93.72 63	86.01 128	97.66 44	95.78 140	97.94 38	99.54 19	99.50 13

canonicalmvs		95.25 68	95.45 72	95.00 71	95.27 98	98.72 71	96.89 70	89.82 137	96.51 63	90.84 76	93.72 63	86.01 128	97.66 44	95.78 140	97.94 38	99.54 19	99.50 13

QAPM		96.78 50	97.14 48	96.36 45	99.05 31	99.14 37	98.02 47	93.26 58	97.27 42	90.84 76	91.16 91	97.31 48	97.64 46	97.70 58	98.20 23	99.33 69	99.18 50

TAPA-MVS	94.18 5	96.38 54	96.49 58	96.25 46	98.26 51	98.66 77	98.00 48	94.96 46	97.17 44	89.48 108	92.91 71	96.35 57	97.53 47	96.59 101	95.90 122	99.28 79	97.82 165
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
PHI-MVS		97.78 29	98.44 21	97.02 38	98.73 41	99.25 31	98.11 44	95.54 42	96.66 60	92.79 47	98.52 9	99.38 11	97.50 48	97.84 52	98.39 16	99.45 40	99.03 70

ETV-MVS		96.31 55	97.47 42	94.96 75	94.79 114	98.78 67	96.08 110	91.41 116	96.16 74	90.50 83	95.76 46	96.20 60	97.39 49	98.42 27	97.82 45	99.57 14	99.18 50

OMC-MVS		97.00 42	96.92 52	97.09 36	98.69 42	98.66 77	97.85 51	95.02 45	98.09 16	94.47 31	93.15 67	96.90 50	97.38 50	97.16 76	96.82 91	99.13 121	97.65 174

MGCFI-Net		95.12 70	95.39 75	94.79 85	95.24 100	98.68 75	96.80 77	89.72 141	96.48 65	90.11 93	93.64 65	85.86 133	97.36 51	95.69 146	97.92 41	99.53 21	99.49 16

MAR-MVS		95.50 59	95.60 68	95.39 64	98.67 43	98.18 121	95.89 125	89.81 139	94.55 127	91.97 57	92.99 69	90.21 95	97.30 52	96.79 89	97.49 54	98.72 176	98.99 75
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
DPM-MVS		96.86 47	96.82 53	96.91 41	98.08 55	98.20 118	98.52 36	97.20 31	97.24 43	91.42 60	91.84 83	98.45 38	97.25 53	97.07 78	97.40 59	98.95 150	97.55 177

MVS_111021_LR		97.16 39	98.01 34	96.16 49	98.47 47	98.98 51	96.94 69	93.89 52	97.64 30	91.44 59	98.89 5	96.41 56	97.20 54	98.02 48	97.29 66	99.04 143	98.85 92

CDPH-MVS		96.84 48	97.49 40	96.09 50	98.92 36	98.85 64	98.61 32	95.09 44	96.00 81	87.29 144	95.45 51	97.42 47	97.16 55	97.83 53	97.94 38	99.44 47	98.92 83

thres400		93.56 131	92.43 158	94.87 81	95.40 92	98.91 59	96.70 83	92.38 71	92.93 161	88.19 137	86.69 141	77.35 207	97.13 56	96.75 91	95.85 124	99.42 53	98.56 127

thres200		93.62 129	92.54 151	94.88 79	95.36 93	98.93 56	96.75 81	92.31 72	92.84 162	88.28 135	86.99 136	77.81 205	97.13 56	96.82 84	95.92 120	99.45 40	98.49 135

tfpn200view9		93.64 128	92.57 150	94.89 78	95.33 94	98.94 54	96.82 74	92.31 72	92.63 166	88.29 133	87.21 134	78.01 198	97.12 58	96.82 84	95.85 124	99.45 40	98.56 127

thres600view7		93.49 133	92.37 161	94.79 85	95.42 91	98.93 56	96.58 88	92.31 72	93.04 159	87.88 140	86.62 144	76.94 210	97.09 59	96.82 84	95.63 131	99.45 40	98.63 118

EIA-MVS		95.50 59	96.19 62	94.69 90	94.83 113	98.88 63	95.93 120	91.50 108	94.47 130	89.43 109	93.14 68	92.72 81	97.05 60	97.82 55	97.13 69	99.43 50	99.15 54

ET-MVSNet_ETH3D		93.34 135	94.33 98	92.18 142	83.26 255	97.66 133	96.72 82	89.89 136	95.62 97	87.17 145	96.00 43	83.69 168	96.99 61	93.78 187	95.34 139	99.06 137	98.18 156

TSAR-MVS + COLMAP		94.79 78	94.51 91	95.11 68	96.50 74	97.54 134	97.99 49	94.54 47	97.81 20	85.88 151	96.73 34	81.28 182	96.99 61	96.29 117	95.21 144	98.76 175	96.73 202

PCF-MVS	93.95 6	95.65 58	95.14 80	96.25 46	97.73 62	98.73 70	97.59 56	97.13 33	92.50 170	89.09 124	89.85 106	96.65 54	96.90 63	94.97 166	94.89 153	99.08 130	98.38 146
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
TSAR-MVS + ACMM		97.71 31	98.60 15	96.66 42	98.64 44	99.05 39	98.85 29	97.23 30	98.45 5	89.40 111	97.51 27	99.27 16	96.88 64	98.53 18	97.81 46	98.96 149	99.59 8

Casviewmamba		94.92 71	94.85 87	95.00 71	94.72 122	98.62 84	96.69 84	91.81 90	96.94 52	90.43 84	88.11 124	86.57 119	96.84 65	97.72 57	97.32 63	99.48 31	98.69 108

OpenMVS	92.33 11	95.50 59	95.22 78	95.82 56	98.98 32	98.97 52	97.67 55	93.04 65	94.64 125	89.18 118	84.44 177	94.79 69	96.79 66	97.23 72	97.61 52	99.24 87	98.88 88

thres100view900		93.55 132	92.47 157	94.81 84	95.33 94	98.74 69	96.78 79	92.30 75	92.63 166	88.29 133	87.21 134	78.01 198	96.78 67	96.38 110	95.92 120	99.38 63	98.40 143

Anonymous20231211		93.49 133	92.33 162	94.84 82	94.78 116	98.00 127	96.11 108	91.85 82	94.86 122	90.91 72	74.69 227	89.18 103	96.73 68	94.82 167	95.51 135	98.67 181	99.24 40

Effi-MVS+		92.93 142	93.86 119	91.86 144	94.07 158	98.09 126	95.59 136	85.98 187	94.27 135	79.54 181	91.12 94	81.81 179	96.71 69	96.67 96	96.06 115	99.27 82	98.98 77

MVS_111021_HR		97.04 41	98.20 29	95.69 57	98.44 49	99.29 26	96.59 87	93.20 61	97.70 26	89.94 100	98.46 11	96.89 51	96.71 69	98.11 45	97.95 37	99.27 82	99.01 73

E2		94.88 74	94.85 87	94.91 77	94.58 133	98.59 85	96.16 103	91.80 91	95.88 86	91.04 70	90.11 104	86.91 116	96.68 71	96.91 83	96.85 86	99.19 111	98.70 107

viewcassd2359sk11		94.63 84	94.45 93	94.84 82	94.58 133	98.57 89	96.13 106	91.79 92	95.32 104	90.67 79	88.73 115	86.13 126	96.65 72	96.82 84	96.87 85	99.21 102	98.68 110

viewdifsd2359ckpt09		94.40 94	94.26 99	94.57 93	94.51 140	98.50 103	95.96 119	91.72 99	95.31 108	89.37 112	88.33 121	85.88 131	96.64 73	96.61 97	96.57 97	99.20 109	98.60 123

E3new		94.34 96	93.98 116	94.75 87	94.56 135	98.56 92	96.13 106	91.78 94	94.54 129	90.22 90	87.24 132	85.36 140	96.62 74	96.61 97	96.90 79	99.22 96	98.68 110

Fast-Effi-MVS+		91.87 151	92.08 165	91.62 150	92.91 174	97.21 147	94.93 147	84.60 216	93.61 146	81.49 171	83.50 182	78.95 190	96.62 74	96.55 103	96.22 109	99.16 117	98.51 132

E3		94.33 97	93.99 115	94.73 88	94.56 135	98.56 92	96.14 104	91.78 94	94.55 127	90.05 95	87.23 133	85.39 138	96.61 76	96.61 97	96.90 79	99.21 102	98.68 110

MGCNet		97.94 26	98.72 12	97.02 38	98.48 46	99.50 9	99.02 22	94.06 50	98.33 8	94.51 30	98.78 7	97.73 46	96.60 77	98.51 19	98.68 5	99.45 40	99.53 12

casdiffmvs		94.38 95	94.15 111	94.64 92	94.70 126	98.51 97	96.03 117	91.66 102	95.70 93	89.36 113	86.48 149	85.03 150	96.60 77	97.40 67	97.30 64	99.52 22	98.67 113
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
casdiffmvs_mvg		94.55 87	94.26 99	94.88 79	94.96 108	98.51 97	97.11 63	91.82 89	94.28 134	89.20 117	86.60 145	86.85 117	96.56 79	97.47 65	97.25 67	99.64 4	98.83 95
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
hybridcas		94.67 83	94.44 94	94.94 76	94.66 129	98.57 89	96.76 80	91.72 99	96.60 61	90.57 81	86.88 137	85.79 134	96.53 80	97.55 63	97.07 70	99.43 50	98.62 120

E5new		93.95 113	93.42 131	94.57 93	94.50 143	98.51 97	96.18 101	91.84 83	93.55 149	89.12 121	85.80 163	84.38 156	96.53 80	96.16 127	96.85 86	99.23 94	98.67 113

E5		93.95 113	93.42 131	94.57 93	94.50 143	98.51 97	96.18 101	91.84 83	93.55 149	89.12 121	85.80 163	84.38 156	96.53 80	96.16 127	96.85 86	99.23 94	98.67 113

E4		93.88 119	93.38 135	94.48 98	94.50 143	98.51 97	96.08 110	91.74 98	93.42 155	88.84 127	85.51 166	84.38 156	96.49 83	96.22 120	96.90 79	99.22 96	98.69 108

Anonymous202405211			92.18 163		95.04 107	98.20 118	96.14 104	91.79 92	93.93 139		74.60 228	88.38 111	96.48 84	95.17 160	95.82 128	99.00 144	99.15 54

MVS_Test		94.82 76	95.66 67	93.84 123	94.79 114	98.35 108	96.49 91	89.10 151	96.12 77	87.09 146	92.58 74	90.61 92	96.48 84	96.51 108	96.89 83	99.11 125	98.54 129

E6new		93.85 120	93.39 133	94.39 104	94.50 143	98.53 95	95.93 120	91.41 116	93.47 151	88.81 128	85.51 166	84.16 162	96.46 86	96.32 115	96.99 75	99.21 102	98.78 100

E6		93.85 120	93.39 133	94.39 104	94.50 143	98.53 95	95.93 120	91.41 116	93.47 151	88.81 128	85.51 166	84.16 162	96.46 86	96.32 115	96.99 75	99.21 102	98.78 100

viewmanbaseed2359cas		94.31 99	94.25 101	94.38 106	94.72 122	98.59 85	96.09 109	91.84 83	95.35 102	87.92 139	87.86 126	85.54 136	96.45 88	96.71 93	97.04 71	99.26 85	98.67 113

viewdifsd2359ckpt13		94.14 106	94.00 113	94.30 110	94.55 137	98.55 94	95.71 133	91.76 96	95.03 119	88.12 138	87.34 129	85.15 145	96.39 89	96.81 88	96.60 95	99.24 87	98.50 133

dtuplus		93.75 126	93.15 141	94.46 99	94.41 152	98.12 125	96.06 114	91.45 112	94.25 136	89.32 116	85.82 161	85.24 143	96.38 90	93.99 186	95.83 126	99.12 123	98.78 100

viewmamba		94.27 102	94.15 111	94.42 102	94.77 117	98.24 115	95.87 128	91.46 111	97.44 38	88.99 125	88.77 114	85.11 147	96.34 91	94.77 168	96.19 112	99.07 132	98.53 130

ACMM	92.75 10	94.41 93	93.84 121	95.09 69	96.41 77	96.80 155	94.88 151	93.54 54	96.41 67	90.16 91	92.31 77	83.11 172	96.32 92	96.22 120	94.65 159	99.22 96	97.35 184
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
OPM-MVS		93.61 130	92.43 158	95.00 71	96.94 71	97.34 142	97.78 52	94.23 49	89.64 204	85.53 152	88.70 117	82.81 175	96.28 93	96.28 118	95.00 152	99.24 87	97.22 187
Ray L. Khuboni, Hongjun Xu: Octagram Propagation Matching for Multi-Scale View Stereopsis (OPM-MVS).
casdiffseed414692147		93.07 140	92.06 166	94.25 113	94.46 148	98.28 110	95.61 135	91.28 120	92.74 164	88.58 131	82.11 189	80.19 185	96.25 94	96.05 130	96.49 98	99.32 71	98.57 125

CANet		96.84 48	97.20 45	96.42 43	97.92 57	99.24 33	98.60 33	93.51 55	97.11 47	93.07 40	91.16 91	97.24 49	96.21 95	98.24 39	98.05 30	99.22 96	99.35 24

viewmacassd2359aftdt		93.65 127	93.29 138	94.07 117	94.61 131	98.51 97	96.04 116	91.75 97	93.61 146	86.56 149	84.89 172	84.41 155	96.17 96	95.97 132	97.03 72	99.28 79	98.63 118

viewmambaseed2359dif		93.92 117	93.38 135	94.54 96	94.55 137	98.15 122	96.41 93	91.47 110	95.10 116	89.58 106	86.64 142	85.10 148	96.17 96	94.08 185	95.77 129	99.09 128	98.84 94

diffmvs_AUTHOR		94.09 109	93.86 119	94.36 107	94.60 132	98.31 109	96.29 97	91.51 107	96.39 68	88.49 132	87.35 128	83.32 171	96.16 98	96.17 126	96.64 93	99.10 126	98.82 97

PMMVS		94.61 85	95.56 69	93.50 128	94.30 154	96.74 159	94.91 148	89.56 144	95.58 99	87.72 141	96.15 39	92.86 79	96.06 99	95.47 151	95.02 150	98.43 200	97.09 190

CLD-MVS		94.79 78	94.36 97	95.30 65	95.21 102	97.46 137	97.23 62	92.24 76	96.43 66	91.77 58	92.69 73	84.31 159	96.06 99	95.52 149	95.03 149	99.31 75	99.06 65
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
PatchMatch-RL		94.69 82	94.41 95	95.02 70	97.63 63	98.15 122	94.50 164	91.99 78	95.32 104	91.31 62	95.47 50	83.44 169	96.02 101	96.56 102	95.23 143	98.69 179	96.67 203

diffmvs		94.31 99	94.21 103	94.42 102	94.64 130	98.28 110	96.36 95	91.56 104	96.77 56	88.89 126	88.97 112	84.23 161	96.01 102	96.05 130	96.41 101	99.05 141	98.79 99
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
ACM-MVS					98.91 37	98.58 88	98.35 39		95.95 83	89.83 102	96.31 37	98.61 37	95.98 103			98.86 162	98.78 100

TPM-MVS					98.94 34	98.47 104	98.04 46			92.62 50	96.51 36	98.76 31	95.94 104			98.92 154	97.55 177
Ray Leroy Khuboni and Hongjun Xu: Textureless Resilient Propagation Matching in Multiple View Stereosis (TPM-MVS). SATNAC 2025
DCV-MVSNet		94.76 81	95.12 82	94.35 108	95.10 106	95.81 190	96.46 92	89.49 145	96.33 70	90.16 91	92.55 75	90.26 94	95.83 105	95.52 149	96.03 117	99.06 137	99.33 26

hybridnocas07		94.25 103	94.18 105	94.33 109	94.75 118	98.23 116	95.86 129	91.49 109	96.88 54	89.13 119	89.37 110	84.73 152	95.73 106	95.14 161	96.27 106	99.05 141	98.62 120

HyFIR lowres test		92.03 149	91.55 174	92.58 137	97.13 69	98.72 71	94.65 157	86.54 178	93.58 148	82.56 164	67.75 258	90.47 93	95.67 107	95.87 136	95.54 134	98.91 156	98.93 82

viewdifsd2359ckpt07		94.23 104	94.19 104	94.27 111	94.69 128	98.45 105	96.06 114	91.72 99	95.09 117	88.79 130	86.81 138	86.35 125	95.64 108	97.38 68	96.88 84	98.68 180	98.40 143

baseline1		94.59 86	94.47 92	94.72 89	95.16 103	97.97 129	96.07 112	91.94 81	94.86 122	89.98 98	91.60 87	85.87 132	95.64 108	97.07 78	96.90 79	99.52 22	97.06 194

GeoE		92.52 147	92.64 149	92.39 140	93.96 159	97.76 131	96.01 118	85.60 198	93.23 156	83.94 157	81.56 192	84.80 151	95.63 110	96.22 120	95.83 126	99.19 111	99.07 64

viewdifsd2359ckpt11		93.27 137	92.72 146	93.91 120	94.46 148	97.42 140	94.91 148	91.42 114	95.74 91	89.57 107	87.34 129	82.87 174	95.61 111	92.62 207	94.62 161	97.49 220	98.44 136

CHOSEN 280x420		95.46 62	97.01 49	93.66 126	97.28 68	97.98 128	96.40 94	85.39 203	96.10 78	91.07 69	96.53 35	96.34 58	95.61 111	97.65 59	96.95 78	96.21 237	97.49 179

viewmsd2359difaftdt		93.27 137	92.72 146	93.91 120	94.46 148	97.42 140	94.91 148	91.42 114	95.69 95	89.59 105	87.34 129	82.90 173	95.60 113	92.62 207	94.62 161	97.49 220	98.44 136

hybrid		94.23 104	94.23 102	94.24 114	94.70 126	98.20 118	95.66 134	91.43 113	96.94 52	89.13 119	89.47 109	84.64 153	95.59 114	95.56 147	96.20 110	98.95 150	98.57 125

HQP-MVS		94.43 91	94.57 90	94.27 111	96.41 77	97.23 146	96.89 70	93.98 51	95.94 84	83.68 159	95.01 55	84.46 154	95.58 115	95.47 151	94.85 157	99.07 132	99.00 74

MSDG		94.82 76	93.73 123	96.09 50	98.34 50	97.43 139	97.06 64	96.05 40	95.84 89	90.56 82	86.30 156	89.10 105	95.55 116	96.13 129	95.61 132	99.00 144	95.73 222

onestephybrid01		94.30 101	94.16 110	94.46 99	94.74 121	98.25 114	95.77 132	91.59 103	96.57 62	90.06 94	88.08 125	85.68 135	95.53 117	95.37 155	96.41 101	99.07 132	98.74 106

DeepPCF-MVS	95.28 2	97.00 42	98.35 24	95.42 63	97.30 67	98.94 54	94.82 152	96.03 41	98.24 12	92.11 55	95.80 45	98.64 35	95.51 118	98.95 7	98.66 6	96.78 227	99.20 46

DI_MVS_pp		94.01 111	93.63 125	94.44 101	94.54 139	98.26 113	97.51 57	90.63 127	95.88 86	89.34 114	80.54 201	89.36 100	95.48 119	96.33 114	96.27 106	99.17 114	98.78 100

FA-MVS(training)		93.94 115	95.16 79	92.53 138	94.87 112	98.57 89	95.42 139	79.49 241	95.37 101	90.98 71	86.54 147	94.26 73	95.44 120	97.80 56	95.19 145	98.97 147	98.38 146

EPP-MVSNet		95.27 67	96.18 63	94.20 115	94.88 111	98.64 80	94.97 146	90.70 126	95.34 103	89.67 104	91.66 86	93.84 74	95.42 121	97.32 70	97.00 74	99.58 11	99.47 18

RPSCF		94.05 110	94.00 113	94.12 116	96.20 79	96.41 169	96.61 86	91.54 105	95.83 90	89.73 103	96.94 33	92.80 80	95.35 122	91.63 224	90.44 227	95.27 254	93.94 242

test2506		94.32 98	93.00 143	95.87 54	96.16 80	99.39 18	96.96 67	92.80 67	95.22 112	94.47 31	91.55 88	70.45 242	95.25 123	98.29 32	97.98 33	99.59 7	98.10 159

ECVR-MVS		94.14 106	92.96 144	95.52 60	96.16 80	99.39 18	96.96 67	92.80 67	95.22 112	92.38 52	81.48 193	80.31 183	95.25 123	98.29 32	97.98 33	99.59 7	98.05 160

LGP-MVS_train		94.12 108	94.62 89	93.53 127	96.44 76	97.54 134	97.40 59	91.84 83	94.66 124	81.09 173	95.70 47	83.36 170	95.10 125	96.36 113	95.71 130	99.32 71	99.03 70

DELS-MVS		96.06 57	96.04 64	96.07 52	97.77 59	99.25 31	98.10 45	93.26 58	94.42 131	92.79 47	88.52 120	93.48 77	95.06 126	98.51 19	98.83 1	99.45 40	99.28 32
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
test1111		93.94 115	92.78 145	95.29 66	96.14 82	99.42 14	96.79 78	92.85 66	95.08 118	91.39 61	80.69 199	79.86 187	95.00 127	98.28 35	98.00 32	99.58 11	98.11 158

ACMP	92.88 9	94.43 91	94.38 96	94.50 97	96.01 85	97.69 132	95.85 130	92.09 77	95.74 91	89.12 121	95.14 53	82.62 177	94.77 128	95.73 143	94.67 158	99.14 120	99.06 65
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
thisisatest0530		94.54 88	95.47 71	93.46 129	94.51 140	98.65 79	94.66 156	90.72 124	95.69 95	86.90 147	93.80 61	89.44 99	94.74 129	96.98 82	94.86 154	99.19 111	98.85 92

tttt0517		94.52 89	95.44 74	93.44 130	94.51 140	98.68 75	94.61 159	90.72 124	95.61 98	86.84 148	93.78 62	89.26 102	94.74 129	97.02 81	94.86 154	99.20 109	98.87 90

baseline		94.83 75	95.82 66	93.68 125	94.75 118	97.80 130	96.51 90	88.53 157	97.02 51	89.34 114	92.93 70	92.18 83	94.69 131	95.78 140	96.08 113	98.27 203	98.97 81

PVSNet_BlendedMVS		95.41 64	95.28 76	95.57 58	97.42 64	99.02 47	95.89 125	93.10 63	96.16 74	93.12 38	91.99 79	85.27 141	94.66 132	98.09 46	97.34 61	99.24 87	99.08 60

PVSNet_Blended		95.41 64	95.28 76	95.57 58	97.42 64	99.02 47	95.89 125	93.10 63	96.16 74	93.12 38	91.99 79	85.27 141	94.66 132	98.09 46	97.34 61	99.24 87	99.08 60

FC-MVSNet-train		93.85 120	93.91 117	93.78 124	94.94 109	96.79 158	94.29 167	91.13 121	93.84 143	88.26 136	90.40 100	85.23 144	94.65 134	96.54 104	95.31 140	99.38 63	99.28 32

CANet_DTU		93.92 117	96.57 56	90.83 161	95.63 88	98.39 107	96.99 66	87.38 168	96.26 71	71.97 227	96.31 37	93.02 78	94.53 135	97.38 68	96.83 90	98.49 194	97.79 166

FMVSNet1		91.54 159	90.93 182	92.26 141	90.35 200	95.27 209	95.22 142	87.16 172	91.37 187	87.62 142	75.45 222	83.84 166	94.43 136	96.52 105	96.30 103	98.82 165	97.74 172

IterMVS-LS		92.56 146	93.18 139	91.84 145	93.90 160	94.97 216	94.99 145	86.20 182	94.18 137	82.68 163	85.81 162	87.36 115	94.43 136	95.31 156	96.02 118	98.87 160	98.60 123
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
GBi-Net		93.81 123	94.18 105	93.38 131	91.34 190	95.86 186	96.22 98	88.68 154	95.23 109	90.40 85	86.39 151	91.16 86	94.40 138	96.52 105	96.30 103	99.21 102	97.79 166

test1		93.81 123	94.18 105	93.38 131	91.34 190	95.86 186	96.22 98	88.68 154	95.23 109	90.40 85	86.39 151	91.16 86	94.40 138	96.52 105	96.30 103	99.21 102	97.79 166

FMVSNet2		93.30 136	93.36 137	93.22 134	91.34 190	95.86 186	96.22 98	88.24 160	95.15 115	89.92 101	81.64 191	89.36 100	94.40 138	96.77 90	96.98 77	99.21 102	97.79 166

0.4-1-1-0.1		89.64 186	88.08 206	91.46 151	86.21 243	94.41 229	94.79 153	86.20 182	88.54 213	91.15 67	86.64 142	78.03 195	94.36 141	84.47 256	88.05 237	96.08 240	96.40 206

0.3-1-1-0.015		89.40 189	87.72 214	91.36 153	86.10 245	94.08 235	94.62 158	86.10 184	88.02 218	91.16 63	86.39 151	77.89 201	94.30 142	83.93 259	87.88 238	95.88 242	95.86 218

IS_MVSNet		95.28 66	96.43 59	93.94 118	95.30 96	99.01 50	95.90 123	91.12 122	94.13 138	87.50 143	91.23 90	94.45 71	94.17 143	98.45 24	98.50 9	99.65 3	99.23 41

FMVSNet3		93.79 125	94.17 108	93.35 133	91.21 193	95.99 179	96.62 85	88.68 154	95.23 109	90.40 85	86.39 151	91.16 86	94.11 144	95.96 133	96.67 92	99.07 132	97.79 166

CHOSEN 1792x2688		92.66 145	92.49 154	92.85 136	97.13 69	98.89 62	95.90 123	88.50 158	95.32 104	83.31 161	71.99 246	88.96 106	94.10 145	96.69 94	96.49 98	98.15 205	99.10 57

0.4-1-1-0.2		89.32 191	87.66 216	91.26 156	86.11 244	93.97 237	94.54 160	85.98 187	87.83 221	91.12 68	86.40 150	78.02 196	94.06 146	84.03 257	87.73 240	95.75 246	95.62 226

UniMVSNet_ETH3D		88.47 204	86.00 236	91.35 154	91.55 187	96.29 172	92.53 194	88.81 153	85.58 246	82.33 165	67.63 259	66.87 257	94.04 147	91.49 225	95.24 142	98.84 164	98.92 83

dtuonly		90.46 173	91.17 178	89.63 180	91.72 185	95.69 194	94.51 163	87.20 171	90.71 196	73.98 221	81.33 194	86.42 122	94.02 148	94.30 179	93.91 183	96.36 236	95.83 219

SCA		90.92 165	93.04 142	88.45 194	93.72 165	97.33 143	92.77 189	76.08 254	96.02 80	78.26 191	91.96 81	90.86 89	93.99 149	90.98 229	90.04 230	95.88 242	94.06 241

EPMVS		90.88 166	92.12 164	89.44 184	94.71 124	97.24 145	93.55 175	76.81 248	95.89 85	81.77 168	91.49 89	86.47 121	93.87 150	90.21 232	90.07 229	95.92 241	93.49 249

COLMAP_ROB	90.49 14	93.27 137	92.71 148	93.93 119	97.75 61	97.44 138	96.07 112	93.17 62	95.40 100	83.86 158	83.76 181	88.72 107	93.87 150	94.25 181	94.11 177	98.87 160	95.28 229
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
dmvs_re		91.84 152	91.60 173	92.12 143	91.60 186	97.26 144	95.14 143	91.96 79	91.02 191	80.98 174	86.56 146	77.96 200	93.84 152	94.71 169	95.08 147	99.22 96	98.62 120

ACMH+	90.88 12	91.41 161	91.13 179	91.74 147	95.11 105	96.95 150	93.13 185	89.48 146	92.42 172	79.93 178	85.13 170	78.02 196	93.82 153	93.49 194	93.88 184	98.94 152	97.99 162

ACMH	90.77 13	91.51 160	91.63 172	91.38 152	95.62 89	96.87 153	91.76 213	89.66 142	91.58 185	78.67 186	86.73 140	78.12 194	93.77 154	94.59 171	94.54 168	98.78 173	98.98 77
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
CostFormer		90.69 167	90.48 187	90.93 159	94.18 155	96.08 177	94.03 169	78.20 244	93.47 151	89.96 99	90.97 96	80.30 184	93.72 155	87.66 243	88.75 234	95.51 250	96.12 213

MVSTER		94.89 73	95.07 83	94.68 91	94.71 124	96.68 161	97.00 65	90.57 128	95.18 114	93.05 42	95.21 52	86.41 123	93.72 155	97.59 61	95.88 123	99.00 144	98.50 133

USDC		90.69 167	90.52 186	90.88 160	94.17 156	96.43 168	95.82 131	86.76 175	93.92 140	76.27 204	86.49 148	74.30 225	93.67 157	95.04 165	93.36 194	98.61 187	94.13 238

Effi-MVS+-dtu		91.78 154	93.59 127	89.68 179	92.44 180	97.11 148	94.40 165	84.94 212	92.43 171	75.48 208	91.09 95	83.75 167	93.55 158	96.61 97	95.47 136	97.24 223	98.67 113

PatchmatchNet		90.56 170	92.49 154	88.31 197	93.83 163	96.86 154	92.42 197	76.50 251	95.96 82	78.31 190	91.96 81	89.66 98	93.48 159	90.04 234	89.20 233	95.32 251	93.73 247
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
TinyColmap		89.42 187	88.58 198	90.40 168	93.80 164	95.45 203	93.96 171	86.54 178	92.24 178	76.49 201	80.83 197	70.44 243	93.37 160	94.45 175	93.30 197	98.26 204	93.37 250

LTVRE_ROB	87.32 16	87.55 219	88.25 202	86.73 229	90.66 195	95.80 191	93.05 186	84.77 213	83.35 252	60.32 265	83.12 184	67.39 255	93.32 161	94.36 178	94.86 154	98.28 202	98.87 90
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
ADS-MVSNet		89.80 183	91.33 177	88.00 206	94.43 151	96.71 160	92.29 201	74.95 259	96.07 79	77.39 194	88.67 118	86.09 127	93.26 162	88.44 238	89.57 232	95.68 247	93.81 245

MDTV_nov1_ep13		91.57 158	93.18 139	89.70 177	93.39 168	96.97 149	93.53 176	80.91 238	95.70 93	81.86 167	92.40 76	89.93 96	93.25 163	91.97 221	90.80 224	95.25 255	94.46 233

UniMVSNet_NR-MVSNet		90.35 175	89.96 188	90.80 162	89.66 209	95.83 189	92.48 195	90.53 129	90.96 193	79.57 179	79.33 205	77.14 208	93.21 164	92.91 204	94.50 171	99.37 65	99.05 67

DU-MVS		89.67 185	88.84 196	90.63 165	89.26 220	95.61 196	92.48 195	89.91 134	91.22 188	79.57 179	77.72 215	71.18 239	93.21 164	92.53 210	94.57 165	99.35 68	99.05 67

pmmvs4		90.55 171	89.91 189	91.30 155	90.26 202	94.95 217	92.73 191	87.94 163	93.44 154	85.35 153	82.28 188	76.09 216	93.02 166	93.56 192	92.26 217	98.51 193	96.77 201

tpmrst		88.86 201	89.62 190	87.97 207	94.33 153	95.98 180	92.62 193	76.36 252	94.62 126	76.94 198	85.98 160	82.80 176	92.80 167	86.90 245	87.15 243	94.77 259	93.93 243

RPMNet		90.19 178	92.03 168	88.05 203	93.46 166	95.95 183	93.41 179	74.59 260	92.40 173	75.91 206	84.22 178	86.41 123	92.49 168	94.42 176	93.85 186	98.44 198	96.96 195

FMVSNet5		90.36 174	90.93 182	89.70 177	87.99 236	92.25 247	92.03 208	83.51 223	92.20 179	84.13 156	85.59 165	86.48 120	92.43 169	94.61 170	94.52 169	98.13 206	90.85 258

dps		90.11 181	89.37 194	90.98 158	93.89 161	96.21 174	93.49 178	77.61 246	91.95 181	92.74 49	88.85 113	78.77 192	92.37 170	87.71 242	87.71 241	95.80 245	94.38 234

Baseline_NR-MVSNet		89.27 193	88.01 207	90.73 164	89.26 220	93.71 239	92.71 192	89.78 140	90.73 194	81.28 172	73.53 237	72.85 231	92.30 171	92.53 210	93.84 187	99.07 132	98.88 88

CR-MVSNet		90.16 179	91.96 169	88.06 202	93.32 169	95.95 183	93.36 181	75.99 255	92.40 173	75.19 212	83.18 183	85.37 139	92.05 172	95.21 158	94.56 166	98.47 196	97.08 192

PatchT		89.13 196	91.71 170	86.11 235	92.92 173	95.59 198	83.64 257	75.09 258	91.87 182	75.19 212	82.63 186	85.06 149	92.05 172	95.21 158	94.56 166	97.76 214	97.08 192

v2v482		88.25 207	87.71 215	88.88 189	89.23 224	95.28 207	92.10 205	87.89 164	88.69 212	73.31 223	75.32 223	71.64 236	91.89 174	92.10 218	92.92 203	98.86 162	97.99 162

tfpnnormal		88.50 202	87.01 224	90.23 169	91.36 189	95.78 192	92.74 190	90.09 132	83.65 251	76.33 203	71.46 249	69.58 248	91.84 175	95.54 148	94.02 180	99.06 137	99.03 70

TranMVSNet+NR-MVSNet		89.23 194	88.48 200	90.11 175	89.07 226	95.25 210	92.91 188	90.43 130	90.31 200	77.10 197	76.62 220	71.57 237	91.83 176	92.12 216	94.59 164	99.32 71	98.92 83

EPNet		96.27 56	96.97 50	95.46 62	98.47 47	98.28 110	97.41 58	93.67 53	95.86 88	92.86 46	97.51 27	93.79 75	91.76 177	97.03 80	97.03 72	98.61 187	99.28 32
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
Fast-Effi-MVS+-dtu		91.19 162	93.64 124	88.33 196	92.19 182	96.46 167	93.99 170	81.52 236	92.59 168	71.82 228	92.17 78	85.54 136	91.68 178	95.73 143	94.64 160	98.80 170	98.34 148

usedtu_dtu_shiyan1		90.61 169	91.45 176	89.62 181	85.03 250	96.03 178	93.51 177	89.17 149	93.13 158	79.51 182	81.79 190	84.24 160	91.63 179	95.06 164	93.79 189	98.88 158	96.12 213

tpm		87.95 211	89.44 193	86.21 234	92.53 179	94.62 226	91.40 222	76.36 252	91.46 186	69.80 242	87.43 127	75.14 219	91.55 180	89.85 236	90.60 226	95.61 248	96.96 195

tpm cat1		88.90 199	87.78 213	90.22 170	93.88 162	95.39 205	93.79 172	78.11 245	92.55 169	89.43 109	81.31 195	79.84 188	91.40 181	84.95 253	86.34 247	94.68 261	94.09 239

baseline2		93.01 141	94.17 108	91.64 148	92.83 176	97.49 136	93.40 180	87.53 166	93.67 145	86.07 150	91.83 84	86.58 118	91.36 182	96.38 110	95.06 148	98.67 181	98.20 155

v10		88.00 209	87.96 208	88.05 203	89.44 214	94.68 223	92.36 198	83.35 224	89.37 206	72.96 224	73.98 234	72.79 232	91.35 183	93.59 189	92.88 204	98.81 168	98.42 141

usedtu_blend_shiyan5		87.98 210	86.70 230	89.47 182	77.63 260	92.14 252	94.53 161	85.67 193	86.74 234	91.16 63	86.06 157	77.89 201	91.22 184	85.19 249	82.63 256	96.58 231	96.25 207

blend_shiyan4		88.50 202	86.74 229	90.54 166	85.31 249	92.15 251	93.79 172	85.10 208	87.64 225	91.16 63	86.06 157	77.89 201	91.22 184	84.59 254	82.60 260	96.67 230	96.25 207

FE-MVSNET3		87.75 217	86.69 231	88.99 188	77.63 260	92.14 252	91.64 217	85.67 193	86.75 232	91.16 63	86.06 157	77.89 201	91.22 184	85.19 249	82.63 256	96.58 231	96.18 209

v1192		87.51 220	87.31 218	87.74 211	89.04 227	94.87 221	92.07 206	85.03 209	88.49 215	70.32 235	72.65 242	70.35 244	91.21 187	93.59 189	92.80 206	98.78 173	98.42 141

UniMVSNet (Re)		90.03 182	89.61 191	90.51 167	89.97 206	96.12 176	92.32 199	89.26 147	90.99 192	80.95 175	78.25 212	75.08 221	91.14 188	93.78 187	93.87 185	99.41 56	99.21 45

v1921920		87.31 224	87.13 222	87.52 217	88.87 230	94.72 222	91.96 211	84.59 217	88.28 216	69.86 241	72.50 243	70.03 247	91.10 189	93.33 196	92.61 211	98.71 177	98.44 136

v1144		87.92 214	87.79 212	88.07 200	89.27 219	95.15 212	92.17 204	85.62 197	88.52 214	71.52 229	73.80 235	72.40 234	91.06 190	93.54 193	92.80 206	98.81 168	98.33 149

MIMVSNet		88.99 198	91.07 180	86.57 231	86.78 242	95.62 195	91.20 227	75.40 257	90.65 197	76.57 200	84.05 179	82.44 178	91.01 191	95.84 137	95.38 138	98.48 195	93.50 248

test-LLR		91.62 157	93.56 128	89.35 186	93.31 170	96.57 164	92.02 209	87.06 173	92.34 176	75.05 215	90.20 101	88.64 108	90.93 192	96.19 124	94.07 178	97.75 215	96.90 198

TESTMET0.1,1		91.07 163	93.56 128	88.17 198	90.43 197	96.57 164	92.02 209	82.83 228	92.34 176	75.05 215	90.20 101	88.64 108	90.93 192	96.19 124	94.07 178	97.75 215	96.90 198

SixPastTwentyTwo		88.37 205	89.47 192	87.08 221	90.01 205	95.93 185	87.41 244	85.32 204	90.26 202	70.26 236	86.34 155	71.95 235	90.93 192	92.89 205	91.72 220	98.55 190	97.22 187

test-mter		90.95 164	93.54 130	87.93 208	90.28 201	96.80 155	91.44 221	82.68 229	92.15 180	74.37 219	89.57 108	88.23 113	90.88 195	96.37 112	94.31 174	97.93 212	97.37 183

PVSNet_Blended_VisFu		94.77 80	95.54 70	93.87 122	96.48 75	98.97 52	94.33 166	91.84 83	94.93 121	90.37 88	85.04 171	94.99 67	90.87 196	98.12 44	97.30 64	99.30 77	99.45 19

blended_shiyan8		86.10 234	85.44 240	86.88 225	77.65 259	92.22 248	91.69 214	85.52 200	86.88 230	78.82 185	78.06 214	76.43 215	90.85 197	85.36 248	82.97 254	96.74 228	96.14 212

wanda-best-256-512		86.03 236	85.37 241	86.79 226	77.63 260	92.14 252	91.64 217	85.67 193	86.75 232	78.43 187	78.36 209	76.66 213	90.81 198	85.19 249	82.63 256	96.58 231	95.88 216

FE-blended-shiyan7		86.03 236	85.37 241	86.79 226	77.63 260	92.14 252	91.64 217	85.67 193	86.74 234	78.43 187	78.36 209	76.66 213	90.81 198	85.19 249	82.63 256	96.58 231	95.88 216

blended_shiyan6		86.10 234	85.52 238	86.79 226	77.63 260	92.20 249	91.66 215	85.46 202	86.86 231	78.43 187	78.30 211	76.71 212	90.80 200	85.37 247	82.98 253	96.74 228	96.18 209

CP-MVSNet		87.89 215	87.27 219	88.62 192	89.30 218	95.06 213	90.60 232	85.78 191	87.43 228	75.98 205	74.60 228	68.14 254	90.76 201	93.07 202	93.60 191	99.30 77	98.98 77

v144192		87.40 222	87.20 221	87.64 212	88.89 228	94.88 220	91.65 216	84.70 215	87.80 222	71.17 233	73.20 240	70.91 240	90.75 202	92.69 206	92.49 212	98.71 177	98.43 139

pmmvs5		87.83 216	88.09 204	87.51 218	89.59 212	95.48 201	89.75 238	84.73 214	86.07 244	71.44 230	80.57 200	70.09 246	90.74 203	94.47 174	92.87 205	98.82 165	97.10 189

v8		88.21 208	87.94 210	88.51 193	89.62 210	95.01 215	92.31 200	84.99 210	88.94 207	74.70 217	75.03 224	73.51 229	90.67 204	92.11 217	92.74 209	98.80 170	98.24 153

v1240		86.89 226	86.75 228	87.06 222	88.75 232	94.65 225	91.30 226	84.05 219	87.49 227	68.94 245	71.96 247	68.86 252	90.65 205	93.33 196	92.72 210	98.67 181	98.24 153

gm-plane-assit		83.26 247	85.29 243	80.89 248	89.52 213	89.89 261	70.26 270	78.24 243	77.11 264	58.01 270	74.16 233	66.90 256	90.63 206	97.20 73	96.05 116	98.66 184	95.68 223

MS-PatchMatch		91.82 153	92.51 152	91.02 157	95.83 87	96.88 151	95.05 144	84.55 218	93.85 142	82.01 166	82.51 187	91.71 84	90.52 207	95.07 163	93.03 201	98.13 206	94.52 231

CDS-MVSNet		92.77 143	93.60 126	91.80 146	92.63 178	96.80 155	95.24 141	89.14 150	90.30 201	84.58 155	86.76 139	90.65 91	90.42 208	95.89 135	96.49 98	98.79 172	98.32 151
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
TAMVS		90.54 172	90.87 184	90.16 171	91.48 188	96.61 163	93.26 183	86.08 185	87.71 223	81.66 170	83.11 185	84.04 164	90.42 208	94.54 172	94.60 163	98.04 210	95.48 227

V42		88.31 206	87.95 209	88.73 191	89.44 214	95.34 206	92.23 203	87.21 170	88.83 209	74.49 218	74.89 226	73.43 230	90.41 210	92.08 219	92.77 208	98.60 189	98.33 149

anonymousdsp		88.90 199	91.00 181	86.44 232	88.74 233	95.97 181	90.40 234	82.86 227	88.77 211	67.33 248	81.18 196	81.44 181	90.22 211	96.23 119	94.27 175	99.12 123	99.16 52

PS-CasMVS		87.33 223	86.68 232	88.10 199	89.22 225	94.93 218	90.35 235	85.70 192	86.44 241	74.01 220	73.43 238	66.59 260	90.04 212	92.92 203	93.52 192	99.28 79	98.91 86

IterMVS-SCA-FT		90.24 176	92.48 156	87.63 213	92.85 175	94.30 233	93.79 172	81.47 237	92.66 165	69.95 239	84.66 175	88.38 111	89.99 213	95.39 154	94.34 173	97.74 217	97.63 175

IterMVS		90.20 177	92.43 158	87.61 214	92.82 177	94.31 232	94.11 168	81.54 235	92.97 160	69.90 240	84.71 174	88.16 114	89.96 214	95.25 157	94.17 176	97.31 222	97.46 180
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
gg-mvs-nofinetune		86.17 232	88.57 199	83.36 243	93.44 167	98.15 122	96.58 88	72.05 263	74.12 266	49.23 273	64.81 262	90.85 90	89.90 215	97.83 53	96.84 89	98.97 147	97.41 182

gbinet_0.2-2-1-0.02		86.23 231	85.66 237	86.89 224	78.33 257	92.17 250	91.62 220	85.96 189	86.51 240	79.33 183	78.13 213	77.66 206	89.55 216	85.60 246	82.66 255	96.56 235	96.87 200

GA-MVS		89.28 192	90.75 185	87.57 215	91.77 184	96.48 166	92.29 201	87.58 165	90.61 198	65.77 253	84.48 176	76.84 211	89.46 217	95.84 137	93.68 190	98.52 192	97.34 185

PEN-MVS		87.22 225	86.50 234	88.07 200	88.88 229	94.44 228	90.99 229	86.21 180	86.53 239	73.66 222	74.97 225	66.56 261	89.42 218	91.20 227	93.48 193	99.24 87	98.31 152

NR-MVSNet		89.34 190	88.66 197	90.13 174	90.40 198	95.61 196	93.04 187	89.91 134	91.22 188	78.96 184	77.72 215	68.90 251	89.16 219	94.24 182	93.95 181	99.32 71	98.99 75

pm-mvs1		89.19 195	89.02 195	89.38 185	90.40 198	95.74 193	92.05 207	88.10 162	86.13 242	77.70 192	73.72 236	79.44 189	88.97 220	95.81 139	94.51 170	99.08 130	97.78 171

MVS-HIRNet		85.36 240	86.89 225	83.57 242	90.13 203	94.51 227	83.57 258	72.61 262	88.27 217	71.22 232	68.97 254	81.81 179	88.91 221	93.08 201	91.94 218	94.97 258	89.64 261

PM-MVS		84.72 243	84.47 248	85.03 238	84.67 251	91.57 257	86.27 248	82.31 233	87.65 224	70.62 234	76.54 221	56.41 272	88.75 222	92.59 209	89.85 231	97.54 219	96.66 204

Vis-MVSNet (Re-imp)		94.46 90	96.24 61	92.40 139	95.23 101	98.64 80	95.56 137	90.99 123	94.42 131	85.02 154	90.88 97	94.65 70	88.01 223	98.17 41	98.37 18	99.57 14	98.53 130

v7n		86.43 229	86.52 233	86.33 233	87.91 237	94.93 218	90.15 236	83.05 225	86.57 238	70.21 237	71.48 248	66.78 258	87.72 224	94.19 184	92.96 202	98.92 154	98.76 105

pmmvs6		85.98 238	84.89 247	87.25 220	88.83 231	94.35 231	89.36 239	85.30 206	78.51 263	75.44 209	62.71 264	75.41 218	87.65 225	93.58 191	92.40 214	96.89 225	97.29 186

DTE-MVSNet		86.67 228	86.09 235	87.35 219	88.45 235	94.08 235	90.65 231	86.05 186	86.13 242	72.19 226	74.58 230	66.77 259	87.61 226	90.31 231	93.12 199	99.13 121	97.62 176

MDTV_nov1_ep13_2view		86.30 230	88.27 201	84.01 241	87.71 239	94.67 224	88.08 242	76.78 249	90.59 199	68.66 246	80.46 202	80.12 186	87.58 227	89.95 235	88.20 236	95.25 255	93.90 244

pmmvs-eth3d		84.33 244	82.94 250	85.96 237	84.16 252	90.94 258	86.55 247	83.79 220	84.25 249	75.85 207	70.64 251	56.43 271	87.44 228	92.20 215	90.41 228	97.97 211	95.68 223

v148		87.51 220	86.79 226	88.36 195	89.39 217	95.21 211	89.84 237	88.20 161	87.61 226	77.56 193	73.38 239	70.32 245	86.80 229	90.70 230	92.31 215	98.37 201	97.98 164

TransMVSNet (Re)		87.73 218	86.79 226	88.83 190	90.76 194	94.40 230	91.33 225	89.62 143	84.73 248	75.41 210	72.73 241	71.41 238	86.80 229	94.53 173	93.93 182	99.06 137	95.83 219

pmnet_mix02		86.12 233	87.12 223	84.96 239	89.82 207	94.12 234	84.88 253	86.63 177	91.78 183	65.60 254	80.76 198	76.98 209	86.61 231	87.29 244	84.80 250	96.21 237	94.09 239

WR-MVS_H		87.93 212	87.85 211	88.03 205	89.62 210	95.58 200	90.47 233	85.55 199	87.20 229	76.83 199	74.42 231	72.67 233	86.37 232	93.22 199	93.04 200	99.33 69	98.83 95

UGNet		94.92 71	96.63 55	92.93 135	96.03 84	98.63 82	94.53 161	91.52 106	96.23 72	90.03 96	92.87 72	96.10 62	86.28 233	96.68 95	96.60 95	99.16 117	99.32 30
Wanjuan Su, Qingshan Xu, Wenbing Tao: Uncertainty-guided Multi-view Stereo Network for Depth Estimation. IEEE Transactions on Circuits and Systems for Video Technology, 2022
UA-Net		93.96 112	95.95 65	91.64 148	96.06 83	98.59 85	95.29 140	90.00 133	91.06 190	82.87 162	90.64 98	98.06 43	86.06 234	98.14 43	98.20 23	99.58 11	96.96 195

test0.0.03 1		91.97 150	93.91 117	89.72 176	93.31 170	96.40 170	91.34 224	87.06 173	93.86 141	81.67 169	91.15 93	89.16 104	86.02 235	95.08 162	95.09 146	98.91 156	96.64 205

thisisatest0515		90.12 180	92.06 166	87.85 209	90.03 204	96.17 175	87.83 243	87.45 167	91.71 184	77.15 196	85.40 169	84.01 165	85.74 236	95.41 153	93.30 197	98.88 158	98.43 139

FC-MVSNet-test		91.63 156	93.82 122	89.08 187	92.02 183	96.40 170	93.26 183	87.26 169	93.72 144	77.26 195	88.61 119	89.86 97	85.50 237	95.72 145	95.02 150	99.16 117	97.44 181

CMPMVS	65.18 17	84.76 242	83.10 249	86.69 230	95.29 97	95.05 214	88.37 241	85.51 201	80.27 260	71.31 231	68.37 256	73.85 227	85.25 238	87.72 241	87.75 239	94.38 262	88.70 262
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
N_pmnet		84.80 241	85.10 246	84.45 240	89.25 223	92.86 242	84.04 255	86.21 180	88.78 210	66.73 251	72.41 244	74.87 224	85.21 239	88.32 240	86.45 244	95.30 253	92.04 255

PatchmatchNet1												75.08 221	85.19 240	88.36 239	86.44 245	95.32 251	92.09 254
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo. CVPR 2021
WR-MVS		87.93 212	88.09 204	87.75 210	89.26 220	95.28 207	90.81 230	86.69 176	88.90 208	75.29 211	74.31 232	73.72 228	85.19 240	92.26 213	93.32 196	99.27 82	98.81 98

TDRefinement		89.07 197	88.15 203	90.14 173	95.16 103	96.88 151	95.55 138	90.20 131	89.68 203	76.42 202	76.67 219	74.30 225	84.85 242	93.11 200	91.91 219	98.64 186	94.47 232

CVMVSNet		89.77 184	91.66 171	87.56 216	93.21 172	95.45 203	91.94 212	89.22 148	89.62 205	69.34 244	83.99 180	85.90 130	84.81 243	94.30 179	95.28 141	96.85 226	97.09 190

pmmvs3		79.16 253	80.12 256	78.05 254	79.36 256	86.59 266	78.13 267	73.87 261	76.42 265	57.51 271	70.59 252	57.02 270	84.66 244	90.10 233	88.32 235	94.75 260	91.77 257

Vis-MVSNet		92.77 143	95.00 85	90.16 171	94.10 157	98.79 66	94.76 155	88.26 159	92.37 175	79.95 177	88.19 123	91.58 85	84.38 245	97.59 61	97.58 53	99.52 22	98.91 86
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
EG-PatchMatch MVS		86.68 227	87.24 220	86.02 236	90.58 196	96.26 173	91.08 228	81.59 234	84.96 247	69.80 242	71.35 250	75.08 221	84.23 246	94.24 182	93.35 195	98.82 165	95.46 228

dtuonlycased		84.27 245	85.21 244	83.17 245	85.99 247	92.85 244	83.74 256	82.59 230	86.74 234	66.76 250	77.36 217	78.74 193	84.13 247	83.16 261	83.81 251	95.83 244	93.80 246

testgi		89.42 187	91.50 175	87.00 223	92.40 181	95.59 198	89.15 240	85.27 207	92.78 163	72.42 225	91.75 85	76.00 217	84.09 248	94.38 177	93.82 188	98.65 185	96.15 211

EPNet_dtu		92.45 148	95.02 84	89.46 183	98.02 56	95.47 202	94.79 153	92.62 69	94.97 120	70.11 238	94.76 59	92.61 82	84.07 249	95.94 134	95.56 133	97.15 224	95.82 221
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
FE-MVSNET2		81.81 249	81.15 252	82.57 246	75.40 267	92.39 246	86.04 249	83.61 222	81.61 257	68.16 247	55.75 268	59.22 269	83.77 250	93.31 198	91.54 222	98.45 197	94.24 236

MDA-MVSNet-bldmvs		80.11 251	80.24 255	79.94 250	77.01 265	93.21 240	78.86 266	85.94 190	82.71 255	60.86 262	79.71 204	51.77 275	83.71 251	75.60 265	86.37 246	93.28 264	92.35 252

new_pmnet		81.53 250	82.68 251	80.20 249	83.47 254	89.47 262	82.21 261	78.36 242	87.86 220	60.14 267	67.90 257	69.43 249	82.03 252	89.22 237	87.47 242	94.99 257	87.39 263

DeepMVS_CX						86.86 265	79.50 265	70.43 265	90.73 194	63.66 257	80.36 203	60.83 264	79.68 253	76.23 264		89.46 267	86.53 264

EU-MVSNet		85.62 239	87.65 217	83.24 244	88.54 234	92.77 245	87.12 245	85.32 204	86.71 237	64.54 256	78.52 208	75.11 220	78.35 254	92.25 214	92.28 216	95.58 249	95.93 215

IB-MVS	89.56 15	91.71 155	92.50 153	90.79 163	95.94 86	98.44 106	87.05 246	91.38 119	93.15 157	92.98 45	84.78 173	85.14 146	78.27 255	92.47 212	94.44 172	99.10 126	99.08 60
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
FE-MVSNET		79.15 254	80.25 254	77.87 255	69.65 270	89.30 263	81.34 263	82.42 232	79.49 262	59.18 269	59.18 265	59.41 268	77.03 256	91.12 228	90.65 225	97.57 218	92.63 251

MIMVSNet1		80.03 252	80.93 253	78.97 252	72.46 269	90.73 259	80.81 264	82.44 231	80.39 259	63.64 258	57.57 267	64.93 262	76.37 257	91.66 223	91.55 221	98.07 209	89.70 260

new-patchmatchnet		78.49 255	78.19 258	78.84 253	84.13 253	90.06 260	77.11 268	80.39 239	79.57 261	59.64 268	66.01 260	55.65 273	75.62 258	84.55 255	80.70 263	96.14 239	90.77 259

Anonymous20231206		83.84 246	85.19 245	82.26 247	87.38 240	92.87 241	85.49 251	83.65 221	86.07 244	63.44 260	68.42 255	69.01 250	75.45 259	93.34 195	92.44 213	98.12 208	94.20 237

usedtu_dtu_shiyan2		75.82 257	75.29 260	76.44 257	65.25 272	87.28 264	82.09 262	76.55 250	68.86 268	66.94 249	48.90 271	60.22 266	74.42 260	83.98 258	83.40 252	93.39 263	94.38 234

ambc			73.83 262		76.23 266	85.13 267	82.27 260		84.16 250	65.58 255	52.82 270	23.31 282	73.55 261	91.41 226	85.26 249	92.97 265	94.70 230

test_method		72.96 258	78.68 257	66.28 261	50.17 275	64.90 273	75.45 269	50.90 270	87.89 219	62.54 261	62.98 263	68.34 253	70.45 262	91.90 222	82.41 261	88.19 269	92.35 252

Gipuma		68.35 260	66.71 263	70.27 258	74.16 268	68.78 272	63.93 273	71.77 264	83.34 253	54.57 272	34.37 273	31.88 279	68.69 263	83.30 260	85.53 248	88.48 268	79.78 267
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
test20.03		82.92 248	85.52 238	79.90 251	87.75 238	91.84 256	82.80 259	82.99 226	82.65 256	60.32 265	78.90 207	70.50 241	67.10 264	92.05 220	90.89 223	98.44 198	91.80 256

FPMVS		75.84 256	74.59 261	77.29 256	86.92 241	83.89 268	85.01 252	80.05 240	82.91 254	60.61 264	65.25 261	60.41 265	63.86 265	75.60 265	73.60 267	87.29 270	80.47 266

EMVS		49.98 265	46.76 270	53.74 265	64.96 273	51.29 276	37.81 277	69.35 267	51.83 273	22.69 277	29.57 276	25.06 280	57.28 266	44.81 273	56.11 270	70.32 275	68.64 274

E-PMN		50.67 264	47.85 269	53.96 264	64.13 274	50.98 277	38.06 276	69.51 266	51.40 274	24.60 276	29.46 277	24.39 281	56.07 267	48.17 272	59.70 269	71.40 274	70.84 273

PMVS	63.12 18	67.27 261	66.39 264	68.30 259	77.98 258	60.24 274	59.53 274	76.82 247	66.65 269	60.74 263	54.39 269	59.82 267	51.24 268	73.92 268	70.52 268	83.48 271	79.17 268
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
tmp_tt				66.88 260	86.07 246	73.86 271	68.22 271	33.38 271	96.88 54	80.67 176	88.23 122	78.82 191	49.78 269	82.68 262	77.47 265	83.19 272

MVE	50.86 19	49.54 266	51.43 268	47.33 266	44.14 276	59.20 275	36.45 278	60.59 269	41.47 275	31.14 275	29.58 275	17.06 284	48.52 270	62.22 269	74.63 266	63.12 276	75.87 269
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
PMMVS2		64.36 263	65.94 265	62.52 262	67.37 271	77.44 270	64.39 272	69.32 268	61.47 271	34.59 274	46.09 272	41.03 278	48.02 271	74.56 267	78.23 264	91.43 266	82.76 265

MVS_clip		38.84 268	56.46 266	18.29 269	13.36 278	29.13 278	19.64 281	2.47 274	71.08 267	6.73 280	70.46 253	54.40 274	39.84 272	49.33 271	47.18 272	27.70 278	73.62 271

WB-MVS		69.22 259	76.91 259	60.24 263	85.80 248	79.37 269	56.86 275	84.96 211	81.50 258	18.16 278	76.85 218	61.07 263	34.23 273	82.46 263	81.81 262	81.43 273	75.31 270

VLMVS_CLIP		39.59 267	56.46 266	19.90 268	13.25 279	27.17 279	26.54 280	2.21 276	66.02 270	6.72 281	79.09 206	45.57 277	33.38 274	53.25 270	50.75 271	37.29 277	73.40 272

VLMVS		32.08 269	46.28 271	15.51 270	13.62 277	23.52 280	16.04 282	2.37 275	55.69 272	11.28 279	58.95 266	46.49 276	30.07 275	40.05 274	38.72 273	19.67 279	60.57 275

MVS_baseline		14.46 270	24.65 272	2.57 273	1.29 283	4.82 283	1.07 285	0.00 278	28.62 276	0.00 285	33.80 274	22.58 283	12.80 276	21.00 275	19.40 274	0.25 282	42.31 276

testmvs		12.09 271	16.94 273	6.42 271	3.15 280	6.08 281	9.51 283	3.84 272	21.46 277	5.31 282	27.49 278	6.76 285	10.89 277	17.06 276	15.01 275	5.84 280	24.75 277

test123		9.58 272	13.53 274	4.97 272	1.31 282	5.47 282	8.32 284	2.95 273	18.14 278	2.03 284	20.82 279	2.34 286	10.60 278	10.00 277	14.16 276	4.60 281	23.77 278

GG-mvs-BLEND		66.17 262	94.91 86	32.63 267	1.32 281	96.64 162	91.40 222	0.85 277	94.39 133	2.20 283	90.15 103	95.70 65	2.27 279	96.39 109	95.44 137	97.78 213	95.68 223

uanet_test		0.00 273	0.00 275	0.00 274	0.00 284	0.00 284	0.00 286	0.00 278	0.00 279	0.00 285	0.00 280	0.00 287	0.00 280	0.00 278	0.00 277	0.00 283	0.00 279

sosnet-low-res		0.00 273	0.00 275	0.00 274	0.00 284	0.00 284	0.00 286	0.00 278	0.00 279	0.00 285	0.00 280	0.00 287	0.00 280	0.00 278	0.00 277	0.00 283	0.00 279

sosnet		0.00 273	0.00 275	0.00 274	0.00 284	0.00 284	0.00 286	0.00 278	0.00 279	0.00 285	0.00 280	0.00 287	0.00 280	0.00 278	0.00 277	0.00 283	0.00 279

PatchmatchNet2					89.40 216	92.86 242	84.21 254
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo. CVPR 2021
PatchmatchNet3										66.35 252	72.11 245
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo. CVPR 2021
TestfortrainingZip							99.35 6	97.66 10		98.71 3						99.42 53

RE-MVS-def										63.50 259

9.14												99.28 14

SR-MVS					99.45 11			97.61 17				99.20 18

our_test_3					89.78 208	93.84 238	85.59 250

MTAPA										96.83 13		99.12 23

MTMP										97.18 8		98.83 28

Patchmatch-RL test							34.61 279

XVS					96.60 72	99.35 20	96.82 74			90.85 73		98.72 32				99.46 36

X-MVStestdata					96.60 72	99.35 20	96.82 74			90.85 73		98.72 32				99.46 36

mPP-MVS					99.21 25							98.29 41

NP-MVS									95.32 104

Patchmtry						95.96 182	93.36 181	75.99 255		75.19 212

High-resolution multi-view benchmark