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.

DELS-MVS		96.06 52	96.04 58	96.07 49	97.77 55	99.25 22	98.10 41	93.26 55	94.42 102	92.79 42	88.52 106	93.48 69	95.06 87	98.51 14	98.83 1	99.45 30	99.28 24
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
DeepC-MVS_fast	96.13 1	98.13 19	98.27 24	97.97 24	99.16 26	99.03 38	99.05 17	97.24 26	98.22 8	94.17 31	95.82 37	98.07 38	98.69 17	98.83 9	98.80 2	99.52 14	99.10 51
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
3Dnovator	93.79 8	97.08 36	97.20 37	96.95 37	99.09 28	99.03 38	98.20 39	93.33 53	97.99 13	93.82 32	90.61 89	96.80 48	97.82 34	97.90 43	98.78 3	99.47 25	99.26 29

MSLP-MVS++		98.04 22	97.93 31	98.18 16	99.10 27	99.09 30	98.34 35	96.99 32	97.54 29	96.60 12	94.82 48	98.45 34	98.89 6	97.46 54	98.77 4	99.17 85	99.37 16

DeepPCF-MVS	95.28 2	97.00 39	98.35 19	95.42 57	97.30 61	98.94 47	94.82 110	96.03 38	98.24 7	92.11 47	95.80 38	98.64 31	95.51 82	98.95 5	98.66 5	96.78 182	99.20 38

SMA-MVS		98.66 5	98.89 5	98.39 8	99.60 1	99.41 7	99.00 19	97.63 11	97.78 16	95.83 18	98.33 9	99.83 2	98.85 10	98.93 6	98.56 6	99.41 43	99.40 13

IS_MVSNet		95.28 61	96.43 53	93.94 84	95.30 88	99.01 42	95.90 91	91.12 86	94.13 107	87.50 98	91.23 80	94.45 65	94.17 101	98.45 17	98.50 7	99.65 2	99.23 33

DeepC-MVS	94.87 4	96.76 46	96.50 51	97.05 35	98.21 48	99.28 18	98.67 26	97.38 20	97.31 34	90.36 68	89.19 99	93.58 68	98.19 26	98.31 23	98.50 7	99.51 19	99.36 17
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
SteuartSystems-ACMMP		98.38 13	98.71 8	97.99 23	99.34 20	99.46 6	99.34 5	97.33 24	97.31 34	94.25 29	98.06 12	99.17 17	98.13 27	98.98 3	98.46 9	99.55 11	99.54 8
Skip Steuart: Steuart Systems R&D Blog.
3Dnovator+	93.91 7	97.23 34	97.22 36	97.24 32	98.89 35	98.85 57	98.26 38	93.25 57	97.99 13	95.56 22	90.01 95	98.03 40	98.05 31	97.91 42	98.43 10	99.44 38	99.35 18

MSP-MVS		98.86 2	98.97 1	98.75 2	99.43 13	99.63 1	99.25 11	97.81 1	98.62 1	97.69 1	97.59 19	99.90 1	98.93 5	98.99 2	98.42 11	99.37 52	99.62 3

DVP-MVS		98.73 4	98.93 3	98.50 5	99.44 12	99.57 2	99.36 3	97.65 7	98.14 10	96.51 14	98.49 5	99.65 6	98.67 18	98.60 12	98.42 11	99.40 46	99.63 1

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

APD-MVS		98.36 14	98.32 21	98.41 7	99.47 6	99.26 20	99.12 14	97.77 5	96.73 48	96.12 16	97.27 27	98.88 23	98.46 24	98.47 16	98.39 14	99.52 14	99.22 35
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
PHI-MVS		97.78 25	98.44 16	97.02 36	98.73 37	99.25 22	98.11 40	95.54 39	96.66 51	92.79 42	98.52 4	99.38 7	97.50 41	97.84 44	98.39 14	99.45 30	99.03 63

Vis-MVSNet (Re-imp)		94.46 78	96.24 55	92.40 102	95.23 91	98.64 71	95.56 99	90.99 87	94.42 102	85.02 108	90.88 87	94.65 64	88.01 169	98.17 30	98.37 16	99.57 8	98.53 100

DPE-MVS		98.75 3	98.91 4	98.57 3	99.21 23	99.54 3	99.42 2	97.78 4	97.49 30	96.84 8	98.94 1	99.82 3	98.59 20	98.90 8	98.22 17	99.56 10	99.48 10

UA-Net		93.96 87	95.95 60	91.64 109	96.06 75	98.59 75	95.29 101	90.00 98	91.06 149	82.87 115	90.64 88	98.06 39	86.06 179	98.14 33	98.20 18	99.58 6	96.96 155

QAPM		96.78 45	97.14 41	96.36 43	99.05 29	99.14 29	98.02 42	93.26 55	97.27 36	90.84 59	91.16 81	97.31 43	97.64 39	97.70 48	98.20 18	99.33 55	99.18 43

X-MVS		97.84 23	98.19 26	97.42 30	99.40 15	99.35 11	99.06 16	97.25 25	97.38 33	90.85 56	96.06 35	98.72 28	98.53 23	98.41 21	98.15 20	99.46 26	99.28 24

ACMMPR		98.40 11	98.49 11	98.28 13	99.41 14	99.40 8	99.36 3	97.35 21	98.30 5	95.02 25	97.79 16	98.39 36	99.04 2	98.26 26	98.10 21	99.50 21	99.22 35

CNVR-MVS		98.47 9	98.46 14	98.48 6	99.40 15	99.05 31	99.02 18	97.54 16	97.73 17	96.65 11	97.20 28	99.13 18	98.85 10	98.91 7	98.10 21	99.41 43	99.08 54

HFP-MVS		98.48 8	98.62 9	98.32 11	99.39 18	99.33 15	99.27 9	97.42 18	98.27 6	95.25 23	98.34 8	98.83 25	99.08 1	98.26 26	98.08 23	99.48 22	99.26 29

CANet		96.84 43	97.20 37	96.42 41	97.92 53	99.24 24	98.60 28	93.51 52	97.11 40	93.07 36	91.16 81	97.24 44	96.21 70	98.24 28	98.05 24	99.22 77	99.35 18

TSAR-MVS + MP.		98.49 7	98.78 6	98.15 19	98.14 50	99.17 27	99.34 5	97.18 29	98.44 4	95.72 19	97.84 15	99.28 10	98.87 7	99.05 1	98.05 24	99.66 1	99.60 5
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
SD-MVS		98.52 6	98.77 7	98.23 15	98.15 49	99.26 20	98.79 25	97.59 15	98.52 2	96.25 15	97.99 14	99.75 4	99.01 3	98.27 25	97.97 26	99.59 4	99.63 1
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
MVS_111021_HR		97.04 38	98.20 25	95.69 52	98.44 45	99.29 17	96.59 73	93.20 58	97.70 21	89.94 74	98.46 6	96.89 46	96.71 62	98.11 36	97.95 27	99.27 66	99.01 66

canonicalmvs		95.25 63	95.45 67	95.00 64	95.27 90	98.72 65	96.89 61	89.82 102	96.51 52	90.84 59	93.72 56	86.01 114	97.66 38	95.78 111	97.94 28	99.54 13	99.50 9

CDPH-MVS		96.84 43	97.49 33	96.09 47	98.92 33	98.85 57	98.61 27	95.09 41	96.00 67	87.29 99	95.45 43	97.42 42	97.16 49	97.83 45	97.94 28	99.44 38	98.92 76

MVS_0304		96.31 48	96.91 46	95.62 53	97.21 63	99.20 25	98.55 30	93.10 60	97.04 43	89.73 76	90.30 91	96.35 52	95.71 76	98.14 33	97.93 30	99.38 49	99.40 13

PGM-MVS		97.81 24	98.11 27	97.46 29	99.55 3	99.34 14	99.32 7	94.51 45	96.21 59	93.07 36	98.05 13	97.95 41	98.82 12	98.22 29	97.89 31	99.48 22	99.09 53

ACMMP_NAP		98.20 17	98.49 11	97.85 25	99.50 4	99.40 8	99.26 10	97.64 10	97.47 32	92.62 45	97.59 19	99.09 20	98.71 16	98.82 10	97.86 32	99.40 46	99.19 39

ETV-MVS		96.31 48	97.47 35	94.96 66	94.79 102	98.78 60	96.08 86	91.41 83	96.16 60	90.50 63	95.76 39	96.20 56	97.39 42	98.42 20	97.82 33	99.57 8	99.18 43

TSAR-MVS + ACMM		97.71 27	98.60 10	96.66 40	98.64 40	99.05 31	98.85 24	97.23 27	98.45 3	89.40 82	97.51 23	99.27 12	96.88 58	98.53 13	97.81 34	98.96 113	99.59 6

NCCC		98.10 20	98.05 29	98.17 18	99.38 19	99.05 31	99.00 19	97.53 17	98.04 12	95.12 24	94.80 49	99.18 16	98.58 21	98.49 15	97.78 35	99.39 48	98.98 70

MP-MVS		98.09 21	98.30 23	97.84 26	99.34 20	99.19 26	99.23 12	97.40 19	97.09 41	93.03 39	97.58 21	98.85 24	98.57 22	98.44 19	97.69 36	99.48 22	99.23 33
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
CP-MVS		98.32 16	98.34 20	98.29 12	99.34 20	99.30 16	99.15 13	97.35 21	97.49 30	95.58 21	97.72 17	98.62 32	98.82 12	98.29 24	97.67 37	99.51 19	99.28 24

xxxxxxxxxxxxxcwj		97.07 37	95.99 59	98.33 9	99.45 9	99.05 31	98.27 36	97.65 7	97.73 17	97.02 6	98.18 10	81.99 140	98.11 28	98.15 31	97.62 38	99.45 30	99.19 39

SF-MVS		98.39 12	98.45 15	98.33 9	99.45 9	99.05 31	98.27 36	97.65 7	97.73 17	97.02 6	98.18 10	99.25 13	98.11 28	98.15 31	97.62 38	99.45 30	99.19 39

OpenMVS	92.33 11	95.50 54	95.22 71	95.82 51	98.98 30	98.97 45	97.67 50	93.04 63	94.64 98	89.18 86	84.44 134	94.79 63	96.79 59	97.23 59	97.61 40	99.24 71	98.88 81

Vis-MVSNet		92.77 105	95.00 77	90.16 128	94.10 118	98.79 59	94.76 112	88.26 120	92.37 135	79.95 129	88.19 108	91.58 77	84.38 189	97.59 51	97.58 41	99.52 14	98.91 79
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
zzz-MVS		98.43 10	98.31 22	98.57 3	99.48 5	99.40 8	99.32 7	97.62 12	97.70 21	96.67 10	96.59 31	99.09 20	98.86 8	98.65 11	97.56 42	99.45 30	99.17 45

MAR-MVS		95.50 54	95.60 63	95.39 58	98.67 39	98.18 86	95.89 93	89.81 103	94.55 100	91.97 48	92.99 60	90.21 87	97.30 45	96.79 73	97.49 43	98.72 136	98.99 68
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
MCST-MVS		98.20 17	98.36 17	98.01 22	99.40 15	99.05 31	99.00 19	97.62 12	97.59 28	93.70 33	97.42 26	99.30 9	98.77 14	98.39 22	97.48 44	99.59 4	99.31 23

HPM-MVS++		98.34 15	98.47 13	98.18 16	99.46 8	99.15 28	99.10 15	97.69 6	97.67 24	94.93 26	97.62 18	99.70 5	98.60 19	98.45 17	97.46 45	99.31 60	99.26 29

CS-MVS		96.23 51	97.15 40	95.16 60	95.01 98	98.98 43	97.13 56	90.68 91	96.00 67	91.21 53	94.03 53	96.48 50	97.35 44	98.00 40	97.43 46	99.55 11	99.15 47

LS3D		95.46 57	95.14 72	95.84 50	97.91 54	98.90 54	98.58 29	97.79 3	97.07 42	83.65 113	88.71 102	88.64 100	97.82 34	97.49 53	97.42 47	99.26 70	97.72 134

TSAR-MVS + GP.		97.45 30	98.36 17	96.39 42	95.56 82	98.93 49	97.74 48	93.31 54	97.61 27	94.24 30	98.44 7	99.19 15	98.03 32	97.60 50	97.41 48	99.44 38	99.33 20

DPM-MVS		96.86 42	96.82 47	96.91 38	98.08 51	98.20 84	98.52 32	97.20 28	97.24 37	91.42 51	91.84 74	98.45 34	97.25 46	97.07 65	97.40 49	98.95 114	97.55 138

CSCG		97.44 31	97.18 39	97.75 27	99.47 6	99.52 5	98.55 30	95.41 40	97.69 23	95.72 19	94.29 52	95.53 61	98.10 30	96.20 99	97.38 50	99.24 71	99.62 3

PVSNet_BlendedMVS		95.41 59	95.28 69	95.57 54	97.42 59	99.02 40	95.89 93	93.10 60	96.16 60	93.12 34	91.99 70	85.27 119	94.66 92	98.09 37	97.34 51	99.24 71	99.08 54

PVSNet_Blended		95.41 59	95.28 69	95.57 54	97.42 59	99.02 40	95.89 93	93.10 60	96.16 60	93.12 34	91.99 70	85.27 119	94.66 92	98.09 37	97.34 51	99.24 71	99.08 54

casdiffmvs		94.38 82	94.15 92	94.64 76	94.70 108	98.51 76	96.03 89	91.66 78	95.70 77	89.36 83	86.48 118	85.03 124	96.60 65	97.40 55	97.30 53	99.52 14	98.67 92

PVSNet_Blended_VisFu		94.77 71	95.54 65	93.87 86	96.48 70	98.97 45	94.33 119	91.84 75	94.93 94	90.37 67	85.04 129	94.99 62	90.87 147	98.12 35	97.30 53	99.30 62	99.45 12

MVS_111021_LR		97.16 35	98.01 30	96.16 46	98.47 43	98.98 43	96.94 60	93.89 48	97.64 26	91.44 50	98.89 2	96.41 51	97.20 48	98.02 39	97.29 55	99.04 108	98.85 85

train_agg		97.65 28	98.06 28	97.18 33	98.94 32	98.91 52	98.98 23	97.07 31	96.71 49	90.66 61	97.43 25	99.08 22	98.20 25	97.96 41	97.14 56	99.22 77	99.19 39

EIA-MVS		95.50 54	96.19 56	94.69 74	94.83 101	98.88 56	95.93 90	91.50 82	94.47 101	89.43 80	93.14 59	92.72 73	97.05 54	97.82 47	97.13 57	99.43 41	99.15 47

EPNet		96.27 50	96.97 43	95.46 56	98.47 43	98.28 80	97.41 53	93.67 50	95.86 73	92.86 41	97.51 23	93.79 67	91.76 132	97.03 67	97.03 58	98.61 146	99.28 24
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
EPP-MVSNet		95.27 62	96.18 57	94.20 82	94.88 100	98.64 71	94.97 106	90.70 90	95.34 84	89.67 78	91.66 77	93.84 66	95.42 84	97.32 57	97.00 59	99.58 6	99.47 11

FMVSNet2		93.30 101	93.36 107	93.22 98	91.34 148	95.86 144	96.22 81	88.24 121	95.15 92	89.92 75	81.64 146	89.36 92	94.40 98	96.77 74	96.98 60	99.21 80	97.79 127

CHOSEN 280x420		95.46 57	97.01 42	93.66 90	97.28 62	97.98 91	96.40 79	85.39 151	96.10 64	91.07 54	96.53 32	96.34 54	95.61 79	97.65 49	96.95 61	96.21 183	97.49 139

baseline1		94.59 75	94.47 82	94.72 73	95.16 93	97.97 92	96.07 87	91.94 73	94.86 95	89.98 72	91.60 78	85.87 116	95.64 78	97.07 65	96.90 62	99.52 14	97.06 154

MVS_Test		94.82 67	95.66 62	93.84 87	94.79 102	98.35 79	96.49 77	89.10 113	96.12 63	87.09 101	92.58 65	90.61 84	96.48 66	96.51 88	96.89 63	99.11 95	98.54 99

gg-mvs-nofinetune		86.17 183	88.57 157	83.36 190	93.44 127	98.15 87	96.58 74	72.05 204	74.12 207	49.23 210	64.81 202	90.85 82	89.90 162	97.83 45	96.84 64	98.97 112	97.41 142

CANet_DTU		93.92 88	96.57 50	90.83 119	95.63 80	98.39 78	96.99 59	87.38 129	96.26 56	71.97 172	96.31 33	93.02 70	94.53 95	97.38 56	96.83 65	98.49 153	97.79 127

OMC-MVS		97.00 39	96.92 45	97.09 34	98.69 38	98.66 68	97.85 46	95.02 42	98.09 11	94.47 27	93.15 58	96.90 45	97.38 43	97.16 63	96.82 66	99.13 92	97.65 135

FMVSNet3		93.79 92	94.17 90	93.35 97	91.21 151	95.99 137	96.62 71	88.68 115	95.23 88	90.40 64	86.39 119	91.16 78	94.11 102	95.96 104	96.67 67	99.07 100	97.79 127

CNLPA		96.90 41	96.28 54	97.64 28	98.56 42	98.63 73	96.85 63	96.60 36	97.73 17	97.08 5	89.78 97	96.28 55	97.80 36	96.73 76	96.63 68	98.94 115	98.14 121

UGNet		94.92 64	96.63 49	92.93 99	96.03 76	98.63 73	94.53 116	91.52 81	96.23 58	90.03 71	92.87 63	96.10 58	86.28 178	96.68 78	96.60 69	99.16 88	99.32 22
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
CHOSEN 1792x2688		92.66 107	92.49 116	92.85 100	97.13 64	98.89 55	95.90 91	88.50 119	95.32 85	83.31 114	71.99 187	88.96 98	94.10 103	96.69 77	96.49 70	98.15 163	99.10 51

CDS-MVSNet		92.77 105	93.60 102	91.80 107	92.63 138	96.80 114	95.24 102	89.14 112	90.30 158	84.58 109	86.76 113	90.65 83	90.42 155	95.89 106	96.49 70	98.79 132	98.32 115
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
diffmvs		94.31 83	94.21 87	94.42 79	94.64 109	98.28 80	96.36 80	91.56 79	96.77 47	88.89 89	88.97 100	84.23 127	96.01 74	96.05 103	96.41 72	99.05 107	98.79 89

GBi-Net		93.81 90	94.18 88	93.38 95	91.34 148	95.86 144	96.22 81	88.68 115	95.23 88	90.40 64	86.39 119	91.16 78	94.40 98	96.52 85	96.30 73	99.21 80	97.79 127

test1		93.81 90	94.18 88	93.38 95	91.34 148	95.86 144	96.22 81	88.68 115	95.23 88	90.40 64	86.39 119	91.16 78	94.40 98	96.52 85	96.30 73	99.21 80	97.79 127

FMVSNet1		91.54 119	90.93 140	92.26 103	90.35 158	95.27 166	95.22 103	87.16 132	91.37 146	87.62 97	75.45 164	83.84 130	94.43 96	96.52 85	96.30 73	98.82 125	97.74 133

DI_MVS_plusplus_trai		94.01 86	93.63 101	94.44 78	94.54 110	98.26 82	97.51 52	90.63 92	95.88 72	89.34 84	80.54 151	89.36 92	95.48 83	96.33 94	96.27 76	99.17 85	98.78 90

AdaColmap		97.53 29	96.93 44	98.24 14	99.21 23	98.77 61	98.47 33	97.34 23	96.68 50	96.52 13	95.11 46	96.12 57	98.72 15	97.19 62	96.24 77	99.17 85	98.39 110

Fast-Effi-MVS+		91.87 112	92.08 127	91.62 111	92.91 134	97.21 106	94.93 107	84.60 162	93.61 115	81.49 124	83.50 139	78.95 149	96.62 64	96.55 83	96.22 78	99.16 88	98.51 101

PLC	94.95 3	97.37 32	96.77 48	98.07 20	98.97 31	98.21 83	97.94 45	96.85 35	97.66 25	97.58 2	93.33 57	96.84 47	98.01 33	97.13 64	96.20 79	99.09 97	98.01 122
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
baseline		94.83 66	95.82 61	93.68 89	94.75 105	97.80 93	96.51 76	88.53 118	97.02 45	89.34 84	92.93 61	92.18 75	94.69 91	95.78 111	96.08 80	98.27 161	98.97 74

CPTT-MVS		97.78 25	97.54 32	98.05 21	98.91 34	99.05 31	99.00 19	96.96 33	97.14 39	95.92 17	95.50 41	98.78 27	98.99 4	97.20 60	96.07 81	98.54 150	99.04 62

Effi-MVS+		92.93 104	93.86 96	91.86 105	94.07 119	98.09 89	95.59 98	85.98 144	94.27 105	79.54 133	91.12 84	81.81 141	96.71 62	96.67 79	96.06 82	99.27 66	98.98 70

gm-plane-assit		83.26 192	85.29 189	80.89 193	89.52 170	89.89 203	70.26 208	78.24 185	77.11 205	58.01 207	74.16 175	66.90 198	90.63 153	97.20 60	96.05 83	98.66 143	95.68 171

DCV-MVSNet		94.76 72	95.12 74	94.35 80	95.10 96	95.81 148	96.46 78	89.49 108	96.33 55	90.16 69	92.55 66	90.26 86	95.83 75	95.52 117	96.03 84	99.06 103	99.33 20

IterMVS-LS		92.56 108	93.18 108	91.84 106	93.90 120	94.97 173	94.99 105	86.20 141	94.18 106	82.68 116	85.81 125	87.36 107	94.43 96	95.31 123	96.02 85	98.87 121	98.60 96
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
ACMMP		97.37 32	97.48 34	97.25 31	98.88 36	99.28 18	98.47 33	96.86 34	97.04 43	92.15 46	97.57 22	96.05 59	97.67 37	97.27 58	95.99 86	99.46 26	99.14 50
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
thres100view900		93.55 97	92.47 119	94.81 71	95.33 86	98.74 62	96.78 67	92.30 69	92.63 126	88.29 90	87.21 110	78.01 154	96.78 60	96.38 90	95.92 87	99.38 49	98.40 109

thres200		93.62 94	92.54 113	94.88 68	95.36 85	98.93 49	96.75 68	92.31 66	92.84 123	88.28 92	86.99 112	77.81 156	97.13 50	96.82 70	95.92 87	99.45 30	98.49 103

TAPA-MVS	94.18 5	96.38 47	96.49 52	96.25 44	98.26 47	98.66 68	98.00 43	94.96 43	97.17 38	89.48 79	92.91 62	96.35 52	97.53 40	96.59 81	95.90 89	99.28 64	97.82 126
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
MVSTER		94.89 65	95.07 75	94.68 75	94.71 106	96.68 120	97.00 58	90.57 93	95.18 91	93.05 38	95.21 44	86.41 111	93.72 110	97.59 51	95.88 90	99.00 109	98.50 102

tfpn200view9		93.64 93	92.57 112	94.89 67	95.33 86	98.94 47	96.82 64	92.31 66	92.63 126	88.29 90	87.21 110	78.01 154	97.12 52	96.82 70	95.85 91	99.45 30	98.56 97

thres400		93.56 96	92.43 120	94.87 69	95.40 84	98.91 52	96.70 70	92.38 65	92.93 122	88.19 94	86.69 115	77.35 157	97.13 50	96.75 75	95.85 91	99.42 42	98.56 97

Anonymous202405211			92.18 125		95.04 97	98.20 84	96.14 84	91.79 77	93.93 108		74.60 170	88.38 103	96.48 66	95.17 127	95.82 93	99.00 109	99.15 47

LGP-MVS_train		94.12 84	94.62 79	93.53 91	96.44 71	97.54 96	97.40 54	91.84 75	94.66 97	81.09 126	95.70 40	83.36 134	95.10 86	96.36 93	95.71 94	99.32 57	99.03 63

thres600view7		93.49 98	92.37 123	94.79 72	95.42 83	98.93 49	96.58 74	92.31 66	93.04 120	87.88 95	86.62 116	76.94 159	97.09 53	96.82 70	95.63 95	99.45 30	98.63 94

MSDG		94.82 67	93.73 99	96.09 47	98.34 46	97.43 101	97.06 57	96.05 37	95.84 74	90.56 62	86.30 123	89.10 97	95.55 81	96.13 102	95.61 96	99.00 109	95.73 170

EPNet_dtu		92.45 109	95.02 76	89.46 137	98.02 52	95.47 159	94.79 111	92.62 64	94.97 93	70.11 183	94.76 50	92.61 74	84.07 192	95.94 105	95.56 97	97.15 179	95.82 169
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
HyFIR lowres test		92.03 110	91.55 134	92.58 101	97.13 64	98.72 65	94.65 114	86.54 137	93.58 116	82.56 117	67.75 198	90.47 85	95.67 77	95.87 107	95.54 98	98.91 118	98.93 75

Anonymous20231211		93.49 98	92.33 124	94.84 70	94.78 104	98.00 90	96.11 85	91.85 74	94.86 95	90.91 55	74.69 169	89.18 95	96.73 61	94.82 132	95.51 99	98.67 140	99.24 32

Effi-MVS+-dtu		91.78 114	93.59 103	89.68 136	92.44 140	97.11 107	94.40 118	84.94 158	92.43 131	75.48 154	91.09 85	83.75 131	93.55 113	96.61 80	95.47 100	97.24 178	98.67 92

GG-mvs-BLEND		66.17 202	94.91 78	32.63 208	1.32 216	96.64 121	91.40 167	0.85 214	94.39 104	2.20 216	90.15 94	95.70 60	2.27 213	96.39 89	95.44 101	97.78 171	95.68 171

MIMVSNet		88.99 153	91.07 138	86.57 179	86.78 198	95.62 152	91.20 172	75.40 198	90.65 154	76.57 146	84.05 136	82.44 139	91.01 142	95.84 108	95.38 102	98.48 154	93.50 191

ET-MVSNet_ETH3D		93.34 100	94.33 86	92.18 104	83.26 204	97.66 95	96.72 69	89.89 101	95.62 80	87.17 100	96.00 36	83.69 132	96.99 55	93.78 147	95.34 103	99.06 103	98.18 120

FC-MVSNet-train		93.85 89	93.91 94	93.78 88	94.94 99	96.79 117	94.29 120	91.13 85	93.84 112	88.26 93	90.40 90	85.23 121	94.65 94	96.54 84	95.31 104	99.38 49	99.28 24

CVMVSNet		89.77 142	91.66 132	87.56 169	93.21 132	95.45 160	91.94 163	89.22 111	89.62 162	69.34 189	83.99 137	85.90 115	84.81 187	94.30 142	95.28 105	96.85 181	97.09 150

UniMVSNet_ETH3D		88.47 158	86.00 187	91.35 113	91.55 145	96.29 131	92.53 145	88.81 114	85.58 190	82.33 118	67.63 199	66.87 199	94.04 104	91.49 181	95.24 106	98.84 124	98.92 76

PatchMatch-RL		94.69 73	94.41 83	95.02 63	97.63 58	98.15 87	94.50 117	91.99 72	95.32 85	91.31 52	95.47 42	83.44 133	96.02 73	96.56 82	95.23 107	98.69 139	96.67 162

TSAR-MVS + COLMAP		94.79 69	94.51 81	95.11 61	96.50 69	97.54 96	97.99 44	94.54 44	97.81 15	85.88 105	96.73 30	81.28 144	96.99 55	96.29 95	95.21 108	98.76 135	96.73 161

test0.0.03 1		91.97 111	93.91 94	89.72 133	93.31 130	96.40 129	91.34 169	87.06 133	93.86 110	81.67 122	91.15 83	89.16 96	86.02 180	95.08 128	95.09 109	98.91 118	96.64 164

baseline2		93.01 103	94.17 90	91.64 109	92.83 136	97.49 98	93.40 131	87.53 127	93.67 114	86.07 104	91.83 75	86.58 108	91.36 136	96.38 90	95.06 110	98.67 140	98.20 119

CLD-MVS		94.79 69	94.36 85	95.30 59	95.21 92	97.46 99	97.23 55	92.24 70	96.43 53	91.77 49	92.69 64	84.31 126	96.06 71	95.52 117	95.03 111	99.31 60	99.06 58
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
FC-MVSNet-test		91.63 116	93.82 98	89.08 141	92.02 143	96.40 129	93.26 134	87.26 130	93.72 113	77.26 141	88.61 105	89.86 89	85.50 182	95.72 115	95.02 112	99.16 88	97.44 141

PMMVS		94.61 74	95.56 64	93.50 92	94.30 115	96.74 118	94.91 108	89.56 107	95.58 82	87.72 96	96.15 34	92.86 71	96.06 71	95.47 119	95.02 112	98.43 158	97.09 150

OPM-MVS		93.61 95	92.43 120	95.00 64	96.94 66	97.34 102	97.78 47	94.23 46	89.64 161	85.53 106	88.70 103	82.81 136	96.28 69	96.28 96	95.00 114	99.24 71	97.22 147

PCF-MVS	93.95 6	95.65 53	95.14 72	96.25 44	97.73 57	98.73 64	97.59 51	97.13 30	92.50 130	89.09 88	89.85 96	96.65 49	96.90 57	94.97 131	94.89 115	99.08 98	98.38 111
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
thisisatest0530		94.54 76	95.47 66	93.46 93	94.51 111	98.65 70	94.66 113	90.72 88	95.69 79	86.90 102	93.80 54	89.44 91	94.74 89	96.98 69	94.86 116	99.19 84	98.85 85

tttt0517		94.52 77	95.44 68	93.44 94	94.51 111	98.68 67	94.61 115	90.72 88	95.61 81	86.84 103	93.78 55	89.26 94	94.74 89	97.02 68	94.86 116	99.20 83	98.87 83

LTVRE_ROB	87.32 16	87.55 171	88.25 160	86.73 177	90.66 153	95.80 149	93.05 137	84.77 159	83.35 196	60.32 203	83.12 141	67.39 197	93.32 116	94.36 141	94.86 116	98.28 160	98.87 83
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
HQP-MVS		94.43 79	94.57 80	94.27 81	96.41 72	97.23 105	96.89 61	93.98 47	95.94 70	83.68 112	95.01 47	84.46 125	95.58 80	95.47 119	94.85 119	99.07 100	99.00 67

ACMP	92.88 9	94.43 79	94.38 84	94.50 77	96.01 77	97.69 94	95.85 96	92.09 71	95.74 76	89.12 87	95.14 45	82.62 138	94.77 88	95.73 113	94.67 120	99.14 91	99.06 58
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
ACMM	92.75 10	94.41 81	93.84 97	95.09 62	96.41 72	96.80 114	94.88 109	93.54 51	96.41 54	90.16 69	92.31 68	83.11 135	96.32 68	96.22 98	94.65 121	99.22 77	97.35 144
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
Fast-Effi-MVS+-dtu		91.19 122	93.64 100	88.33 149	92.19 142	96.46 126	93.99 123	81.52 179	92.59 128	71.82 173	92.17 69	85.54 117	91.68 133	95.73 113	94.64 122	98.80 130	98.34 112

TAMVS		90.54 131	90.87 142	90.16 128	91.48 146	96.61 122	93.26 134	86.08 142	87.71 176	81.66 123	83.11 142	84.04 128	90.42 155	94.54 135	94.60 123	98.04 168	95.48 174

TranMVSNet+NR-MVSNet		89.23 149	88.48 158	90.11 132	89.07 182	95.25 167	92.91 139	90.43 95	90.31 157	77.10 143	76.62 162	71.57 181	91.83 131	92.12 173	94.59 124	99.32 57	98.92 76

DU-MVS		89.67 143	88.84 154	90.63 123	89.26 176	95.61 153	92.48 146	89.91 99	91.22 147	79.57 131	77.72 159	71.18 183	93.21 119	92.53 167	94.57 125	99.35 54	99.05 60

CR-MVSNet		90.16 137	91.96 130	88.06 155	93.32 129	95.95 141	93.36 132	75.99 196	92.40 133	75.19 158	83.18 140	85.37 118	92.05 127	95.21 125	94.56 126	98.47 155	97.08 152

PatchT		89.13 151	91.71 131	86.11 183	92.92 133	95.59 155	83.64 198	75.09 199	91.87 142	75.19 158	82.63 143	85.06 123	92.05 127	95.21 125	94.56 126	97.76 172	97.08 152

ACMH	90.77 13	91.51 120	91.63 133	91.38 112	95.62 81	96.87 112	91.76 164	89.66 105	91.58 144	78.67 135	86.73 114	78.12 152	93.77 109	94.59 134	94.54 128	98.78 133	98.98 70
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
FMVSNet5		90.36 132	90.93 140	89.70 134	87.99 192	92.25 197	92.03 159	83.51 168	92.20 139	84.13 110	85.59 126	86.48 109	92.43 124	94.61 133	94.52 129	98.13 164	90.85 198

pm-mvs1		89.19 150	89.02 153	89.38 139	90.40 156	95.74 151	92.05 158	88.10 123	86.13 186	77.70 138	73.72 178	79.44 148	88.97 166	95.81 110	94.51 130	99.08 98	97.78 132

UniMVSNet_NR-MVSNet		90.35 133	89.96 146	90.80 120	89.66 166	95.83 147	92.48 146	90.53 94	90.96 151	79.57 131	79.33 155	77.14 158	93.21 119	92.91 163	94.50 131	99.37 52	99.05 60

IB-MVS	89.56 15	91.71 115	92.50 115	90.79 121	95.94 78	98.44 77	87.05 191	91.38 84	93.15 119	92.98 40	84.78 130	85.14 122	78.27 197	92.47 169	94.44 132	99.10 96	99.08 54
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
IterMVS-SCA-FT		90.24 134	92.48 118	87.63 166	92.85 135	94.30 189	93.79 125	81.47 180	92.66 125	69.95 184	84.66 132	88.38 103	89.99 160	95.39 122	94.34 133	97.74 175	97.63 136

test-mter		90.95 124	93.54 106	87.93 161	90.28 159	96.80 114	91.44 166	82.68 174	92.15 140	74.37 165	89.57 98	88.23 105	90.88 146	96.37 92	94.31 134	97.93 170	97.37 143

anonymousdsp		88.90 154	91.00 139	86.44 180	88.74 189	95.97 139	90.40 179	82.86 172	88.77 168	67.33 192	81.18 148	81.44 143	90.22 158	96.23 97	94.27 135	99.12 94	99.16 46

IterMVS		90.20 135	92.43 120	87.61 167	92.82 137	94.31 188	94.11 121	81.54 178	92.97 121	69.90 185	84.71 131	88.16 106	89.96 161	95.25 124	94.17 136	97.31 177	97.46 140
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
COLMAP_ROB	90.49 14	93.27 102	92.71 111	93.93 85	97.75 56	97.44 100	96.07 87	93.17 59	95.40 83	83.86 111	83.76 138	88.72 99	93.87 106	94.25 143	94.11 137	98.87 121	95.28 176
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
test-LLR		91.62 117	93.56 104	89.35 140	93.31 130	96.57 123	92.02 160	87.06 133	92.34 136	75.05 161	90.20 92	88.64 100	90.93 143	96.19 100	94.07 138	97.75 173	96.90 158

TESTMET0.1,1		91.07 123	93.56 104	88.17 151	90.43 155	96.57 123	92.02 160	82.83 173	92.34 136	75.05 161	90.20 92	88.64 100	90.93 143	96.19 100	94.07 138	97.75 173	96.90 158

tfpnnormal		88.50 157	87.01 178	90.23 126	91.36 147	95.78 150	92.74 141	90.09 97	83.65 195	76.33 149	71.46 190	69.58 191	91.84 130	95.54 116	94.02 140	99.06 103	99.03 63

NR-MVSNet		89.34 146	88.66 155	90.13 131	90.40 156	95.61 153	93.04 138	89.91 99	91.22 147	78.96 134	77.72 159	68.90 194	89.16 165	94.24 144	93.95 141	99.32 57	98.99 68

TransMVSNet (Re)		87.73 170	86.79 180	88.83 143	90.76 152	94.40 186	91.33 170	89.62 106	84.73 192	75.41 156	72.73 183	71.41 182	86.80 175	94.53 136	93.93 142	99.06 103	95.83 168

ACMH+	90.88 12	91.41 121	91.13 137	91.74 108	95.11 95	96.95 109	93.13 136	89.48 109	92.42 132	79.93 130	85.13 128	78.02 153	93.82 108	93.49 154	93.88 143	98.94 115	97.99 123

UniMVSNet (Re)		90.03 140	89.61 149	90.51 124	89.97 164	96.12 135	92.32 150	89.26 110	90.99 150	80.95 127	78.25 158	75.08 166	91.14 139	93.78 147	93.87 144	99.41 43	99.21 37

RPMNet		90.19 136	92.03 129	88.05 156	93.46 126	95.95 141	93.41 130	74.59 201	92.40 133	75.91 152	84.22 135	86.41 111	92.49 123	94.42 139	93.85 145	98.44 156	96.96 155

Baseline_NR-MVSNet		89.27 148	88.01 164	90.73 122	89.26 176	93.71 192	92.71 143	89.78 104	90.73 152	81.28 125	73.53 179	72.85 175	92.30 126	92.53 167	93.84 146	99.07 100	98.88 81

testgi		89.42 144	91.50 135	87.00 176	92.40 141	95.59 155	89.15 185	85.27 155	92.78 124	72.42 170	91.75 76	76.00 162	84.09 191	94.38 140	93.82 147	98.65 144	96.15 165

GA-MVS		89.28 147	90.75 143	87.57 168	91.77 144	96.48 125	92.29 152	87.58 126	90.61 155	65.77 194	84.48 133	76.84 160	89.46 163	95.84 108	93.68 148	98.52 151	97.34 145

CP-MVSNet		87.89 168	87.27 174	88.62 145	89.30 174	95.06 170	90.60 177	85.78 146	87.43 180	75.98 151	74.60 170	68.14 196	90.76 148	93.07 161	93.60 149	99.30 62	98.98 70

PS-CasMVS		87.33 175	86.68 183	88.10 152	89.22 181	94.93 175	90.35 180	85.70 147	86.44 185	74.01 166	73.43 180	66.59 202	90.04 159	92.92 162	93.52 150	99.28 64	98.91 79

PEN-MVS		87.22 177	86.50 185	88.07 153	88.88 185	94.44 185	90.99 174	86.21 139	86.53 184	73.66 167	74.97 167	66.56 203	89.42 164	91.20 183	93.48 151	99.24 71	98.31 116

USDC		90.69 127	90.52 144	90.88 118	94.17 117	96.43 127	95.82 97	86.76 135	93.92 109	76.27 150	86.49 117	74.30 169	93.67 112	95.04 130	93.36 152	98.61 146	94.13 183

EG-PatchMatch MVS		86.68 179	87.24 175	86.02 184	90.58 154	96.26 132	91.08 173	81.59 177	84.96 191	69.80 187	71.35 191	75.08 166	84.23 190	94.24 144	93.35 153	98.82 125	95.46 175

WR-MVS		87.93 165	88.09 162	87.75 163	89.26 176	95.28 164	90.81 175	86.69 136	88.90 165	75.29 157	74.31 174	73.72 172	85.19 185	92.26 170	93.32 154	99.27 66	98.81 88

thisisatest0515		90.12 138	92.06 128	87.85 162	90.03 162	96.17 134	87.83 188	87.45 128	91.71 143	77.15 142	85.40 127	84.01 129	85.74 181	95.41 121	93.30 155	98.88 120	98.43 105

TinyColmap		89.42 144	88.58 156	90.40 125	93.80 124	95.45 160	93.96 124	86.54 137	92.24 138	76.49 147	80.83 149	70.44 186	93.37 115	94.45 138	93.30 155	98.26 162	93.37 193

DTE-MVSNet		86.67 180	86.09 186	87.35 172	88.45 191	94.08 190	90.65 176	86.05 143	86.13 186	72.19 171	74.58 172	66.77 201	87.61 172	90.31 186	93.12 157	99.13 92	97.62 137

WR-MVS_H		87.93 165	87.85 168	88.03 158	89.62 167	95.58 157	90.47 178	85.55 149	87.20 181	76.83 145	74.42 173	72.67 177	86.37 177	93.22 158	93.04 158	99.33 55	98.83 87

MS-PatchMatch		91.82 113	92.51 114	91.02 115	95.83 79	96.88 110	95.05 104	84.55 164	93.85 111	82.01 119	82.51 144	91.71 76	90.52 154	95.07 129	93.03 159	98.13 164	94.52 178

v7n		86.43 181	86.52 184	86.33 181	87.91 193	94.93 175	90.15 181	83.05 170	86.57 183	70.21 182	71.48 189	66.78 200	87.72 170	94.19 146	92.96 160	98.92 117	98.76 91

v2v482		88.25 161	87.71 171	88.88 142	89.23 180	95.28 164	92.10 156	87.89 125	88.69 169	73.31 168	75.32 165	71.64 180	91.89 129	92.10 175	92.92 161	98.86 123	97.99 123

v10		88.00 163	87.96 165	88.05 156	89.44 171	94.68 180	92.36 149	83.35 169	89.37 163	72.96 169	73.98 176	72.79 176	91.35 137	93.59 149	92.88 162	98.81 128	98.42 107

pmmvs5		87.83 169	88.09 162	87.51 171	89.59 169	95.48 158	89.75 183	84.73 160	86.07 188	71.44 175	80.57 150	70.09 189	90.74 150	94.47 137	92.87 163	98.82 125	97.10 149

v1192		87.51 172	87.31 173	87.74 164	89.04 183	94.87 178	92.07 157	85.03 156	88.49 171	70.32 180	72.65 184	70.35 187	91.21 138	93.59 149	92.80 164	98.78 133	98.42 107

v1144		87.92 167	87.79 169	88.07 153	89.27 175	95.15 169	92.17 155	85.62 148	88.52 170	71.52 174	73.80 177	72.40 178	91.06 141	93.54 153	92.80 164	98.81 128	98.33 113

V42		88.31 160	87.95 166	88.73 144	89.44 171	95.34 163	92.23 154	87.21 131	88.83 166	74.49 164	74.89 168	73.43 174	90.41 157	92.08 176	92.77 166	98.60 148	98.33 113

v8		88.21 162	87.94 167	88.51 146	89.62 167	95.01 172	92.31 151	84.99 157	88.94 164	74.70 163	75.03 166	73.51 173	90.67 151	92.11 174	92.74 167	98.80 130	98.24 117

v1240		86.89 178	86.75 182	87.06 175	88.75 188	94.65 182	91.30 171	84.05 165	87.49 179	68.94 190	71.96 188	68.86 195	90.65 152	93.33 156	92.72 168	98.67 140	98.24 117

v1921920		87.31 176	87.13 177	87.52 170	88.87 186	94.72 179	91.96 162	84.59 163	88.28 172	69.86 186	72.50 185	70.03 190	91.10 140	93.33 156	92.61 169	98.71 137	98.44 104

v144192		87.40 174	87.20 176	87.64 165	88.89 184	94.88 177	91.65 165	84.70 161	87.80 175	71.17 178	73.20 182	70.91 184	90.75 149	92.69 165	92.49 170	98.71 137	98.43 105

Anonymous20231206		83.84 191	85.19 190	82.26 192	87.38 196	92.87 194	85.49 195	83.65 167	86.07 188	63.44 199	68.42 195	69.01 193	75.45 200	93.34 155	92.44 171	98.12 166	94.20 182

pmmvs6		85.98 184	84.89 192	87.25 173	88.83 187	94.35 187	89.36 184	85.30 154	78.51 204	75.44 155	62.71 203	75.41 163	87.65 171	93.58 151	92.40 172	96.89 180	97.29 146

v148		87.51 172	86.79 180	88.36 148	89.39 173	95.21 168	89.84 182	88.20 122	87.61 178	77.56 139	73.38 181	70.32 188	86.80 175	90.70 185	92.31 173	98.37 159	97.98 125

EU-MVSNet		85.62 185	87.65 172	83.24 191	88.54 190	92.77 196	87.12 190	85.32 152	86.71 182	64.54 196	78.52 157	75.11 165	78.35 196	92.25 171	92.28 174	95.58 190	95.93 167

pmmvs4		90.55 130	89.91 147	91.30 114	90.26 160	94.95 174	92.73 142	87.94 124	93.44 118	85.35 107	82.28 145	76.09 161	93.02 121	93.56 152	92.26 175	98.51 152	96.77 160

MVS-HIRNet		85.36 186	86.89 179	83.57 189	90.13 161	94.51 184	83.57 199	72.61 203	88.27 173	71.22 177	68.97 194	81.81 141	88.91 167	93.08 160	91.94 176	94.97 198	89.64 201

TDRefinement		89.07 152	88.15 161	90.14 130	95.16 93	96.88 110	95.55 100	90.20 96	89.68 160	76.42 148	76.67 161	74.30 169	84.85 186	93.11 159	91.91 177	98.64 145	94.47 179

SixPastTwentyTwo		88.37 159	89.47 150	87.08 174	90.01 163	95.93 143	87.41 189	85.32 152	90.26 159	70.26 181	86.34 122	71.95 179	90.93 143	92.89 164	91.72 178	98.55 149	97.22 147

MIMVSNet1		80.03 196	80.93 197	78.97 197	72.46 210	90.73 201	80.81 203	82.44 175	80.39 201	63.64 198	57.57 204	64.93 204	76.37 198	91.66 179	91.55 179	98.07 167	89.70 200

test20.03		82.92 193	85.52 188	79.90 196	87.75 194	91.84 198	82.80 200	82.99 171	82.65 200	60.32 203	78.90 156	70.50 185	67.10 203	92.05 177	90.89 180	98.44 156	91.80 196

MDTV_nov1_ep13		91.57 118	93.18 108	89.70 134	93.39 128	96.97 108	93.53 128	80.91 181	95.70 77	81.86 120	92.40 67	89.93 88	93.25 118	91.97 178	90.80 181	95.25 195	94.46 180

tpm		87.95 164	89.44 151	86.21 182	92.53 139	94.62 183	91.40 167	76.36 193	91.46 145	69.80 187	87.43 109	75.14 164	91.55 134	89.85 191	90.60 182	95.61 189	96.96 155

RPSCF		94.05 85	94.00 93	94.12 83	96.20 74	96.41 128	96.61 72	91.54 80	95.83 75	89.73 76	96.94 29	92.80 72	95.35 85	91.63 180	90.44 183	95.27 194	93.94 186

pmmvs-eth3d		84.33 190	82.94 195	85.96 185	84.16 201	90.94 200	86.55 192	83.79 166	84.25 193	75.85 153	70.64 192	56.43 209	87.44 174	92.20 172	90.41 184	97.97 169	95.68 171

EPMVS		90.88 126	92.12 126	89.44 138	94.71 106	97.24 104	93.55 127	76.81 190	95.89 71	81.77 121	91.49 79	86.47 110	93.87 106	90.21 187	90.07 185	95.92 185	93.49 192

SCA		90.92 125	93.04 110	88.45 147	93.72 125	97.33 103	92.77 140	76.08 195	96.02 66	78.26 137	91.96 72	90.86 81	93.99 105	90.98 184	90.04 186	95.88 186	94.06 185

PM-MVS		84.72 189	84.47 193	85.03 186	84.67 200	91.57 199	86.27 193	82.31 176	87.65 177	70.62 179	76.54 163	56.41 210	88.75 168	92.59 166	89.85 187	97.54 176	96.66 163

ADS-MVSNet		89.80 141	91.33 136	88.00 159	94.43 113	96.71 119	92.29 152	74.95 200	96.07 65	77.39 140	88.67 104	86.09 113	93.26 117	88.44 193	89.57 188	95.68 188	93.81 189

PatchmatchNet		90.56 129	92.49 116	88.31 150	93.83 123	96.86 113	92.42 148	76.50 192	95.96 69	78.31 136	91.96 72	89.66 90	93.48 114	90.04 189	89.20 189	95.32 192	93.73 190
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
CostFormer		90.69 127	90.48 145	90.93 117	94.18 116	96.08 136	94.03 122	78.20 186	93.47 117	89.96 73	90.97 86	80.30 145	93.72 110	87.66 197	88.75 190	95.51 191	96.12 166

pmmvs3		79.16 197	80.12 199	78.05 199	79.36 205	86.59 206	78.13 206	73.87 202	76.42 206	57.51 208	70.59 193	57.02 208	84.66 188	90.10 188	88.32 191	94.75 200	91.77 197

MDTV_nov1_ep13_2view		86.30 182	88.27 159	84.01 188	87.71 195	94.67 181	88.08 187	76.78 191	90.59 156	68.66 191	80.46 152	80.12 146	87.58 173	89.95 190	88.20 192	95.25 195	93.90 188

CMPMVS	65.18 17	84.76 188	83.10 194	86.69 178	95.29 89	95.05 171	88.37 186	85.51 150	80.27 202	71.31 176	68.37 196	73.85 171	85.25 183	87.72 195	87.75 193	94.38 202	88.70 202
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
dps		90.11 139	89.37 152	90.98 116	93.89 121	96.21 133	93.49 129	77.61 188	91.95 141	92.74 44	88.85 101	78.77 151	92.37 125	87.71 196	87.71 194	95.80 187	94.38 181

new_pmnet		81.53 194	82.68 196	80.20 194	83.47 203	89.47 204	82.21 202	78.36 184	87.86 174	60.14 205	67.90 197	69.43 192	82.03 194	89.22 192	87.47 195	94.99 197	87.39 203

tpmrst		88.86 156	89.62 148	87.97 160	94.33 114	95.98 138	92.62 144	76.36 193	94.62 99	76.94 144	85.98 124	82.80 137	92.80 122	86.90 198	87.15 196	94.77 199	93.93 187

N_pmnet		84.80 187	85.10 191	84.45 187	89.25 179	92.86 195	84.04 197	86.21 139	88.78 167	66.73 193	72.41 186	74.87 168	85.21 184	88.32 194	86.45 197	95.30 193	92.04 195

MDA-MVSNet-bldmvs		80.11 195	80.24 198	79.94 195	77.01 207	93.21 193	78.86 205	85.94 145	82.71 199	60.86 200	79.71 154	51.77 212	83.71 193	75.60 204	86.37 198	93.28 203	92.35 194

tpm cat1		88.90 154	87.78 170	90.22 127	93.88 122	95.39 162	93.79 125	78.11 187	92.55 129	89.43 80	81.31 147	79.84 147	91.40 135	84.95 199	86.34 199	94.68 201	94.09 184

Gipuma		68.35 200	66.71 203	70.27 201	74.16 209	68.78 211	63.93 211	71.77 205	83.34 197	54.57 209	34.37 208	31.88 214	68.69 202	83.30 201	85.53 200	88.48 207	79.78 207
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
ambc			73.83 202		76.23 208	85.13 207	82.27 201		84.16 194	65.58 195	52.82 206	23.31 217	73.55 201	91.41 182	85.26 201	92.97 204	94.70 177

new-patchmatchnet		78.49 198	78.19 200	78.84 198	84.13 202	90.06 202	77.11 207	80.39 182	79.57 203	59.64 206	66.01 200	55.65 211	75.62 199	84.55 200	80.70 202	96.14 184	90.77 199

PMMVS2		64.36 203	65.94 205	62.52 204	67.37 211	77.44 209	64.39 210	69.32 209	61.47 209	34.59 211	46.09 207	41.03 213	48.02 210	74.56 206	78.23 203	91.43 205	82.76 205

tmp_tt				66.88 203	86.07 199	73.86 210	68.22 209	33.38 211	96.88 46	80.67 128	88.23 107	78.82 150	49.78 208	82.68 202	77.47 204	83.19 210

MVE	50.86 19	49.54 206	51.43 206	47.33 207	44.14 214	59.20 213	36.45 215	60.59 210	41.47 212	31.14 212	29.58 209	17.06 218	48.52 209	62.22 208	74.63 205	63.12 213	75.87 209
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
FPMVS		75.84 199	74.59 201	77.29 200	86.92 197	83.89 208	85.01 196	80.05 183	82.91 198	60.61 202	65.25 201	60.41 206	63.86 204	75.60 204	73.60 206	87.29 208	80.47 206

PMVS	63.12 18	67.27 201	66.39 204	68.30 202	77.98 206	60.24 212	59.53 212	76.82 189	66.65 208	60.74 201	54.39 205	59.82 207	51.24 207	73.92 207	70.52 207	83.48 209	79.17 208
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
E-PMN		50.67 204	47.85 207	53.96 205	64.13 213	50.98 215	38.06 213	69.51 207	51.40 211	24.60 213	29.46 211	24.39 216	56.07 206	48.17 209	59.70 208	71.40 211	70.84 210

EMVS		49.98 205	46.76 208	53.74 206	64.96 212	51.29 214	37.81 214	69.35 208	51.83 210	22.69 214	29.57 210	25.06 215	57.28 205	44.81 210	56.11 209	70.32 212	68.64 211

testmvs		12.09 207	16.94 209	6.42 209	3.15 215	6.08 216	9.51 217	3.84 212	21.46 213	5.31 215	27.49 212	6.76 219	10.89 211	17.06 211	15.01 210	5.84 214	24.75 212

test123		9.58 208	13.53 210	4.97 210	1.31 217	5.47 217	8.32 218	2.95 213	18.14 214	2.03 217	20.82 213	2.34 220	10.60 212	10.00 212	14.16 211	4.60 215	23.77 213

uanet_test		0.00 209	0.00 211	0.00 211	0.00 218	0.00 218	0.00 219	0.00 215	0.00 215	0.00 218	0.00 214	0.00 221	0.00 214	0.00 213	0.00 212	0.00 216	0.00 214

sosnet-low-res		0.00 209	0.00 211	0.00 211	0.00 218	0.00 218	0.00 219	0.00 215	0.00 215	0.00 218	0.00 214	0.00 221	0.00 214	0.00 213	0.00 212	0.00 216	0.00 214

sosnet		0.00 209	0.00 211	0.00 211	0.00 218	0.00 218	0.00 219	0.00 215	0.00 215	0.00 218	0.00 214	0.00 221	0.00 214	0.00 213	0.00 212	0.00 216	0.00 214

9.14												99.28 10

SR-MVS					99.45 9			97.61 14				99.20 14

our_test_3					89.78 165	93.84 191	85.59 194

test_part1																	99.38 15

MTAPA										96.83 9		99.12 19

MTMP										97.18 4		98.83 25

Patchmatch-RL test							34.61 216

XVS					96.60 67	99.35 11	96.82 64			90.85 56		98.72 28				99.46 26

X-MVStestdata					96.60 67	99.35 11	96.82 64			90.85 56		98.72 28				99.46 26

abl_6				96.82 39	98.60 41	98.74 62	97.74 48	93.73 49	96.25 57	94.37 28	94.55 51	98.60 33	97.25 46			99.27 66	98.61 95

mPP-MVS					99.21 23							98.29 37

NP-MVS									95.32 85

Patchmtry						95.96 140	93.36 132	75.99 196		75.19 158

DeepMVS_CX						86.86 205	79.50 204	70.43 206	90.73 152	63.66 197	80.36 153	60.83 205	79.68 195	76.23 203		89.46 206	86.53 204

High-resolution multi-view benchmark