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.

DVP-MVS++		98.92 1	99.18 1	98.61 4	99.47 5	99.61 2	99.39 3	97.82 1	98.80 1	96.86 8	98.90 2	99.92 1	98.67 17	99.02 2	98.20 19	99.43 45	99.82 1

MSP-MVS		98.73 6	98.93 5	98.50 6	99.44 11	99.57 4	99.36 4	97.65 9	98.14 12	96.51 14	98.49 7	99.65 8	98.67 17	98.60 14	98.42 11	99.40 51	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
SD-MVS		98.52 8	98.77 9	98.23 15	98.15 48	99.26 26	98.79 26	97.59 15	98.52 3	96.25 15	97.99 15	99.75 6	99.01 3	98.27 32	97.97 31	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
DVP-MVS		98.86 4	98.97 3	98.75 2	99.43 12	99.63 1	99.25 12	97.81 2	98.62 2	97.69 1	97.59 20	99.90 2	98.93 5	98.99 4	98.42 11	99.37 57	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
CSCG		97.44 32	97.18 43	97.75 27	99.47 5	99.52 8	98.55 31	95.41 40	97.69 23	95.72 19	94.29 54	95.53 62	98.10 33	96.20 107	97.38 55	99.24 76	99.62 4

SED-MVS		98.90 2	99.07 2	98.69 3	99.38 18	99.61 2	99.33 8	97.80 4	98.25 8	97.60 2	98.87 4	99.89 3	98.67 17	99.02 2	98.26 17	99.36 59	99.61 6

TSAR-MVS + MP.		98.49 9	98.78 8	98.15 19	98.14 49	99.17 33	99.34 6	97.18 29	98.44 5	95.72 19	97.84 16	99.28 12	98.87 7	99.05 1	98.05 26	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
TSAR-MVS + ACMM		97.71 28	98.60 12	96.66 39	98.64 40	99.05 37	98.85 25	97.23 27	98.45 4	89.40 87	97.51 24	99.27 14	96.88 59	98.53 15	97.81 41	98.96 120	99.59 8

APDe-MVS		98.87 3	98.96 4	98.77 1	99.58 2	99.53 7	99.44 1	97.81 2	98.22 10	97.33 4	98.70 5	99.33 10	98.86 8	98.96 6	98.40 13	99.63 5	99.57 9

DROMVSNet		96.49 49	97.63 34	95.16 64	94.75 110	98.69 70	97.39 54	88.97 118	96.34 57	92.02 51	96.04 37	96.46 50	98.21 26	98.41 24	97.96 32	99.61 6	99.55 10

SteuartSystems-ACMMP		98.38 14	98.71 10	97.99 23	99.34 20	99.46 10	99.34 6	97.33 24	97.31 35	94.25 29	98.06 13	99.17 19	98.13 31	98.98 5	98.46 9	99.55 18	99.54 11
Skip Steuart: Steuart Systems R&D Blog.
CS-MVS		96.87 43	97.41 39	96.24 45	97.42 59	99.48 9	97.30 55	91.83 80	97.17 39	93.02 40	94.80 51	94.45 66	98.16 30	98.61 13	97.85 39	99.69 1	99.50 12

canonicalmvs		95.25 65	95.45 69	95.00 68	95.27 94	98.72 68	96.89 64	89.82 106	96.51 54	90.84 64	93.72 57	86.01 118	97.66 41	95.78 119	97.94 34	99.54 19	99.50 12

DPE-MVS		98.75 5	98.91 6	98.57 5	99.21 23	99.54 6	99.42 2	97.78 6	97.49 31	96.84 9	98.94 1	99.82 5	98.59 21	98.90 10	98.22 18	99.56 17	99.48 14
Kehua Chen, Zhenlong Yuan, Tianlu Mao, Zhaoqi Wang: Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization. AAAI2025
EPP-MVSNet		95.27 64	96.18 60	94.20 87	94.88 104	98.64 75	94.97 113	90.70 95	95.34 86	89.67 83	91.66 78	93.84 69	95.42 88	97.32 64	97.00 65	99.58 11	99.47 15

CS-MVS-test		97.00 39	97.85 33	96.00 50	97.77 54	99.56 5	96.35 85	91.95 75	97.54 29	92.20 48	96.14 35	96.00 60	98.19 28	98.46 19	97.78 42	99.57 14	99.45 16

PVSNet_Blended_VisFu		94.77 73	95.54 67	93.87 91	96.48 71	98.97 49	94.33 126	91.84 78	94.93 99	90.37 72	85.04 133	94.99 63	90.87 154	98.12 41	97.30 58	99.30 68	99.45 16

SMA-MVS		98.66 7	98.89 7	98.39 9	99.60 1	99.41 12	99.00 20	97.63 12	97.78 18	95.83 18	98.33 11	99.83 4	98.85 9	98.93 8	98.56 6	99.41 48	99.40 18
Yufeng Yin; Xiaoyan Liu; Zichao Zhang: SMA-MVS: Segmentation-Guided Multi-Scale Anchor Deformation Patch Multi-View Stereo. IEEE Transactions on Circuits and Systems for Video Technology
MVS_0304		96.31 51	96.91 49	95.62 55	97.21 64	99.20 31	98.55 31	93.10 59	97.04 45	89.73 81	90.30 93	96.35 52	95.71 78	98.14 39	97.93 36	99.38 54	99.40 18

MSLP-MVS++		98.04 23	97.93 32	98.18 16	99.10 27	99.09 36	98.34 36	96.99 32	97.54 29	96.60 12	94.82 50	98.45 34	98.89 6	97.46 61	98.77 4	99.17 91	99.37 20

DeepC-MVS	94.87 4	96.76 48	96.50 54	97.05 35	98.21 47	99.28 24	98.67 27	97.38 20	97.31 35	90.36 73	89.19 101	93.58 71	98.19 28	98.31 27	98.50 7	99.51 25	99.36 21
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
CANet		96.84 45	97.20 41	96.42 40	97.92 52	99.24 30	98.60 29	93.51 51	97.11 42	93.07 36	91.16 83	97.24 44	96.21 72	98.24 35	98.05 26	99.22 82	99.35 22

3Dnovator+	93.91 7	97.23 35	97.22 40	97.24 32	98.89 35	98.85 61	98.26 38	93.25 56	97.99 15	95.56 22	90.01 97	98.03 40	98.05 34	97.91 47	98.43 10	99.44 42	99.35 22

DCV-MVSNet		94.76 74	95.12 77	94.35 85	95.10 100	95.81 155	96.46 82	89.49 112	96.33 58	90.16 74	92.55 67	90.26 89	95.83 77	95.52 125	96.03 90	99.06 109	99.33 24

TSAR-MVS + GP.		97.45 31	98.36 19	96.39 41	95.56 86	98.93 53	97.74 48	93.31 53	97.61 27	94.24 30	98.44 9	99.19 17	98.03 35	97.60 56	97.41 53	99.44 42	99.33 24

UGNet		94.92 66	96.63 52	92.93 104	96.03 80	98.63 77	94.53 123	91.52 86	96.23 60	90.03 76	92.87 64	96.10 58	86.28 186	96.68 85	96.60 75	99.16 94	99.32 26
Wanjuan Su, Qingshan Xu, Wenbing Tao: Uncertainty-guided Multi-view Stereo Network for Depth Estimation. IEEE Transactions on Circuits and Systems for Video Technology, 2022
MCST-MVS		98.20 18	98.36 19	98.01 22	99.40 14	99.05 37	99.00 20	97.62 13	97.59 28	93.70 33	97.42 27	99.30 11	98.77 13	98.39 26	97.48 50	99.59 7	99.31 27

FC-MVSNet-train		93.85 96	93.91 98	93.78 93	94.94 103	96.79 124	94.29 127	91.13 90	93.84 118	88.26 99	90.40 92	85.23 125	94.65 101	96.54 91	95.31 111	99.38 54	99.28 28

X-MVS		97.84 24	98.19 27	97.42 30	99.40 14	99.35 17	99.06 17	97.25 25	97.38 34	90.85 61	96.06 36	98.72 29	98.53 24	98.41 24	98.15 22	99.46 32	99.28 28

EPNet		96.27 53	96.97 46	95.46 59	98.47 42	98.28 86	97.41 52	93.67 49	95.86 74	92.86 42	97.51 24	93.79 70	91.76 139	97.03 74	97.03 64	98.61 153	99.28 28
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
CP-MVS		98.32 17	98.34 22	98.29 12	99.34 20	99.30 22	99.15 14	97.35 21	97.49 31	95.58 21	97.72 18	98.62 33	98.82 11	98.29 28	97.67 45	99.51 25	99.28 28

DELS-MVS		96.06 54	96.04 61	96.07 49	97.77 54	99.25 28	98.10 41	93.26 54	94.42 107	92.79 43	88.52 108	93.48 72	95.06 93	98.51 16	98.83 1	99.45 36	99.28 28
Christian Sormann, Emanuele Santellani, Mattia Rossi, Andreas Kuhn, Friedrich Fraundorfer: DELS-MVS: Deep Epipolar Line Search for Multi-View Stereo. Winter Conference on Applications of Computer Vision (WACV), 2023
HFP-MVS		98.48 10	98.62 11	98.32 11	99.39 17	99.33 21	99.27 10	97.42 18	98.27 7	95.25 23	98.34 10	98.83 26	99.08 1	98.26 33	98.08 25	99.48 28	99.26 33

HPM-MVS++		98.34 16	98.47 15	98.18 16	99.46 8	99.15 34	99.10 16	97.69 8	97.67 24	94.93 26	97.62 19	99.70 7	98.60 20	98.45 20	97.46 51	99.31 66	99.26 33

3Dnovator	93.79 8	97.08 37	97.20 41	96.95 37	99.09 28	99.03 43	98.20 39	93.33 52	97.99 15	93.82 32	90.61 91	96.80 48	97.82 37	97.90 48	98.78 3	99.47 31	99.26 33

Anonymous20231211		93.49 105	92.33 132	94.84 75	94.78 109	98.00 96	96.11 90	91.85 77	94.86 100	90.91 60	74.69 177	89.18 98	96.73 62	94.82 140	95.51 106	98.67 147	99.24 36

MP-MVS		98.09 22	98.30 24	97.84 26	99.34 20	99.19 32	99.23 13	97.40 19	97.09 43	93.03 39	97.58 22	98.85 25	98.57 23	98.44 22	97.69 44	99.48 28	99.23 37
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
IS_MVSNet		95.28 63	96.43 56	93.94 89	95.30 92	99.01 47	95.90 97	91.12 91	94.13 113	87.50 104	91.23 82	94.45 66	94.17 108	98.45 20	98.50 7	99.65 3	99.23 37

ACMMPR		98.40 12	98.49 13	98.28 13	99.41 13	99.40 13	99.36 4	97.35 21	98.30 6	95.02 25	97.79 17	98.39 36	99.04 2	98.26 33	98.10 23	99.50 27	99.22 39

APD-MVS		98.36 15	98.32 23	98.41 8	99.47 5	99.26 26	99.12 15	97.77 7	96.73 50	96.12 16	97.27 28	98.88 24	98.46 25	98.47 18	98.39 14	99.52 20	99.22 39
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
UniMVSNet (Re)		90.03 148	89.61 157	90.51 131	89.97 171	96.12 142	92.32 157	89.26 114	90.99 158	80.95 134	78.25 166	75.08 173	91.14 146	93.78 155	93.87 152	99.41 48	99.21 41

DeepPCF-MVS	95.28 2	97.00 39	98.35 21	95.42 60	97.30 62	98.94 51	94.82 117	96.03 38	98.24 9	92.11 50	95.80 40	98.64 32	95.51 85	98.95 7	98.66 5	96.78 189	99.20 42

SF-MVS		98.39 13	98.45 17	98.33 10	99.45 9	99.05 37	98.27 37	97.65 9	97.73 19	97.02 7	98.18 12	99.25 15	98.11 32	98.15 38	97.62 46	99.45 36	99.19 43

ACMMP_NAP		98.20 18	98.49 13	97.85 25	99.50 4	99.40 13	99.26 11	97.64 11	97.47 33	92.62 46	97.59 20	99.09 22	98.71 15	98.82 12	97.86 38	99.40 51	99.19 43

train_agg		97.65 29	98.06 29	97.18 33	98.94 32	98.91 56	98.98 24	97.07 31	96.71 51	90.66 66	97.43 26	99.08 23	98.20 27	97.96 46	97.14 62	99.22 82	99.19 43

ETV-MVS		96.31 51	97.47 38	94.96 70	94.79 107	98.78 64	96.08 91	91.41 88	96.16 62	90.50 68	95.76 41	96.20 56	97.39 45	98.42 23	97.82 40	99.57 14	99.18 46

QAPM		96.78 47	97.14 44	96.36 42	99.05 29	99.14 35	98.02 42	93.26 54	97.27 37	90.84 64	91.16 83	97.31 43	97.64 42	97.70 54	98.20 19	99.33 61	99.18 46

anonymousdsp		88.90 162	91.00 147	86.44 187	88.74 197	95.97 146	90.40 186	82.86 179	88.77 176	67.33 199	81.18 154	81.44 147	90.22 165	96.23 104	94.27 143	99.12 100	99.16 48

EIA-MVS		95.50 56	96.19 59	94.69 79	94.83 106	98.88 60	95.93 96	91.50 87	94.47 106	89.43 85	93.14 60	92.72 76	97.05 55	97.82 52	97.13 63	99.43 45	99.15 49

Anonymous202405211			92.18 133		95.04 101	98.20 90	96.14 89	91.79 82	93.93 114		74.60 178	88.38 106	96.48 68	95.17 135	95.82 100	99.00 115	99.15 49

ACMMP		97.37 33	97.48 37	97.25 31	98.88 36	99.28 24	98.47 34	96.86 34	97.04 45	92.15 49	97.57 23	96.05 59	97.67 40	97.27 65	95.99 92	99.46 32	99.14 51
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
CHOSEN 1792x2688		92.66 114	92.49 124	92.85 105	97.13 65	98.89 59	95.90 97	88.50 124	95.32 87	83.31 121	71.99 195	88.96 101	94.10 110	96.69 84	96.49 76	98.15 170	99.10 52

DeepC-MVS_fast	96.13 1	98.13 20	98.27 25	97.97 24	99.16 26	99.03 43	99.05 18	97.24 26	98.22 10	94.17 31	95.82 39	98.07 38	98.69 16	98.83 11	98.80 2	99.52 20	99.10 52
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
PGM-MVS		97.81 25	98.11 28	97.46 29	99.55 3	99.34 20	99.32 9	94.51 45	96.21 61	93.07 36	98.05 14	97.95 41	98.82 11	98.22 36	97.89 37	99.48 28	99.09 54

CNVR-MVS		98.47 11	98.46 16	98.48 7	99.40 14	99.05 37	99.02 19	97.54 16	97.73 19	96.65 11	97.20 29	99.13 20	98.85 9	98.91 9	98.10 23	99.41 48	99.08 55

PVSNet_BlendedMVS		95.41 61	95.28 71	95.57 56	97.42 59	99.02 45	95.89 99	93.10 59	96.16 62	93.12 34	91.99 71	85.27 123	94.66 99	98.09 43	97.34 56	99.24 76	99.08 55

PVSNet_Blended		95.41 61	95.28 71	95.57 56	97.42 59	99.02 45	95.89 99	93.10 59	96.16 62	93.12 34	91.99 71	85.27 123	94.66 99	98.09 43	97.34 56	99.24 76	99.08 55

IB-MVS	89.56 15	91.71 123	92.50 123	90.79 128	95.94 82	98.44 83	87.05 198	91.38 89	93.15 126	92.98 41	84.78 134	85.14 126	78.27 205	92.47 177	94.44 140	99.10 102	99.08 55
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
GeoE		92.52 116	92.64 119	92.39 109	93.96 126	97.76 100	96.01 95	85.60 155	93.23 125	83.94 117	81.56 151	84.80 129	95.63 81	96.22 105	95.83 99	99.19 89	99.07 59

ACMP	92.88 9	94.43 82	94.38 87	94.50 82	96.01 81	97.69 101	95.85 102	92.09 73	95.74 77	89.12 93	95.14 47	82.62 143	94.77 95	95.73 121	94.67 128	99.14 97	99.06 60
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
CLD-MVS		94.79 71	94.36 88	95.30 62	95.21 96	97.46 106	97.23 56	92.24 72	96.43 55	91.77 53	92.69 65	84.31 131	96.06 73	95.52 125	95.03 119	99.31 66	99.06 60
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
UniMVSNet_NR-MVSNet		90.35 141	89.96 154	90.80 127	89.66 174	95.83 154	92.48 153	90.53 98	90.96 159	79.57 138	79.33 163	77.14 164	93.21 126	92.91 171	94.50 139	99.37 57	99.05 62

DU-MVS		89.67 151	88.84 162	90.63 130	89.26 184	95.61 160	92.48 153	89.91 103	91.22 155	79.57 138	77.72 167	71.18 190	93.21 126	92.53 175	94.57 133	99.35 60	99.05 62

CPTT-MVS		97.78 26	97.54 35	98.05 21	98.91 34	99.05 37	99.00 20	96.96 33	97.14 41	95.92 17	95.50 43	98.78 28	98.99 4	97.20 67	96.07 87	98.54 157	99.04 64

tfpnnormal		88.50 165	87.01 187	90.23 133	91.36 154	95.78 157	92.74 148	90.09 101	83.65 204	76.33 156	71.46 198	69.58 199	91.84 137	95.54 124	94.02 148	99.06 109	99.03 65

LGP-MVS_train		94.12 89	94.62 82	93.53 96	96.44 72	97.54 103	97.40 53	91.84 78	94.66 102	81.09 133	95.70 42	83.36 139	95.10 92	96.36 100	95.71 101	99.32 63	99.03 65

PHI-MVS		97.78 26	98.44 18	97.02 36	98.73 37	99.25 28	98.11 40	95.54 39	96.66 53	92.79 43	98.52 6	99.38 9	97.50 44	97.84 49	98.39 14	99.45 36	99.03 65

MVS_111021_HR		97.04 38	98.20 26	95.69 54	98.44 44	99.29 23	96.59 77	93.20 57	97.70 22	89.94 79	98.46 8	96.89 46	96.71 63	98.11 42	97.95 33	99.27 72	99.01 68

HQP-MVS		94.43 82	94.57 83	94.27 86	96.41 73	97.23 112	96.89 64	93.98 47	95.94 71	83.68 119	95.01 49	84.46 130	95.58 83	95.47 127	94.85 127	99.07 106	99.00 69

NR-MVSNet		89.34 154	88.66 163	90.13 138	90.40 163	95.61 160	93.04 145	89.91 103	91.22 155	78.96 141	77.72 167	68.90 202	89.16 172	94.24 152	93.95 149	99.32 63	98.99 70

MAR-MVS		95.50 56	95.60 65	95.39 61	98.67 39	98.18 92	95.89 99	89.81 107	94.55 105	91.97 52	92.99 61	90.21 90	97.30 47	96.79 80	97.49 49	98.72 143	98.99 70
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
Effi-MVS+		92.93 111	93.86 100	91.86 112	94.07 125	98.09 95	95.59 104	85.98 150	94.27 111	79.54 140	91.12 86	81.81 145	96.71 63	96.67 86	96.06 88	99.27 72	98.98 72

CP-MVSNet		87.89 176	87.27 182	88.62 152	89.30 182	95.06 177	90.60 184	85.78 152	87.43 189	75.98 158	74.60 178	68.14 205	90.76 155	93.07 169	93.60 157	99.30 68	98.98 72

NCCC		98.10 21	98.05 30	98.17 18	99.38 18	99.05 37	99.00 20	97.53 17	98.04 14	95.12 24	94.80 51	99.18 18	98.58 22	98.49 17	97.78 42	99.39 53	98.98 72

ACMH	90.77 13	91.51 128	91.63 141	91.38 119	95.62 85	96.87 119	91.76 171	89.66 109	91.58 152	78.67 142	86.73 116	78.12 158	93.77 116	94.59 142	94.54 136	98.78 140	98.98 72
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
baseline		94.83 68	95.82 63	93.68 94	94.75 110	97.80 99	96.51 80	88.53 123	97.02 47	89.34 89	92.93 62	92.18 78	94.69 98	95.78 119	96.08 86	98.27 168	98.97 76

HyFIR lowres test		92.03 118	91.55 142	92.58 106	97.13 65	98.72 68	94.65 121	86.54 143	93.58 122	82.56 124	67.75 206	90.47 88	95.67 79	95.87 115	95.54 105	98.91 125	98.93 77

UniMVSNet_ETH3D		88.47 166	86.00 196	91.35 120	91.55 152	96.29 138	92.53 152	88.81 119	85.58 199	82.33 125	67.63 207	66.87 208	94.04 111	91.49 190	95.24 113	98.84 131	98.92 78

CDPH-MVS		96.84 45	97.49 36	96.09 47	98.92 33	98.85 61	98.61 28	95.09 41	96.00 69	87.29 105	95.45 45	97.42 42	97.16 50	97.83 50	97.94 34	99.44 42	98.92 78

TranMVSNet+NR-MVSNet		89.23 157	88.48 166	90.11 139	89.07 190	95.25 174	92.91 146	90.43 99	90.31 165	77.10 150	76.62 170	71.57 188	91.83 138	92.12 181	94.59 132	99.32 63	98.92 78

PS-CasMVS		87.33 183	86.68 192	88.10 159	89.22 189	94.93 182	90.35 187	85.70 153	86.44 194	74.01 173	73.43 188	66.59 211	90.04 166	92.92 170	93.52 158	99.28 70	98.91 81

Vis-MVSNet		92.77 112	95.00 80	90.16 135	94.10 124	98.79 63	94.76 119	88.26 125	92.37 142	79.95 136	88.19 110	91.58 80	84.38 197	97.59 57	97.58 48	99.52 20	98.91 81
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
Baseline_NR-MVSNet		89.27 156	88.01 172	90.73 129	89.26 184	93.71 200	92.71 150	89.78 108	90.73 160	81.28 132	73.53 187	72.85 182	92.30 133	92.53 175	93.84 154	99.07 106	98.88 83

OpenMVS	92.33 11	95.50 56	95.22 73	95.82 53	98.98 30	98.97 49	97.67 49	93.04 62	94.64 103	89.18 92	84.44 138	94.79 64	96.79 60	97.23 66	97.61 47	99.24 76	98.88 83

tttt0517		94.52 80	95.44 70	93.44 99	94.51 117	98.68 71	94.61 122	90.72 93	95.61 82	86.84 109	93.78 56	89.26 97	94.74 96	97.02 75	94.86 124	99.20 88	98.87 85

LTVRE_ROB	87.32 16	87.55 179	88.25 168	86.73 184	90.66 160	95.80 156	93.05 144	84.77 166	83.35 205	60.32 213	83.12 145	67.39 206	93.32 123	94.36 149	94.86 124	98.28 167	98.87 85
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
thisisatest0530		94.54 79	95.47 68	93.46 98	94.51 117	98.65 74	94.66 120	90.72 93	95.69 80	86.90 108	93.80 55	89.44 94	94.74 96	96.98 76	94.86 124	99.19 89	98.85 87

MVS_111021_LR		97.16 36	98.01 31	96.16 46	98.47 42	98.98 48	96.94 63	93.89 48	97.64 26	91.44 54	98.89 3	96.41 51	97.20 49	98.02 45	97.29 60	99.04 114	98.85 87

WR-MVS_H		87.93 173	87.85 176	88.03 165	89.62 175	95.58 164	90.47 185	85.55 156	87.20 190	76.83 152	74.42 181	72.67 184	86.37 185	93.22 166	93.04 166	99.33 61	98.83 89

casdiffmvs_mvg		94.55 78	94.26 90	94.88 72	94.96 102	98.51 81	97.11 57	91.82 81	94.28 110	89.20 91	86.60 119	86.85 111	96.56 67	97.47 60	97.25 61	99.64 4	98.83 89
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
WR-MVS		87.93 173	88.09 170	87.75 170	89.26 184	95.28 171	90.81 182	86.69 141	88.90 173	75.29 164	74.31 182	73.72 179	85.19 193	92.26 178	93.32 162	99.27 72	98.81 91

diffmvs		94.31 87	94.21 91	94.42 84	94.64 115	98.28 86	96.36 84	91.56 84	96.77 49	88.89 95	88.97 102	84.23 132	96.01 76	96.05 111	96.41 78	99.05 113	98.79 92
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
DI_MVS_plusplus_trai		94.01 91	93.63 105	94.44 83	94.54 116	98.26 88	97.51 51	90.63 96	95.88 73	89.34 89	80.54 159	89.36 95	95.48 86	96.33 101	96.27 82	99.17 91	98.78 93

v7n		86.43 189	86.52 193	86.33 188	87.91 201	94.93 182	90.15 188	83.05 177	86.57 192	70.21 189	71.48 197	66.78 209	87.72 177	94.19 154	92.96 168	98.92 124	98.76 94

Effi-MVS+-dtu		91.78 122	93.59 107	89.68 143	92.44 147	97.11 114	94.40 125	84.94 165	92.43 138	75.48 161	91.09 87	83.75 136	93.55 120	96.61 87	95.47 107	97.24 185	98.67 95

casdiffmvs		94.38 85	94.15 96	94.64 81	94.70 114	98.51 81	96.03 94	91.66 83	95.70 78	89.36 88	86.48 122	85.03 128	96.60 66	97.40 62	97.30 58	99.52 20	98.67 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
thres600view7		93.49 105	92.37 131	94.79 77	95.42 87	98.93 53	96.58 78	92.31 68	93.04 127	87.88 101	86.62 118	76.94 166	97.09 54	96.82 77	95.63 102	99.45 36	98.63 97

IterMVS-LS		92.56 115	93.18 112	91.84 113	93.90 127	94.97 180	94.99 112	86.20 147	94.18 112	82.68 123	85.81 129	87.36 110	94.43 103	95.31 131	96.02 91	98.87 128	98.60 98
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
tfpn200view9		93.64 100	92.57 120	94.89 71	95.33 90	98.94 51	96.82 67	92.31 68	92.63 133	88.29 96	87.21 112	78.01 160	97.12 53	96.82 77	95.85 97	99.45 36	98.56 99

thres400		93.56 103	92.43 128	94.87 74	95.40 88	98.91 56	96.70 74	92.38 67	92.93 129	88.19 100	86.69 117	77.35 163	97.13 51	96.75 82	95.85 97	99.42 47	98.56 99

MVS_Test		94.82 69	95.66 64	93.84 92	94.79 107	98.35 85	96.49 81	89.10 117	96.12 65	87.09 107	92.58 66	90.61 87	96.48 68	96.51 95	96.89 69	99.11 101	98.54 101

Vis-MVSNet (Re-imp)		94.46 81	96.24 58	92.40 108	95.23 95	98.64 75	95.56 105	90.99 92	94.42 107	85.02 114	90.88 89	94.65 65	88.01 176	98.17 37	98.37 16	99.57 14	98.53 102

Fast-Effi-MVS+		91.87 120	92.08 135	91.62 118	92.91 141	97.21 113	94.93 114	84.60 169	93.61 121	81.49 131	83.50 143	78.95 155	96.62 65	96.55 90	96.22 84	99.16 94	98.51 103

MVSTER		94.89 67	95.07 78	94.68 80	94.71 112	96.68 127	97.00 59	90.57 97	95.18 95	93.05 38	95.21 46	86.41 115	93.72 117	97.59 57	95.88 96	99.00 115	98.50 104

thres200		93.62 101	92.54 121	94.88 72	95.36 89	98.93 53	96.75 72	92.31 68	92.84 130	88.28 98	86.99 114	77.81 162	97.13 51	96.82 77	95.92 93	99.45 36	98.49 105

v1921920		87.31 184	87.13 185	87.52 177	88.87 194	94.72 186	91.96 169	84.59 170	88.28 180	69.86 193	72.50 193	70.03 198	91.10 147	93.33 164	92.61 177	98.71 144	98.44 106

thisisatest0515		90.12 146	92.06 136	87.85 169	90.03 169	96.17 141	87.83 195	87.45 133	91.71 151	77.15 149	85.40 131	84.01 134	85.74 189	95.41 129	93.30 163	98.88 127	98.43 107

v144192		87.40 182	87.20 184	87.64 172	88.89 192	94.88 184	91.65 172	84.70 168	87.80 184	71.17 185	73.20 190	70.91 191	90.75 156	92.69 173	92.49 178	98.71 144	98.43 107

v1192		87.51 180	87.31 181	87.74 171	89.04 191	94.87 185	92.07 164	85.03 163	88.49 179	70.32 187	72.65 192	70.35 195	91.21 145	93.59 157	92.80 172	98.78 140	98.42 109

v10		88.00 171	87.96 173	88.05 163	89.44 179	94.68 187	92.36 156	83.35 176	89.37 171	72.96 176	73.98 184	72.79 183	91.35 144	93.59 157	92.88 170	98.81 135	98.42 109

thres100view900		93.55 104	92.47 127	94.81 76	95.33 90	98.74 66	96.78 71	92.30 71	92.63 133	88.29 96	87.21 112	78.01 160	96.78 61	96.38 97	95.92 93	99.38 54	98.40 111

AdaColmap		97.53 30	96.93 47	98.24 14	99.21 23	98.77 65	98.47 34	97.34 23	96.68 52	96.52 13	95.11 48	96.12 57	98.72 14	97.19 69	96.24 83	99.17 91	98.39 112

FA-MVS(training)		93.94 93	95.16 74	92.53 107	94.87 105	98.57 80	95.42 107	79.49 191	95.37 85	90.98 59	86.54 120	94.26 68	95.44 87	97.80 53	95.19 116	98.97 118	98.38 113

PCF-MVS	93.95 6	95.65 55	95.14 75	96.25 43	97.73 57	98.73 67	97.59 50	97.13 30	92.50 137	89.09 94	89.85 98	96.65 49	96.90 58	94.97 139	94.89 123	99.08 104	98.38 113
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
Fast-Effi-MVS+-dtu		91.19 130	93.64 104	88.33 156	92.19 149	96.46 133	93.99 130	81.52 186	92.59 135	71.82 180	92.17 70	85.54 121	91.68 140	95.73 121	94.64 130	98.80 137	98.34 115

v1144		87.92 175	87.79 177	88.07 160	89.27 183	95.15 176	92.17 162	85.62 154	88.52 178	71.52 181	73.80 185	72.40 185	91.06 148	93.54 161	92.80 172	98.81 135	98.33 116

V42		88.31 168	87.95 174	88.73 151	89.44 179	95.34 170	92.23 161	87.21 136	88.83 174	74.49 171	74.89 176	73.43 181	90.41 164	92.08 184	92.77 174	98.60 155	98.33 116

CDS-MVSNet		92.77 112	93.60 106	91.80 114	92.63 145	96.80 121	95.24 109	89.14 116	90.30 166	84.58 115	86.76 115	90.65 86	90.42 162	95.89 114	96.49 76	98.79 139	98.32 118
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
PEN-MVS		87.22 185	86.50 194	88.07 160	88.88 193	94.44 192	90.99 181	86.21 145	86.53 193	73.66 174	74.97 175	66.56 212	89.42 171	91.20 192	93.48 159	99.24 76	98.31 119

v1240		86.89 186	86.75 191	87.06 182	88.75 196	94.65 189	91.30 178	84.05 172	87.49 188	68.94 197	71.96 196	68.86 203	90.65 159	93.33 164	92.72 176	98.67 147	98.24 120

v8		88.21 170	87.94 175	88.51 153	89.62 175	95.01 179	92.31 158	84.99 164	88.94 172	74.70 170	75.03 174	73.51 180	90.67 158	92.11 182	92.74 175	98.80 137	98.24 120

baseline2		93.01 110	94.17 94	91.64 116	92.83 143	97.49 105	93.40 138	87.53 132	93.67 120	86.07 110	91.83 76	86.58 112	91.36 143	96.38 97	95.06 118	98.67 147	98.20 122

ET-MVSNet_ETH3D		93.34 107	94.33 89	92.18 111	83.26 212	97.66 102	96.72 73	89.89 105	95.62 81	87.17 106	96.00 38	83.69 137	96.99 56	93.78 155	95.34 110	99.06 109	98.18 123

CNLPA		96.90 42	96.28 57	97.64 28	98.56 41	98.63 77	96.85 66	96.60 36	97.73 19	97.08 6	89.78 99	96.28 55	97.80 39	96.73 83	96.63 74	98.94 122	98.14 124

test1111		93.94 93	92.78 117	95.29 63	96.14 78	99.42 11	96.79 70	92.85 63	95.08 97	91.39 56	80.69 157	79.86 152	95.00 94	98.28 31	98.00 28	99.58 11	98.11 125

test2506		94.32 86	93.00 115	95.87 51	96.16 76	99.39 15	96.96 61	92.80 64	95.22 93	94.47 27	91.55 80	70.45 193	95.25 90	98.29 28	97.98 29	99.59 7	98.10 126

ECVR-MVS		94.14 88	92.96 116	95.52 58	96.16 76	99.39 15	96.96 61	92.80 64	95.22 93	92.38 47	81.48 152	80.31 149	95.25 90	98.29 28	97.98 29	99.59 7	98.05 127

PLC	94.95 3	97.37 33	96.77 51	98.07 20	98.97 31	98.21 89	97.94 45	96.85 35	97.66 25	97.58 3	93.33 58	96.84 47	98.01 36	97.13 71	96.20 85	99.09 103	98.01 128
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
v2v482		88.25 169	87.71 179	88.88 149	89.23 188	95.28 171	92.10 163	87.89 130	88.69 177	73.31 175	75.32 173	71.64 187	91.89 136	92.10 183	92.92 169	98.86 130	97.99 129

ACMH+	90.88 12	91.41 129	91.13 145	91.74 115	95.11 99	96.95 116	93.13 143	89.48 113	92.42 139	79.93 137	85.13 132	78.02 159	93.82 115	93.49 162	93.88 151	98.94 122	97.99 129

v148		87.51 180	86.79 189	88.36 155	89.39 181	95.21 175	89.84 189	88.20 127	87.61 187	77.56 146	73.38 189	70.32 196	86.80 182	90.70 194	92.31 181	98.37 166	97.98 131

TAPA-MVS	94.18 5	96.38 50	96.49 55	96.25 43	98.26 46	98.66 72	98.00 43	94.96 43	97.17 39	89.48 84	92.91 63	96.35 52	97.53 43	96.59 88	95.90 95	99.28 70	97.82 132
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
CANet_DTU		93.92 95	96.57 53	90.83 126	95.63 84	98.39 84	96.99 60	87.38 134	96.26 59	71.97 179	96.31 33	93.02 73	94.53 102	97.38 63	96.83 71	98.49 160	97.79 133

GBi-Net		93.81 97	94.18 92	93.38 100	91.34 155	95.86 151	96.22 86	88.68 120	95.23 90	90.40 69	86.39 123	91.16 81	94.40 105	96.52 92	96.30 79	99.21 85	97.79 133

test1		93.81 97	94.18 92	93.38 100	91.34 155	95.86 151	96.22 86	88.68 120	95.23 90	90.40 69	86.39 123	91.16 81	94.40 105	96.52 92	96.30 79	99.21 85	97.79 133

FMVSNet3		93.79 99	94.17 94	93.35 102	91.21 158	95.99 144	96.62 75	88.68 120	95.23 90	90.40 69	86.39 123	91.16 81	94.11 109	95.96 112	96.67 73	99.07 106	97.79 133

FMVSNet2		93.30 108	93.36 111	93.22 103	91.34 155	95.86 151	96.22 86	88.24 126	95.15 96	89.92 80	81.64 150	89.36 95	94.40 105	96.77 81	96.98 66	99.21 85	97.79 133

pm-mvs1		89.19 158	89.02 161	89.38 146	90.40 163	95.74 158	92.05 165	88.10 128	86.13 195	77.70 145	73.72 186	79.44 154	88.97 173	95.81 118	94.51 138	99.08 104	97.78 138

FMVSNet1		91.54 127	90.93 148	92.26 110	90.35 165	95.27 173	95.22 110	87.16 137	91.37 154	87.62 103	75.45 172	83.84 135	94.43 103	96.52 92	96.30 79	98.82 132	97.74 139

LS3D		95.46 59	95.14 75	95.84 52	97.91 53	98.90 58	98.58 30	97.79 5	97.07 44	83.65 120	88.71 104	88.64 103	97.82 37	97.49 59	97.42 52	99.26 75	97.72 140

OMC-MVS		97.00 39	96.92 48	97.09 34	98.69 38	98.66 72	97.85 46	95.02 42	98.09 13	94.47 27	93.15 59	96.90 45	97.38 46	97.16 70	96.82 72	99.13 98	97.65 141

IterMVS-SCA-FT		90.24 142	92.48 126	87.63 173	92.85 142	94.30 196	93.79 132	81.47 187	92.66 132	69.95 191	84.66 136	88.38 106	89.99 167	95.39 130	94.34 141	97.74 182	97.63 142

DTE-MVSNet		86.67 188	86.09 195	87.35 179	88.45 199	94.08 198	90.65 183	86.05 149	86.13 195	72.19 178	74.58 180	66.77 210	87.61 179	90.31 195	93.12 165	99.13 98	97.62 143

DPM-MVS		96.86 44	96.82 50	96.91 38	98.08 50	98.20 90	98.52 33	97.20 28	97.24 38	91.42 55	91.84 75	98.45 34	97.25 48	97.07 72	97.40 54	98.95 121	97.55 144

CHOSEN 280x420		95.46 59	97.01 45	93.66 95	97.28 63	97.98 97	96.40 83	85.39 158	96.10 66	91.07 58	96.53 32	96.34 54	95.61 82	97.65 55	96.95 67	96.21 190	97.49 145

IterMVS		90.20 143	92.43 128	87.61 174	92.82 144	94.31 195	94.11 128	81.54 185	92.97 128	69.90 192	84.71 135	88.16 109	89.96 168	95.25 132	94.17 144	97.31 184	97.46 146
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
FC-MVSNet-test		91.63 124	93.82 102	89.08 148	92.02 150	96.40 136	93.26 141	87.26 135	93.72 119	77.26 148	88.61 107	89.86 92	85.50 190	95.72 123	95.02 120	99.16 94	97.44 147

gg-mvs-nofinetune		86.17 191	88.57 165	83.36 198	93.44 134	98.15 93	96.58 78	72.05 212	74.12 216	49.23 220	64.81 210	90.85 85	89.90 169	97.83 50	96.84 70	98.97 118	97.41 148

test-mter		90.95 132	93.54 110	87.93 168	90.28 166	96.80 121	91.44 173	82.68 181	92.15 147	74.37 172	89.57 100	88.23 108	90.88 153	96.37 99	94.31 142	97.93 177	97.37 149

ACMM	92.75 10	94.41 84	93.84 101	95.09 66	96.41 73	96.80 121	94.88 116	93.54 50	96.41 56	90.16 74	92.31 69	83.11 140	96.32 70	96.22 105	94.65 129	99.22 82	97.35 150
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
GA-MVS		89.28 155	90.75 151	87.57 175	91.77 151	96.48 132	92.29 159	87.58 131	90.61 163	65.77 201	84.48 137	76.84 167	89.46 170	95.84 116	93.68 156	98.52 158	97.34 151

pmmvs6		85.98 193	84.89 201	87.25 180	88.83 195	94.35 194	89.36 191	85.30 161	78.51 213	75.44 162	62.71 212	75.41 170	87.65 178	93.58 159	92.40 180	96.89 187	97.29 152

OPM-MVS		93.61 102	92.43 128	95.00 68	96.94 67	97.34 109	97.78 47	94.23 46	89.64 169	85.53 112	88.70 105	82.81 141	96.28 71	96.28 103	95.00 122	99.24 76	97.22 153
Ray L. Khuboni, Hongjun Xu: Octagram Propagation Matching for Multi-Scale View Stereopsis (OPM-MVS).
SixPastTwentyTwo		88.37 167	89.47 158	87.08 181	90.01 170	95.93 150	87.41 196	85.32 159	90.26 167	70.26 188	86.34 126	71.95 186	90.93 150	92.89 172	91.72 186	98.55 156	97.22 153

pmmvs5		87.83 177	88.09 170	87.51 178	89.59 177	95.48 165	89.75 190	84.73 167	86.07 197	71.44 182	80.57 158	70.09 197	90.74 157	94.47 145	92.87 171	98.82 132	97.10 155

CVMVSNet		89.77 150	91.66 140	87.56 176	93.21 139	95.45 167	91.94 170	89.22 115	89.62 170	69.34 196	83.99 141	85.90 119	84.81 195	94.30 150	95.28 112	96.85 188	97.09 156

PMMVS		94.61 76	95.56 66	93.50 97	94.30 121	96.74 125	94.91 115	89.56 111	95.58 83	87.72 102	96.15 34	92.86 74	96.06 73	95.47 127	95.02 120	98.43 165	97.09 156

CR-MVSNet		90.16 145	91.96 138	88.06 162	93.32 136	95.95 148	93.36 139	75.99 204	92.40 140	75.19 165	83.18 144	85.37 122	92.05 134	95.21 133	94.56 134	98.47 162	97.08 158

PatchT		89.13 159	91.71 139	86.11 190	92.92 140	95.59 162	83.64 206	75.09 207	91.87 149	75.19 165	82.63 147	85.06 127	92.05 134	95.21 133	94.56 134	97.76 179	97.08 158

baseline1		94.59 77	94.47 85	94.72 78	95.16 97	97.97 98	96.07 92	91.94 76	94.86 100	89.98 77	91.60 79	85.87 120	95.64 80	97.07 72	96.90 68	99.52 20	97.06 160

UA-Net		93.96 92	95.95 62	91.64 116	96.06 79	98.59 79	95.29 108	90.00 102	91.06 157	82.87 122	90.64 90	98.06 39	86.06 187	98.14 39	98.20 19	99.58 11	96.96 161

tpm		87.95 172	89.44 159	86.21 189	92.53 146	94.62 190	91.40 174	76.36 201	91.46 153	69.80 194	87.43 111	75.14 171	91.55 141	89.85 200	90.60 190	95.61 197	96.96 161

RPMNet		90.19 144	92.03 137	88.05 163	93.46 133	95.95 148	93.41 137	74.59 209	92.40 140	75.91 159	84.22 139	86.41 115	92.49 130	94.42 147	93.85 153	98.44 163	96.96 161

test-LLR		91.62 125	93.56 108	89.35 147	93.31 137	96.57 130	92.02 167	87.06 138	92.34 143	75.05 168	90.20 94	88.64 103	90.93 150	96.19 108	94.07 146	97.75 180	96.90 164

TESTMET0.1,1		91.07 131	93.56 108	88.17 158	90.43 162	96.57 130	92.02 167	82.83 180	92.34 143	75.05 168	90.20 94	88.64 103	90.93 150	96.19 108	94.07 146	97.75 180	96.90 164

pmmvs4		90.55 138	89.91 155	91.30 121	90.26 167	94.95 181	92.73 149	87.94 129	93.44 124	85.35 113	82.28 149	76.09 168	93.02 128	93.56 160	92.26 183	98.51 159	96.77 166

TSAR-MVS + COLMAP		94.79 71	94.51 84	95.11 65	96.50 70	97.54 103	97.99 44	94.54 44	97.81 17	85.88 111	96.73 31	81.28 148	96.99 56	96.29 102	95.21 115	98.76 142	96.73 167

PatchMatch-RL		94.69 75	94.41 86	95.02 67	97.63 58	98.15 93	94.50 124	91.99 74	95.32 87	91.31 57	95.47 44	83.44 138	96.02 75	96.56 89	95.23 114	98.69 146	96.67 168

PM-MVS		84.72 198	84.47 202	85.03 193	84.67 208	91.57 207	86.27 200	82.31 183	87.65 186	70.62 186	76.54 171	56.41 219	88.75 175	92.59 174	89.85 195	97.54 183	96.66 169

test0.0.03 1		91.97 119	93.91 98	89.72 140	93.31 137	96.40 136	91.34 176	87.06 138	93.86 116	81.67 129	91.15 85	89.16 99	86.02 188	95.08 136	95.09 117	98.91 125	96.64 170

testgi		89.42 152	91.50 143	87.00 183	92.40 148	95.59 162	89.15 192	85.27 162	92.78 131	72.42 177	91.75 77	76.00 169	84.09 199	94.38 148	93.82 155	98.65 151	96.15 171

CostFormer		90.69 135	90.48 153	90.93 124	94.18 122	96.08 143	94.03 129	78.20 194	93.47 123	89.96 78	90.97 88	80.30 150	93.72 117	87.66 206	88.75 198	95.51 199	96.12 172

EU-MVSNet		85.62 194	87.65 180	83.24 199	88.54 198	92.77 204	87.12 197	85.32 159	86.71 191	64.54 204	78.52 165	75.11 172	78.35 204	92.25 179	92.28 182	95.58 198	95.93 173

TransMVSNet (Re)		87.73 178	86.79 189	88.83 150	90.76 159	94.40 193	91.33 177	89.62 110	84.73 201	75.41 163	72.73 191	71.41 189	86.80 182	94.53 144	93.93 150	99.06 109	95.83 174

EPNet_dtu		92.45 117	95.02 79	89.46 144	98.02 51	95.47 166	94.79 118	92.62 66	94.97 98	70.11 190	94.76 53	92.61 77	84.07 200	95.94 113	95.56 104	97.15 186	95.82 175
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
MSDG		94.82 69	93.73 103	96.09 47	98.34 45	97.43 108	97.06 58	96.05 37	95.84 75	90.56 67	86.30 127	89.10 100	95.55 84	96.13 110	95.61 103	99.00 115	95.73 176

pmmvs-eth3d		84.33 199	82.94 204	85.96 192	84.16 209	90.94 208	86.55 199	83.79 173	84.25 202	75.85 160	70.64 200	56.43 218	87.44 181	92.20 180	90.41 192	97.97 176	95.68 177

GG-mvs-BLEND		66.17 212	94.91 81	32.63 217	1.32 225	96.64 128	91.40 174	0.85 223	94.39 109	2.20 226	90.15 96	95.70 61	2.27 222	96.39 96	95.44 108	97.78 178	95.68 177

gm-plane-assit		83.26 201	85.29 198	80.89 201	89.52 178	89.89 211	70.26 217	78.24 193	77.11 214	58.01 217	74.16 183	66.90 207	90.63 160	97.20 67	96.05 89	98.66 150	95.68 177

TAMVS		90.54 139	90.87 150	90.16 135	91.48 153	96.61 129	93.26 141	86.08 148	87.71 185	81.66 130	83.11 146	84.04 133	90.42 162	94.54 143	94.60 131	98.04 175	95.48 180

EG-PatchMatch MVS		86.68 187	87.24 183	86.02 191	90.58 161	96.26 139	91.08 180	81.59 184	84.96 200	69.80 194	71.35 199	75.08 173	84.23 198	94.24 152	93.35 161	98.82 132	95.46 181

COLMAP_ROB	90.49 14	93.27 109	92.71 118	93.93 90	97.75 56	97.44 107	96.07 92	93.17 58	95.40 84	83.86 118	83.76 142	88.72 102	93.87 113	94.25 151	94.11 145	98.87 128	95.28 182
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
ambc			73.83 212		76.23 216	85.13 215	82.27 209		84.16 203	65.58 203	52.82 215	23.31 226	73.55 209	91.41 191	85.26 209	92.97 212	94.70 183

MS-PatchMatch		91.82 121	92.51 122	91.02 122	95.83 83	96.88 117	95.05 111	84.55 171	93.85 117	82.01 126	82.51 148	91.71 79	90.52 161	95.07 137	93.03 167	98.13 171	94.52 184

TDRefinement		89.07 160	88.15 169	90.14 137	95.16 97	96.88 117	95.55 106	90.20 100	89.68 168	76.42 155	76.67 169	74.30 176	84.85 194	93.11 167	91.91 185	98.64 152	94.47 185

MDTV_nov1_ep13		91.57 126	93.18 112	89.70 141	93.39 135	96.97 115	93.53 135	80.91 188	95.70 78	81.86 127	92.40 68	89.93 91	93.25 125	91.97 186	90.80 189	95.25 203	94.46 186

dps		90.11 147	89.37 160	90.98 123	93.89 128	96.21 140	93.49 136	77.61 196	91.95 148	92.74 45	88.85 103	78.77 157	92.37 132	87.71 205	87.71 202	95.80 195	94.38 187

Anonymous20231206		83.84 200	85.19 199	82.26 200	87.38 204	92.87 202	85.49 202	83.65 174	86.07 197	63.44 208	68.42 203	69.01 201	75.45 208	93.34 163	92.44 179	98.12 173	94.20 188

USDC		90.69 135	90.52 152	90.88 125	94.17 123	96.43 134	95.82 103	86.76 140	93.92 115	76.27 157	86.49 121	74.30 176	93.67 119	95.04 138	93.36 160	98.61 153	94.13 189

pmnet_mix02		86.12 192	87.12 186	84.96 194	89.82 172	94.12 197	84.88 204	86.63 142	91.78 150	65.60 202	80.76 156	76.98 165	86.61 184	87.29 207	84.80 210	96.21 190	94.09 190

tpm cat1		88.90 162	87.78 178	90.22 134	93.88 129	95.39 169	93.79 132	78.11 195	92.55 136	89.43 85	81.31 153	79.84 153	91.40 142	84.95 209	86.34 207	94.68 209	94.09 190

SCA		90.92 133	93.04 114	88.45 154	93.72 132	97.33 110	92.77 147	76.08 203	96.02 68	78.26 144	91.96 73	90.86 84	93.99 112	90.98 193	90.04 194	95.88 194	94.06 192

RPSCF		94.05 90	94.00 97	94.12 88	96.20 75	96.41 135	96.61 76	91.54 85	95.83 76	89.73 81	96.94 30	92.80 75	95.35 89	91.63 189	90.44 191	95.27 202	93.94 193

tpmrst		88.86 164	89.62 156	87.97 167	94.33 120	95.98 145	92.62 151	76.36 201	94.62 104	76.94 151	85.98 128	82.80 142	92.80 129	86.90 208	87.15 204	94.77 207	93.93 194

MDTV_nov1_ep13_2view		86.30 190	88.27 167	84.01 196	87.71 203	94.67 188	88.08 194	76.78 199	90.59 164	68.66 198	80.46 160	80.12 151	87.58 180	89.95 199	88.20 200	95.25 203	93.90 195

ADS-MVSNet		89.80 149	91.33 144	88.00 166	94.43 119	96.71 126	92.29 159	74.95 208	96.07 67	77.39 147	88.67 106	86.09 117	93.26 124	88.44 202	89.57 196	95.68 196	93.81 196

PatchmatchNet		90.56 137	92.49 124	88.31 157	93.83 130	96.86 120	92.42 155	76.50 200	95.96 70	78.31 143	91.96 73	89.66 93	93.48 121	90.04 198	89.20 197	95.32 200	93.73 197
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
MIMVSNet		88.99 161	91.07 146	86.57 186	86.78 206	95.62 159	91.20 179	75.40 206	90.65 162	76.57 153	84.05 140	82.44 144	91.01 149	95.84 116	95.38 109	98.48 161	93.50 198

EPMVS		90.88 134	92.12 134	89.44 145	94.71 112	97.24 111	93.55 134	76.81 198	95.89 72	81.77 128	91.49 81	86.47 114	93.87 113	90.21 196	90.07 193	95.92 193	93.49 199

TinyColmap		89.42 152	88.58 164	90.40 132	93.80 131	95.45 167	93.96 131	86.54 143	92.24 145	76.49 154	80.83 155	70.44 194	93.37 122	94.45 146	93.30 163	98.26 169	93.37 200

test_method		72.96 209	78.68 209	66.28 212	50.17 222	64.90 220	75.45 216	50.90 219	87.89 182	62.54 209	62.98 211	68.34 204	70.45 210	91.90 187	82.41 211	88.19 216	92.35 201

MDA-MVSNet-bldmvs		80.11 204	80.24 207	79.94 203	77.01 215	93.21 201	78.86 213	85.94 151	82.71 208	60.86 210	79.71 162	51.77 221	83.71 201	75.60 214	86.37 206	93.28 211	92.35 201

N_pmnet		84.80 196	85.10 200	84.45 195	89.25 187	92.86 203	84.04 205	86.21 145	88.78 175	66.73 200	72.41 194	74.87 175	85.21 192	88.32 203	86.45 205	95.30 201	92.04 203

test20.03		82.92 202	85.52 197	79.90 204	87.75 202	91.84 206	82.80 208	82.99 178	82.65 209	60.32 213	78.90 164	70.50 192	67.10 212	92.05 185	90.89 188	98.44 163	91.80 204

pmmvs3		79.16 206	80.12 208	78.05 207	79.36 213	86.59 214	78.13 214	73.87 210	76.42 215	57.51 218	70.59 201	57.02 217	84.66 196	90.10 197	88.32 199	94.75 208	91.77 205

FMVSNet5		90.36 140	90.93 148	89.70 141	87.99 200	92.25 205	92.03 166	83.51 175	92.20 146	84.13 116	85.59 130	86.48 113	92.43 131	94.61 141	94.52 137	98.13 171	90.85 206

new-patchmatchnet		78.49 207	78.19 210	78.84 206	84.13 210	90.06 210	77.11 215	80.39 189	79.57 212	59.64 216	66.01 208	55.65 220	75.62 207	84.55 210	80.70 212	96.14 192	90.77 207

MIMVSNet1		80.03 205	80.93 206	78.97 205	72.46 218	90.73 209	80.81 211	82.44 182	80.39 210	63.64 206	57.57 213	64.93 213	76.37 206	91.66 188	91.55 187	98.07 174	89.70 208

MVS-HIRNet		85.36 195	86.89 188	83.57 197	90.13 168	94.51 191	83.57 207	72.61 211	88.27 181	71.22 184	68.97 202	81.81 145	88.91 174	93.08 168	91.94 184	94.97 206	89.64 209

CMPMVS	65.18 17	84.76 197	83.10 203	86.69 185	95.29 93	95.05 178	88.37 193	85.51 157	80.27 211	71.31 183	68.37 204	73.85 178	85.25 191	87.72 204	87.75 201	94.38 210	88.70 210
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
new_pmnet		81.53 203	82.68 205	80.20 202	83.47 211	89.47 212	82.21 210	78.36 192	87.86 183	60.14 215	67.90 205	69.43 200	82.03 202	89.22 201	87.47 203	94.99 205	87.39 211

DeepMVS_CX						86.86 213	79.50 212	70.43 214	90.73 160	63.66 205	80.36 161	60.83 214	79.68 203	76.23 213		89.46 214	86.53 212

PMMVS2		64.36 213	65.94 215	62.52 213	67.37 219	77.44 217	64.39 219	69.32 217	61.47 218	34.59 221	46.09 216	41.03 222	48.02 219	74.56 216	78.23 213	91.43 213	82.76 213

FPMVS		75.84 208	74.59 211	77.29 208	86.92 205	83.89 216	85.01 203	80.05 190	82.91 207	60.61 212	65.25 209	60.41 215	63.86 213	75.60 214	73.60 216	87.29 217	80.47 214

Gipuma		68.35 210	66.71 213	70.27 209	74.16 217	68.78 219	63.93 220	71.77 213	83.34 206	54.57 219	34.37 217	31.88 223	68.69 211	83.30 211	85.53 208	88.48 215	79.78 215
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
PMVS	63.12 18	67.27 211	66.39 214	68.30 210	77.98 214	60.24 221	59.53 221	76.82 197	66.65 217	60.74 211	54.39 214	59.82 216	51.24 216	73.92 217	70.52 217	83.48 218	79.17 216
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
MVE	50.86 19	49.54 216	51.43 216	47.33 216	44.14 223	59.20 222	36.45 224	60.59 218	41.47 221	31.14 222	29.58 218	17.06 227	48.52 218	62.22 218	74.63 215	63.12 222	75.87 217
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
E-PMN		50.67 214	47.85 217	53.96 214	64.13 221	50.98 224	38.06 222	69.51 215	51.40 220	24.60 223	29.46 220	24.39 225	56.07 215	48.17 219	59.70 218	71.40 220	70.84 218

EMVS		49.98 215	46.76 218	53.74 215	64.96 220	51.29 223	37.81 223	69.35 216	51.83 219	22.69 224	29.57 219	25.06 224	57.28 214	44.81 220	56.11 219	70.32 221	68.64 219

testmvs		12.09 217	16.94 219	6.42 218	3.15 224	6.08 225	9.51 226	3.84 221	21.46 222	5.31 225	27.49 221	6.76 228	10.89 220	17.06 221	15.01 220	5.84 223	24.75 220

test123		9.58 218	13.53 220	4.97 219	1.31 226	5.47 226	8.32 227	2.95 222	18.14 223	2.03 227	20.82 222	2.34 229	10.60 221	10.00 222	14.16 221	4.60 224	23.77 221

uanet_test		0.00 219	0.00 221	0.00 220	0.00 227	0.00 227	0.00 228	0.00 224	0.00 224	0.00 228	0.00 223	0.00 230	0.00 223	0.00 223	0.00 222	0.00 225	0.00 222

sosnet-low-res		0.00 219	0.00 221	0.00 220	0.00 227	0.00 227	0.00 228	0.00 224	0.00 224	0.00 228	0.00 223	0.00 230	0.00 223	0.00 223	0.00 222	0.00 225	0.00 222

sosnet		0.00 219	0.00 221	0.00 220	0.00 227	0.00 227	0.00 228	0.00 224	0.00 224	0.00 228	0.00 223	0.00 230	0.00 223	0.00 223	0.00 222	0.00 225	0.00 222

RE-MVS-def										63.50 207

9.14												99.28 12

SR-MVS					99.45 9			97.61 14				99.20 16

our_test_3					89.78 173	93.84 199	85.59 201

MTAPA										96.83 10		99.12 21

MTMP										97.18 5		98.83 26

Patchmatch-RL test							34.61 225

tmp_tt				66.88 211	86.07 207	73.86 218	68.22 218	33.38 220	96.88 48	80.67 135	88.23 109	78.82 156	49.78 217	82.68 212	77.47 214	83.19 219

XVS					96.60 68	99.35 17	96.82 67			90.85 61		98.72 29				99.46 32

X-MVStestdata					96.60 68	99.35 17	96.82 67			90.85 61		98.72 29				99.46 32

mPP-MVS					99.21 23							98.29 37

NP-MVS									95.32 87

Patchmtry						95.96 147	93.36 139	75.99 204		75.19 165

High-resolution multi-view benchmark