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.

ESAPD		97.83 1	98.13 1	97.48 2	98.83 19	99.19 1	98.99 1	96.70 2	96.05 14	94.39 6	98.30 1	99.47 2	97.02 5	97.75 5	97.02 10	98.98 1	99.10 5

HSP-MVS		97.51 4	97.70 5	97.29 4	99.00 11	99.17 2	98.61 4	96.41 6	95.88 16	94.34 8	97.72 3	99.04 5	96.93 9	97.29 14	95.90 37	98.45 24	98.94 8

v1.0		90.03 86	83.83 157	97.27 5	99.12 3	99.14 3	98.66 3	96.80 1	95.74 17	93.46 15	97.72 3	99.48 1	96.76 15	97.77 3	96.92 14	98.83 5	0.00 246

APDe-MVS		97.79 2	97.96 2	97.60 1	99.20 2	99.10 4	98.88 2	96.68 3	96.81 3	94.64 4	97.84 2	98.02 9	97.24 2	97.74 6	97.02 10	98.97 2	99.16 2

CSCG		95.68 27	95.46 32	95.93 25	98.71 21	99.07 5	97.13 32	93.55 33	95.48 22	93.35 17	90.61 41	93.82 41	95.16 32	94.60 74	95.57 42	97.70 105	99.08 6

SMA-MVS		97.53 3	97.93 3	97.07 8	99.21 1	99.02 6	98.08 18	96.25 8	96.36 8	93.57 12	96.56 11	99.27 3	96.78 14	97.91 2	97.43 3	98.51 14	98.94 8

ACMMP_Plus		96.93 13	97.27 11	96.53 21	99.06 6	98.95 7	98.24 12	96.06 12	95.66 19	90.96 31	95.63 21	97.71 13	96.53 19	97.66 8	96.68 17	98.30 48	98.61 16

SteuartSystems-ACMMP		97.10 11	97.49 7	96.65 16	98.97 13	98.95 7	98.43 7	95.96 14	95.12 26	91.46 26	96.85 7	97.60 15	96.37 23	97.76 4	97.16 7	98.68 7	98.97 7
Skip Steuart: Steuart Systems R&D Blog.
ACMMPR		96.92 14	96.96 15	96.87 13	98.99 12	98.78 9	98.38 9	95.52 21	96.57 6	92.81 22	96.06 17	95.90 31	97.07 4	96.60 28	96.34 28	98.46 21	98.42 28

HFP-MVS		97.11 10	97.19 12	97.00 10	98.97 13	98.73 10	98.37 10	95.69 18	96.60 5	93.28 18	96.87 6	96.64 24	97.27 1	96.64 26	96.33 29	98.44 25	98.56 17

zzz-MVS		96.98 12	96.68 20	97.33 3	99.09 4	98.71 11	98.43 7	96.01 13	96.11 13	95.19 3	92.89 30	97.32 19	96.84 10	97.20 15	96.09 34	98.44 25	98.46 27

XVS					95.68 58	98.66 12	94.96 57			88.03 49		96.06 27				98.46 21

X-MVStestdata					95.68 58	98.66 12	94.96 57			88.03 49		96.06 27				98.46 21

X-MVS		96.07 23	96.33 25	95.77 27	98.94 15	98.66 12	97.94 22	95.41 26	95.12 26	88.03 49	93.00 29	96.06 27	95.85 25	96.65 25	96.35 26	98.47 19	98.48 24

SD-MVS		97.35 5	97.73 4	96.90 12	97.35 40	98.66 12	97.85 24	96.25 8	96.86 2	94.54 5	96.75 9	99.13 4	96.99 6	96.94 21	96.58 20	98.39 34	99.20 1

PHI-MVS		95.86 25	96.93 18	94.61 38	97.60 38	98.65 16	96.49 37	93.13 36	94.07 40	87.91 52	97.12 5	97.17 20	93.90 47	96.46 31	96.93 13	98.64 9	98.10 43

PGM-MVS		96.16 21	96.33 25	95.95 24	99.04 7	98.63 17	98.32 11	92.76 38	93.42 46	90.49 36	96.30 13	95.31 36	96.71 17	96.46 31	96.02 35	98.38 35	98.19 36

APD-MVS		97.12 9	97.05 14	97.19 6	99.04 7	98.63 17	98.45 6	96.54 4	94.81 33	93.50 13	96.10 16	97.40 18	96.81 11	97.05 18	96.82 16	98.80 6	98.56 17
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
CNVR-MVS		97.30 7	97.41 8	97.18 7	99.02 10	98.60 19	98.15 15	96.24 10	96.12 12	94.10 9	95.54 22	97.99 10	96.99 6	97.97 1	97.17 6	98.57 12	98.50 23

CP-MVS		96.68 17	96.59 23	96.77 15	98.85 18	98.58 20	98.18 14	95.51 22	95.34 23	92.94 21	95.21 25	96.25 26	96.79 13	96.44 33	95.77 39	98.35 37	98.56 17

TSAR-MVS + MP.		97.31 6	97.64 6	96.92 11	97.28 42	98.56 21	98.61 4	95.48 24	96.72 4	94.03 11	96.73 10	98.29 7	97.15 3	97.61 10	96.42 23	98.96 3	99.13 3

HPM-MVS++		97.22 8	97.40 9	97.01 9	99.08 5	98.55 22	98.19 13	96.48 5	96.02 15	93.28 18	96.26 14	98.71 6	96.76 15	97.30 13	96.25 31	98.30 48	98.68 11

DeepC-MVS	92.10 3	95.22 31	94.77 35	95.75 28	97.77 34	98.54 23	97.63 26	95.96 14	95.07 29	88.85 44	85.35 67	91.85 48	95.82 26	96.88 23	97.10 8	98.44 25	98.63 13
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
ACMMP		95.54 28	95.49 31	95.61 30	98.27 27	98.53 24	97.16 31	94.86 28	94.88 32	89.34 40	95.36 24	91.74 49	95.50 30	95.51 48	94.16 61	98.50 17	98.22 34
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
CANet		94.85 34	94.92 34	94.78 34	97.25 43	98.52 25	97.20 29	91.81 44	93.25 47	91.06 30	86.29 58	94.46 39	92.99 56	97.02 19	96.68 17	98.34 39	98.20 35

MVS_0304		94.30 41	94.68 36	93.86 46	96.33 54	98.48 26	97.41 27	91.20 50	92.75 50	86.96 59	86.03 61	93.81 42	92.64 62	96.89 22	96.54 22	98.61 11	98.24 33

MP-MVS		96.56 18	96.72 19	96.37 22	98.93 16	98.48 26	98.04 19	95.55 20	94.32 37	90.95 33	95.88 19	97.02 21	96.29 24	96.77 24	96.01 36	98.47 19	98.56 17
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
MCST-MVS		96.83 15	97.06 13	96.57 17	98.88 17	98.47 28	98.02 20	96.16 11	95.58 21	90.96 31	95.78 20	97.84 12	96.46 21	97.00 20	96.17 33	98.94 4	98.55 22

MVS_111021_HR		94.84 35	95.91 27	93.60 48	97.35 40	98.46 29	95.08 56	91.19 51	94.18 39	85.97 64	95.38 23	92.56 46	93.61 50	96.61 27	96.25 31	98.40 32	97.92 49

NCCC		96.75 16	96.67 21	96.85 14	99.03 9	98.44 30	98.15 15	96.28 7	96.32 9	92.39 23	92.16 32	97.55 16	96.68 18	97.32 11	96.65 19	98.55 13	98.26 32

TSAR-MVS + ACMM		96.19 20	97.39 10	94.78 34	97.70 36	98.41 31	97.72 25	95.49 23	96.47 7	86.66 61	96.35 12	97.85 11	93.99 44	97.19 16	96.37 25	97.12 127	99.13 3

DELS-MVS		93.71 44	93.47 44	94.00 41	96.82 49	98.39 32	96.80 35	91.07 53	89.51 89	89.94 39	83.80 84	89.29 63	90.95 85	97.32 11	97.65 2	98.42 28	98.32 31
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
QAPM		94.13 42	94.33 41	93.90 44	97.82 33	98.37 33	96.47 38	90.89 55	92.73 51	85.63 70	85.35 67	93.87 40	94.17 42	95.71 46	95.90 37	98.40 32	98.42 28

DeepC-MVS_fast	93.32 1	96.48 19	96.42 24	96.56 18	98.70 22	98.31 34	97.97 21	95.76 17	96.31 10	92.01 25	91.43 37	95.42 35	96.46 21	97.65 9	97.69 1	98.49 18	98.12 41
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
DeepPCF-MVS	92.65 2	95.50 30	96.96 15	93.79 47	96.44 52	98.21 35	93.51 90	94.08 32	96.94 1	89.29 41	93.08 28	96.77 23	93.82 49	97.68 7	97.40 4	95.59 192	98.65 12

3Dnovator	90.28 7	94.70 38	94.34 40	95.11 32	98.06 29	98.21 35	96.89 34	91.03 54	94.72 34	91.45 27	82.87 88	93.10 44	94.61 36	96.24 39	97.08 9	98.63 10	98.16 37

MSLP-MVS++		96.05 24	95.63 28	96.55 19	98.33 26	98.17 37	96.94 33	94.61 30	94.70 35	94.37 7	89.20 47	95.96 30	96.81 11	95.57 47	97.33 5	98.24 60	98.47 25

3Dnovator+	90.56 5	95.06 32	94.56 37	95.65 29	98.11 28	98.15 38	97.19 30	91.59 48	95.11 28	93.23 20	81.99 97	94.71 38	95.43 31	96.48 30	96.88 15	98.35 37	98.63 13

TSAR-MVS + GP.		95.86 25	96.95 17	94.60 39	94.07 81	98.11 39	96.30 40	91.76 46	95.67 18	91.07 29	96.82 8	97.69 14	95.71 28	95.96 42	95.75 40	98.68 7	98.63 13

CDPH-MVS		94.80 37	95.50 30	93.98 43	98.34 25	98.06 40	97.41 27	93.23 35	92.81 49	82.98 88	92.51 31	94.82 37	93.53 51	96.08 41	96.30 30	98.42 28	97.94 47

train_agg		96.15 22	96.64 22	95.58 31	98.44 24	98.03 41	98.14 17	95.40 27	93.90 43	87.72 53	96.26 14	98.10 8	95.75 27	96.25 38	95.45 44	98.01 86	98.47 25

OpenMVS	88.18 11	92.51 53	91.61 71	93.55 49	97.74 35	98.02 42	95.66 50	90.46 58	89.14 91	86.50 62	75.80 132	90.38 61	92.69 61	94.99 53	95.30 45	98.27 55	97.63 59

abl_6				94.78 34	97.46 39	97.99 43	95.76 48	91.80 45	93.72 44	91.25 28	91.33 38	96.47 25	94.28 41			98.14 68	97.39 68

CPTT-MVS		95.54 28	95.07 33	96.10 23	97.88 32	97.98 44	97.92 23	94.86 28	94.56 36	92.16 24	91.01 39	95.71 32	96.97 8	94.56 75	93.50 83	96.81 167	98.14 39

PCF-MVS	90.19 8	92.98 49	92.07 64	94.04 40	96.39 53	97.87 45	96.03 44	95.47 25	87.16 109	85.09 79	84.81 76	93.21 43	93.46 53	91.98 131	91.98 126	97.78 98	97.51 63
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
canonicalmvs		93.08 48	93.09 48	93.07 58	94.24 74	97.86 46	95.45 54	87.86 102	94.00 42	87.47 55	88.32 50	82.37 94	95.13 33	93.96 91	96.41 24	98.27 55	98.73 10

PVSNet_Blended_VisFu		91.92 61	92.39 59	91.36 81	95.45 66	97.85 47	92.25 110	89.54 74	88.53 99	87.47 55	79.82 106	90.53 58	85.47 157	96.31 37	95.16 48	97.99 88	98.56 17

PVSNet_BlendedMVS		92.80 50	92.44 57	93.23 52	96.02 56	97.83 48	93.74 85	90.58 56	91.86 57	90.69 34	85.87 64	82.04 96	90.01 93	96.39 34	95.26 46	98.34 39	97.81 55

PVSNet_Blended		92.80 50	92.44 57	93.23 52	96.02 56	97.83 48	93.74 85	90.58 56	91.86 57	90.69 34	85.87 64	82.04 96	90.01 93	96.39 34	95.26 46	98.34 39	97.81 55

AdaColmap		95.02 33	93.71 42	96.54 20	98.51 23	97.76 50	96.69 36	95.94 16	93.72 44	93.50 13	89.01 48	90.53 58	96.49 20	94.51 77	93.76 73	98.07 79	96.69 90

IS_MVSNet		91.87 62	93.35 46	90.14 97	94.09 80	97.73 51	93.09 97	88.12 94	88.71 95	79.98 107	84.49 77	90.63 57	87.49 121	97.07 17	96.96 12	98.07 79	97.88 54

OMC-MVS		94.49 39	94.36 39	94.64 37	97.17 44	97.73 51	95.49 53	92.25 40	96.18 11	90.34 37	88.51 49	92.88 45	94.90 35	94.92 57	94.17 60	97.69 106	96.15 116

MVS_111021_LR		94.84 35	95.57 29	94.00 41	97.11 45	97.72 53	94.88 59	91.16 52	95.24 25	88.74 45	96.03 18	91.52 52	94.33 40	95.96 42	95.01 49	97.79 97	97.49 64

casdiffmvs1		93.20 47	93.17 47	93.25 51	94.35 73	97.64 54	95.59 52	87.34 110	94.26 38	90.22 38	89.46 44	85.25 75	93.90 47	92.68 115	94.94 51	98.11 72	97.92 49

TAPA-MVS	90.35 6	93.69 45	93.52 43	93.90 44	96.89 48	97.62 55	96.15 41	91.67 47	94.94 30	85.97 64	87.72 52	91.96 47	94.40 37	93.76 93	93.06 103	98.30 48	95.58 134
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
Vis-MVSNet		89.36 97	91.49 73	86.88 133	92.10 123	97.60 56	92.16 116	85.89 119	84.21 143	75.20 126	82.58 92	87.13 64	77.40 206	95.90 44	95.63 41	98.51 14	97.36 69
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
UA-Net		90.81 74	92.58 54	88.74 111	94.87 70	97.44 57	92.61 102	88.22 92	82.35 156	78.93 112	85.20 70	95.61 33	79.56 199	96.52 29	96.57 21	98.23 61	94.37 161

casdiffmvs		92.13 57	91.95 67	92.34 63	93.87 92	97.44 57	94.36 64	86.99 113	92.00 55	88.04 48	87.23 53	81.81 99	92.73 59	93.78 92	94.06 66	98.03 83	97.30 73

CNLPA		93.69 45	92.50 55	95.06 33	97.11 45	97.36 59	93.88 81	93.30 34	95.64 20	93.44 16	80.32 104	90.73 56	94.99 34	93.58 99	93.33 87	97.67 108	96.57 100

conf0.002		89.25 103	87.21 124	91.62 69	93.87 92	97.35 60	94.31 68	89.83 63	85.87 120	81.62 91	78.72 112	63.89 203	91.76 71	94.90 64	93.98 68	98.33 43	95.77 127

conf0.01		89.34 99	87.39 123	91.61 70	93.88 91	97.34 61	94.31 68	89.82 65	85.87 120	81.53 92	77.93 116	66.15 186	91.76 71	94.90 64	93.51 77	98.32 44	96.05 120

EPP-MVSNet		92.13 57	93.06 49	91.05 86	93.66 100	97.30 62	92.18 113	87.90 98	90.24 75	83.63 83	86.14 60	90.52 60	90.76 87	94.82 67	94.38 57	98.18 66	97.98 45

tfpn111		90.16 85	88.99 94	91.52 75	93.90 87	97.26 63	94.31 68	89.75 66	85.87 120	81.10 98	84.41 78	70.38 146	91.76 71	94.92 57	93.51 77	98.29 52	96.61 93

conf200view11		89.55 93	87.86 110	91.52 75	93.90 87	97.26 63	94.31 68	89.75 66	85.87 120	81.10 98	76.46 126	70.38 146	91.76 71	94.92 57	93.51 77	98.29 52	96.61 93

tfpn200view9		89.55 93	87.86 110	91.53 73	93.90 87	97.26 63	94.31 68	89.74 69	85.87 120	81.15 96	76.46 126	70.38 146	91.76 71	94.92 57	93.51 77	98.28 54	96.61 93

thres600view7		89.28 102	87.47 121	91.39 78	94.12 78	97.25 66	93.94 80	89.74 69	85.62 130	80.63 103	75.24 139	69.33 154	91.66 77	94.92 57	93.23 91	98.27 55	96.72 89

thres200		89.49 95	87.72 113	91.55 72	93.95 84	97.25 66	94.34 66	89.74 69	85.66 128	81.18 95	76.12 131	70.19 151	91.80 69	94.92 57	93.51 77	98.27 55	96.40 104

view600		89.29 101	87.48 120	91.41 77	94.10 79	97.21 68	93.96 77	89.70 72	85.67 127	80.75 102	75.29 136	69.35 153	91.70 76	94.92 57	93.23 91	98.26 59	96.69 90

thres400		89.40 96	87.58 118	91.53 73	94.06 82	97.21 68	94.19 76	89.83 63	85.69 126	81.08 100	75.50 134	69.76 152	91.80 69	94.79 69	93.51 77	98.20 64	96.60 98

view800		89.21 104	87.44 122	91.27 82	94.13 76	97.18 70	93.74 85	89.53 75	85.60 131	80.34 105	75.29 136	68.89 155	91.57 78	94.97 54	93.36 86	98.34 39	96.79 87

tfpn		88.67 107	86.57 128	91.12 85	94.14 75	97.15 71	93.51 90	89.37 76	85.49 132	79.91 108	75.26 138	62.24 209	91.39 79	95.00 52	93.95 69	98.41 30	96.88 84

UGNet		91.52 68	93.41 45	89.32 104	94.13 76	97.15 71	91.83 123	89.01 83	90.62 69	85.86 68	86.83 54	91.73 50	77.40 206	94.68 71	94.43 56	97.71 103	98.40 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
MAR-MVS		92.71 52	92.63 53	92.79 60	97.70 36	97.15 71	93.75 84	87.98 96	90.71 67	85.76 69	86.28 59	86.38 67	94.35 39	94.95 55	95.49 43	97.22 121	97.44 66
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
LS3D		91.97 60	90.98 76	93.12 56	97.03 47	97.09 74	95.33 55	95.59 19	92.47 52	79.26 111	81.60 100	82.77 89	94.39 38	94.28 80	94.23 59	97.14 126	94.45 160

tttt0517		91.01 73	91.71 69	90.19 95	92.98 110	97.07 75	91.96 122	87.63 107	90.61 70	81.42 94	86.76 56	82.26 95	89.23 102	94.86 66	93.03 105	97.90 92	97.36 69

thisisatest0530		91.04 72	91.74 68	90.21 93	92.93 114	97.00 76	92.06 118	87.63 107	90.74 66	81.51 93	86.81 55	82.48 91	89.23 102	94.81 68	93.03 105	97.90 92	97.33 71

HyFIR lowres test		87.87 117	86.42 130	89.57 100	95.56 61	96.99 77	92.37 105	84.15 142	86.64 113	77.17 119	57.65 222	83.97 80	91.08 84	92.09 129	92.44 113	97.09 129	95.16 152

MVS_Test		91.81 64	92.19 61	91.37 80	93.24 106	96.95 78	94.43 61	86.25 115	91.45 62	83.45 85	86.31 57	85.15 76	92.93 57	93.99 87	94.71 54	97.92 91	96.77 88

Vis-MVSNet (Re-imp)		90.54 79	92.76 52	87.94 121	93.73 98	96.94 79	92.17 115	87.91 97	88.77 94	76.12 124	83.68 85	90.80 54	79.49 200	96.34 36	96.35 26	98.21 63	96.46 102

diffmvs1		91.72 66	92.13 63	91.24 83	93.20 107	96.92 80	94.37 62	86.24 116	94.05 41	84.30 82	85.80 66	83.64 82	92.71 60	93.47 104	93.92 71	96.60 172	97.11 77

CHOSEN 1792x2688		88.57 111	87.82 112	89.44 103	95.46 64	96.89 81	93.74 85	85.87 120	89.63 87	77.42 118	61.38 217	83.31 84	88.80 113	93.44 106	93.16 97	95.37 197	96.95 81

thres100view900		89.36 97	87.61 116	91.39 78	93.90 87	96.86 82	94.35 65	89.66 73	85.87 120	81.15 96	76.46 126	70.38 146	91.17 81	94.09 85	93.43 85	98.13 69	96.16 115

IB-MVS	85.10 14	87.98 115	87.97 109	87.99 120	94.55 71	96.86 82	84.52 208	88.21 93	86.48 118	88.54 47	74.41 141	77.74 117	74.10 216	89.65 169	92.85 107	98.06 81	97.80 57
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
diffmvs		90.76 76	90.92 77	90.57 88	92.71 118	96.70 84	93.37 94	86.13 117	91.95 56	83.12 87	85.24 69	80.56 104	91.17 81	92.08 130	93.08 100	96.95 144	96.82 85

tfpn_ndepth		89.72 90	89.91 86	89.49 101	93.56 104	96.67 85	92.34 107	89.25 79	90.85 65	78.68 114	84.25 81	77.39 120	84.84 163	93.58 99	92.76 110	98.30 48	93.90 168

CANet_DTU		90.74 77	92.93 51	88.19 115	94.36 72	96.61 86	94.34 66	84.66 136	90.66 68	68.75 183	90.41 42	86.89 65	89.78 95	95.46 49	94.87 52	97.25 120	95.62 132

DI_MVS_plusplus_trai		91.05 71	90.15 82	92.11 65	92.67 119	96.61 86	96.03 44	88.44 90	90.25 74	85.92 66	73.73 142	84.89 78	91.92 68	94.17 84	94.07 65	97.68 107	97.31 72

EPNet		93.92 43	94.40 38	93.36 50	97.89 31	96.55 88	96.08 43	92.14 41	91.65 60	89.16 42	94.07 27	90.17 62	87.78 116	95.24 50	94.97 50	97.09 129	98.15 38
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
gg-mvs-nofinetune		81.83 203	83.58 159	79.80 211	91.57 129	96.54 89	93.79 83	68.80 231	62.71 234	43.01 240	55.28 226	85.06 77	83.65 176	96.13 40	94.86 53	97.98 90	94.46 159

tfpn1000		89.30 100	89.72 88	88.81 109	93.83 95	96.50 90	91.53 127	88.74 87	91.20 64	76.74 121	84.96 74	75.44 129	83.50 178	93.63 97	92.42 114	98.51 14	93.88 169

PLC	90.69 4	94.32 40	92.99 50	95.87 26	97.91 30	96.49 91	95.95 47	94.12 31	94.94 30	94.09 10	85.90 62	90.77 55	95.58 29	94.52 76	93.32 89	97.55 113	95.00 155
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
TSAR-MVS + COLMAP		92.39 55	92.31 60	92.47 61	95.35 68	96.46 92	96.13 42	92.04 43	95.33 24	80.11 106	94.95 26	77.35 121	94.05 43	94.49 78	93.08 100	97.15 124	94.53 158

Effi-MVS+		89.79 89	89.83 87	89.74 98	92.98 110	96.45 93	93.48 92	84.24 140	87.62 106	76.45 122	81.76 98	77.56 119	93.48 52	94.61 73	93.59 76	97.82 96	97.22 74

ACMP	89.13 9	92.03 59	91.70 70	92.41 62	94.92 69	96.44 94	93.95 79	89.96 61	91.81 59	85.48 75	90.97 40	79.12 109	92.42 64	93.28 110	92.55 112	97.76 99	97.74 58
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
thresconf0.02		88.86 105	88.70 99	89.04 107	93.59 103	96.40 95	92.97 99	89.75 66	90.16 79	74.34 128	84.41 78	71.00 142	85.16 159	93.32 108	93.12 99	98.41 30	92.52 189

conf0.05thres1000		87.90 116	85.88 139	90.26 92	93.74 97	96.39 96	92.67 101	88.94 84	80.97 165	77.71 117	70.15 161	68.40 160	90.42 92	94.46 79	93.29 90	98.09 75	97.49 64

Anonymous202405211			88.00 107		93.16 109	96.38 97	93.58 89	89.34 77	87.92 104		65.04 204	83.03 86	92.07 67	92.67 116	93.33 87	96.96 142	97.63 59

tfpn_n400		88.58 109	88.91 96	88.19 115	93.63 101	96.34 98	92.22 111	89.04 81	87.37 107	72.14 142	85.12 71	73.93 131	84.04 174	93.65 95	93.20 94	98.09 75	92.77 182

tfpnconf		88.58 109	88.91 96	88.19 115	93.63 101	96.34 98	92.22 111	89.04 81	87.37 107	72.14 142	85.12 71	73.93 131	84.04 174	93.65 95	93.20 94	98.09 75	92.77 182

tfpnview11		88.80 106	89.21 91	88.31 114	93.70 99	96.24 100	92.35 106	89.11 80	89.90 85	72.14 142	85.12 71	73.93 131	84.20 169	93.75 94	92.85 107	98.38 35	92.68 187

CLD-MVS		92.50 54	91.96 66	93.13 55	93.93 86	96.24 100	95.69 49	88.77 86	92.92 48	89.01 43	88.19 51	81.74 100	93.13 55	93.63 97	93.08 100	98.23 61	97.91 52
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
LGP-MVS_train		91.83 63	92.04 65	91.58 71	95.46 64	96.18 102	95.97 46	89.85 62	90.45 72	77.76 115	91.92 35	80.07 106	92.34 66	94.27 81	93.47 84	98.11 72	97.90 53

OPM-MVS		91.08 70	89.34 89	93.11 57	96.18 55	96.13 103	96.39 39	92.39 39	82.97 153	81.74 90	82.55 94	80.20 105	93.97 46	94.62 72	93.23 91	98.00 87	95.73 129

Anonymous20231211		89.82 88	88.18 105	91.74 68	92.52 120	96.09 104	93.38 93	89.30 78	88.95 93	85.90 67	64.55 208	84.39 79	92.41 65	92.24 126	93.06 103	96.93 151	97.95 46

HQP-MVS		92.39 55	92.49 56	92.29 64	95.65 60	95.94 105	95.64 51	92.12 42	92.46 53	79.65 109	91.97 34	82.68 90	92.92 58	93.47 104	92.77 109	97.74 101	98.12 41

PatchMatch-RL		90.30 81	88.93 95	91.89 66	95.41 67	95.68 106	90.94 128	88.67 88	89.80 86	86.95 60	85.90 62	72.51 135	92.46 63	93.56 102	92.18 119	96.93 151	92.89 180

Fast-Effi-MVS+		88.56 112	87.99 108	89.22 105	91.56 130	95.21 107	92.29 109	82.69 159	86.82 111	77.73 116	76.24 130	73.39 134	93.36 54	94.22 83	93.64 74	97.65 109	96.43 103

FC-MVSNet-train		90.55 78	90.19 81	90.97 87	93.78 96	95.16 108	92.11 117	88.85 85	87.64 105	83.38 86	84.36 80	78.41 112	89.53 96	94.69 70	93.15 98	98.15 67	97.92 49

ACMM	88.76 10	91.70 67	90.43 79	93.19 54	95.56 61	95.14 109	93.35 95	91.48 49	92.26 54	87.12 57	84.02 82	79.34 108	93.99 44	94.07 86	92.68 111	97.62 112	95.50 135
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
Effi-MVS+-dtu		87.51 120	88.13 106	86.77 135	91.10 135	94.90 110	90.91 129	82.67 160	83.47 149	71.55 148	81.11 103	77.04 122	89.41 98	92.65 118	91.68 132	95.00 203	96.09 118

MVSTER		91.73 65	91.61 71	91.86 67	93.18 108	94.56 111	94.37 62	87.90 98	90.16 79	88.69 46	89.23 46	81.28 102	88.92 110	95.75 45	93.95 69	98.12 70	96.37 105

CDS-MVSNet		88.34 113	88.71 98	87.90 122	90.70 142	94.54 112	92.38 104	86.02 118	80.37 172	79.42 110	79.30 107	83.43 83	82.04 187	93.39 107	94.01 67	96.86 165	95.93 122
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
ACMH	85.51 13	87.31 122	86.59 127	88.14 118	93.96 83	94.51 113	89.00 176	87.99 95	81.58 158	70.15 166	78.41 114	71.78 140	90.60 89	91.30 140	91.99 125	97.17 123	96.58 99
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
COLMAP_ROB	84.39 15	87.61 119	86.03 134	89.46 102	95.54 63	94.48 114	91.77 124	90.14 60	87.16 109	75.50 125	73.41 147	76.86 124	87.33 123	90.05 163	89.76 187	96.48 174	90.46 202
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
GG-mvs-BLEND		62.84 231	90.21 80	30.91 242	0.57 249	94.45 115	86.99 193	0.34 247	88.71 95	0.98 249	81.55 102	91.58 51	0.86 247	92.66 117	91.43 135	95.73 187	91.11 197

UniMVSNet (Re)		86.22 131	85.46 145	87.11 130	88.34 160	94.42 116	89.65 167	87.10 112	84.39 140	74.61 127	70.41 159	68.10 161	85.10 161	91.17 142	91.79 128	97.84 95	97.94 47

ACMH+	85.75 12	87.19 123	86.02 135	88.56 112	93.42 105	94.41 117	89.91 159	87.66 106	83.45 150	72.25 140	76.42 129	71.99 139	90.78 86	89.86 164	90.94 139	97.32 118	95.11 154

MSDG		90.42 80	88.25 104	92.94 59	96.67 51	94.41 117	93.96 77	92.91 37	89.59 88	86.26 63	76.74 124	80.92 103	90.43 91	92.60 119	92.08 123	97.44 117	91.41 193

GA-MVS		85.08 147	85.65 142	84.42 175	89.77 146	94.25 119	89.26 171	84.62 137	81.19 163	62.25 214	75.72 133	68.44 159	84.14 171	93.57 101	91.68 132	96.49 173	94.71 157

TDRefinement		84.97 149	83.39 164	86.81 134	92.97 112	94.12 120	92.18 113	87.77 103	82.78 154	71.31 151	68.43 168	68.07 162	81.10 195	89.70 168	89.03 196	95.55 194	91.62 191

MS-PatchMatch		87.63 118	87.61 116	87.65 125	93.95 84	94.09 121	92.60 103	81.52 175	86.64 113	76.41 123	73.46 146	85.94 71	85.01 162	92.23 127	90.00 180	96.43 176	90.93 199

UniMVSNet_NR-MVSNet		86.80 126	85.86 140	87.89 123	88.17 162	94.07 122	90.15 150	88.51 89	84.20 144	73.45 134	72.38 152	70.30 150	88.95 108	90.25 157	92.21 118	98.12 70	97.62 61

USDC		86.73 128	85.96 137	87.63 126	91.64 128	93.97 123	92.76 100	84.58 138	88.19 100	70.67 159	80.10 105	67.86 163	89.43 97	91.81 132	89.77 186	96.69 171	90.05 206

FMVSNet3		90.19 84	90.06 85	90.34 89	88.69 155	93.85 124	94.58 60	85.78 121	90.03 81	85.56 72	77.38 117	86.13 68	89.22 104	93.29 109	94.36 58	98.20 64	95.40 140

DWT-MVSNet_training		86.83 125	84.44 151	89.61 99	92.75 117	93.82 125	91.66 125	82.85 157	88.57 97	87.48 54	79.00 109	64.24 202	88.82 112	85.18 211	87.50 201	94.07 205	92.79 181

EG-PatchMatch MVS		81.70 205	81.31 204	82.15 205	88.75 153	93.81 126	87.14 192	78.89 199	71.57 223	64.12 211	61.20 219	68.46 158	76.73 209	91.48 136	90.77 143	97.28 119	91.90 190

DU-MVS		86.12 133	84.81 148	87.66 124	87.77 170	93.78 127	90.15 150	87.87 100	84.40 138	73.45 134	70.59 156	64.82 197	88.95 108	90.14 158	92.33 115	97.76 99	97.62 61

NR-MVSNet		85.46 144	84.54 150	86.52 138	88.33 161	93.78 127	90.45 134	87.87 100	84.40 138	71.61 147	70.59 156	62.09 212	82.79 181	91.75 133	91.75 129	98.10 74	97.44 66

EPMVS		85.77 137	86.24 132	85.23 160	92.76 116	93.78 127	89.91 159	73.60 218	90.19 77	74.22 129	82.18 96	78.06 114	87.55 119	85.61 210	85.38 211	93.32 207	88.48 213

Fast-Effi-MVS+-dtu		86.25 130	87.70 114	84.56 173	90.37 144	93.70 130	90.54 133	78.14 200	83.50 148	65.37 206	81.59 101	75.83 128	86.09 146	91.70 134	91.70 130	96.88 163	95.84 126

MDTV_nov1_ep13		86.64 129	87.50 119	85.65 152	90.73 140	93.69 131	89.96 157	78.03 202	89.48 90	76.85 120	84.92 75	82.42 93	86.14 141	86.85 206	86.15 204	92.17 218	88.97 210

thisisatest0515		85.70 138	87.00 125	84.19 178	88.16 164	93.67 132	84.20 210	84.14 143	83.39 151	72.91 136	76.79 123	74.75 130	78.82 202	92.57 120	91.26 137	96.94 147	96.56 101

GBi-Net		90.21 82	90.11 83	90.32 90	88.66 156	93.65 133	94.25 73	85.78 121	90.03 81	85.56 72	77.38 117	86.13 68	89.38 99	93.97 88	94.16 61	98.31 45	95.47 136

test1		90.21 82	90.11 83	90.32 90	88.66 156	93.65 133	94.25 73	85.78 121	90.03 81	85.56 72	77.38 117	86.13 68	89.38 99	93.97 88	94.16 61	98.31 45	95.47 136

FMVSNet2		89.61 92	89.14 92	90.16 96	88.66 156	93.65 133	94.25 73	85.44 128	88.57 97	84.96 80	73.53 144	83.82 81	89.38 99	94.23 82	94.68 55	98.31 45	95.47 136

anonymousdsp		84.51 156	85.85 141	82.95 194	86.30 210	93.51 136	85.77 205	80.38 186	78.25 197	63.42 212	73.51 145	72.20 137	84.64 165	93.21 111	92.16 120	97.19 122	98.14 39

v1neww		84.65 153	83.34 168	86.18 142	87.53 178	93.49 137	90.32 137	85.17 131	80.57 169	71.02 157	66.93 178	67.04 172	86.13 143	89.26 180	90.23 166	96.93 151	95.88 124

v7new		84.65 153	83.34 168	86.18 142	87.53 178	93.49 137	90.32 137	85.17 131	80.57 169	71.02 157	66.93 178	67.04 172	86.13 143	89.26 180	90.23 166	96.93 151	95.88 124

v6		84.67 152	83.36 166	86.20 140	87.53 178	93.49 137	90.34 136	85.16 133	80.58 168	71.13 153	66.97 177	67.10 170	86.11 145	89.25 183	90.22 169	96.93 151	95.89 123

Anonymous20240521		91.24 69	91.26 74	91.22 84	92.84 115	93.44 140	93.82 82	86.75 114	91.33 63	85.61 71	84.00 83	85.46 74	91.27 80	92.91 112	93.62 75	97.02 133	98.05 44

v1141		84.40 161	83.00 176	86.03 144	87.41 184	93.42 141	90.28 145	85.53 125	79.58 181	70.12 168	66.62 186	66.27 183	85.94 147	89.16 186	90.19 171	96.89 160	95.73 129

divwei89l23v2f112		84.40 161	83.00 176	86.02 146	87.42 183	93.42 141	90.28 145	85.52 126	79.57 182	70.11 169	66.64 185	66.29 182	85.91 148	89.16 186	90.19 171	96.90 158	95.73 129

v1		84.40 161	83.01 175	86.03 144	87.41 184	93.42 141	90.31 141	85.52 126	79.51 184	70.13 167	66.66 184	66.40 177	85.89 149	89.15 188	90.19 171	96.89 160	95.74 128

PatchmatchNet		85.70 138	86.65 126	84.60 172	91.79 126	93.40 144	89.27 170	73.62 217	90.19 77	72.63 138	82.74 91	81.93 98	87.64 117	84.99 212	84.29 217	92.64 212	89.00 209
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
pm-mvs1		84.55 155	83.46 160	85.82 147	88.16 164	93.39 145	89.05 175	85.36 130	74.03 217	72.43 139	65.08 203	71.11 141	82.30 186	93.48 103	91.70 130	97.64 110	95.43 139

TranMVSNet+NR-MVSNet		85.57 142	84.41 152	86.92 132	87.67 173	93.34 146	90.31 141	88.43 91	83.07 152	70.11 169	69.99 163	65.28 192	86.96 127	89.73 166	92.27 116	98.06 81	97.17 76

WR-MVS_H		82.86 194	82.66 183	83.10 191	87.44 182	93.33 147	85.71 206	83.20 155	77.36 201	68.20 188	66.37 187	65.23 193	76.05 211	89.35 174	90.13 174	97.99 88	96.89 83

WR-MVS		83.14 189	83.38 165	82.87 195	87.55 177	93.29 148	86.36 199	84.21 141	80.05 176	66.41 200	66.91 180	66.92 174	75.66 212	88.96 191	90.56 149	97.05 131	96.96 80

v2v482		84.51 156	83.05 174	86.20 140	87.25 196	93.28 149	90.22 147	85.40 129	79.94 178	69.78 174	67.74 171	65.15 194	87.57 118	89.12 189	90.55 150	96.97 139	95.60 133

V42		84.48 158	83.36 166	85.79 149	87.14 199	93.28 149	90.03 154	83.98 145	80.30 173	71.20 152	66.90 182	67.17 168	85.55 155	89.35 174	90.27 157	96.82 166	96.27 112

tfpnnormal		83.80 175	81.26 205	86.77 135	89.60 148	93.26 151	89.72 166	87.60 109	72.78 219	70.44 160	60.53 220	61.15 217	85.55 155	92.72 114	91.44 134	97.71 103	96.92 82

CostFormer		86.78 127	86.05 133	87.62 127	92.15 122	93.20 152	91.55 126	75.83 209	88.11 102	85.29 77	81.76 98	76.22 126	87.80 115	84.45 216	85.21 212	93.12 208	93.42 175

v7		84.37 164	83.23 171	85.69 151	87.34 188	93.19 153	90.32 137	83.10 156	79.88 180	69.33 177	66.33 190	65.75 187	87.06 125	90.83 147	90.38 153	96.97 139	96.26 113

pmmvs5		83.37 186	82.68 182	84.18 179	87.13 200	93.18 154	86.74 195	82.08 168	76.48 206	67.28 194	71.26 153	62.70 207	84.71 164	90.77 148	90.12 177	97.15 124	94.24 162

FC-MVSNet-test		86.15 132	89.10 93	82.71 199	89.83 145	93.18 154	87.88 186	84.69 135	86.54 115	62.18 215	82.39 95	83.31 84	74.18 215	92.52 121	91.86 127	97.50 115	93.88 169

TAMVS		84.94 150	84.95 146	84.93 169	88.82 152	93.18 154	88.44 182	81.28 177	77.16 202	73.76 133	75.43 135	76.57 125	82.04 187	90.59 153	90.79 141	95.22 199	90.94 198

v1144		84.03 172	82.88 179	85.37 156	87.17 198	93.15 157	90.18 149	83.31 153	78.83 192	67.85 189	65.99 196	64.99 195	86.79 130	90.75 149	90.33 156	96.90 158	96.15 116

SixPastTwentyTwo		83.12 190	83.44 162	82.74 198	87.71 172	93.11 158	82.30 216	82.33 165	79.24 190	64.33 209	78.77 111	62.75 206	84.11 172	88.11 194	87.89 199	95.70 188	94.21 164

LTVRE_ROB	81.71 16	82.44 198	81.84 197	83.13 189	89.01 151	92.99 159	88.90 177	82.32 166	66.26 231	54.02 229	74.68 140	59.62 224	88.87 111	90.71 151	92.02 124	95.68 189	96.62 92
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
v8		84.45 160	83.30 170	85.80 148	87.53 178	92.95 160	90.31 141	82.46 164	80.46 171	71.43 149	66.99 176	67.16 169	86.14 141	89.26 180	90.22 169	96.94 147	96.06 119

v148		83.61 180	82.10 193	85.37 156	87.34 188	92.94 161	87.48 188	85.72 124	78.92 191	73.87 132	65.71 200	64.69 198	81.78 191	87.82 195	89.35 193	96.01 182	95.26 146

RPSCF		89.68 91	89.24 90	90.20 94	92.97 112	92.93 162	92.30 108	87.69 104	90.44 73	85.12 78	91.68 36	85.84 73	90.69 88	87.34 201	86.07 205	92.46 215	90.37 203

v144192		83.48 185	82.23 191	84.94 168	86.65 206	92.84 163	89.63 168	82.48 163	77.87 198	67.36 193	65.33 202	63.50 204	86.51 132	89.72 167	89.99 181	97.03 132	96.35 106

v1192		83.56 183	82.35 187	84.98 167	86.84 205	92.84 163	90.01 156	82.70 158	78.54 194	66.48 199	64.88 205	62.91 205	86.91 128	90.72 150	90.25 160	96.94 147	96.32 108

FMVSNet1		87.33 121	86.00 136	88.89 108	87.13 200	92.83 165	93.08 98	84.46 139	81.35 162	82.20 89	66.33 190	77.96 115	88.96 107	93.97 88	94.16 61	97.54 114	95.38 141

CHOSEN 280x420		90.77 75	92.14 62	89.17 106	93.86 94	92.81 166	93.16 96	80.22 191	90.21 76	84.67 81	89.89 43	91.38 53	90.57 90	94.94 56	92.11 121	92.52 214	93.65 172

CP-MVSNet		83.11 191	82.15 192	84.23 177	87.20 197	92.70 167	86.42 198	83.53 151	77.83 199	67.67 191	66.89 183	60.53 220	82.47 184	89.23 185	90.65 148	98.08 78	97.20 75

v10		84.18 166	83.17 173	85.37 156	87.34 188	92.68 168	90.32 137	81.33 176	79.93 179	69.23 180	66.33 190	65.74 189	87.03 126	90.84 146	90.38 153	96.97 139	96.29 111

v1921920		83.30 187	82.09 194	84.70 170	86.59 208	92.67 169	89.82 165	82.23 167	78.32 195	65.76 203	64.64 207	62.35 208	86.78 131	90.34 156	90.02 179	97.02 133	96.31 110

test-mter		86.09 135	88.38 101	83.43 187	87.89 167	92.61 170	86.89 194	77.11 205	84.30 141	68.62 185	82.57 93	82.45 92	84.34 166	92.40 122	90.11 178	95.74 186	94.21 164

v7n		82.25 199	81.54 200	83.07 192	85.55 214	92.58 171	86.68 197	81.10 180	76.54 205	65.97 202	62.91 214	60.56 219	82.36 185	91.07 144	90.35 155	96.77 168	96.80 86

dps		85.00 148	83.21 172	87.08 131	90.73 140	92.55 172	89.34 169	75.29 211	84.94 133	87.01 58	79.27 108	67.69 164	87.27 124	84.22 218	83.56 218	92.83 210	90.25 204

test0.0.03 1		85.58 141	87.69 115	83.11 190	91.22 133	92.54 173	85.60 207	83.62 148	85.66 128	67.84 190	82.79 90	79.70 107	73.51 218	91.15 143	90.79 141	96.88 163	91.23 196

PS-CasMVS		82.53 196	81.54 200	83.68 183	87.08 202	92.54 173	86.20 200	83.46 152	76.46 207	65.73 204	65.71 200	59.41 225	81.61 192	89.06 190	90.55 150	98.03 83	97.07 79

V4		82.11 200	81.49 203	82.83 196	84.60 219	92.53 175	85.97 202	80.24 189	76.35 210	66.87 197	63.17 211	64.55 201	82.54 183	87.70 197	89.55 189	96.73 169	96.61 93

v1240		82.88 193	81.66 198	84.29 176	86.46 209	92.52 176	89.06 174	81.82 172	77.16 202	65.09 207	64.17 209	61.50 214	86.36 133	90.12 160	90.13 174	96.95 144	96.04 121

v52		82.11 200	81.50 202	82.82 197	84.59 220	92.51 177	85.96 204	80.24 189	76.38 209	66.83 198	63.12 212	64.62 200	82.56 182	87.70 197	89.55 189	96.73 169	96.61 93

IterMVS-LS		88.60 108	88.45 100	88.78 110	92.02 124	92.44 178	92.00 121	83.57 150	86.52 116	78.90 113	78.61 113	81.34 101	89.12 105	90.68 152	93.18 96	97.10 128	96.35 106
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
Patchmtry						92.39 179	89.18 172	73.30 222		71.08 154

EPNet_dtu		88.32 114	90.61 78	85.64 153	96.79 50	92.27 180	92.03 120	90.31 59	89.05 92	65.44 205	89.43 45	85.90 72	74.22 214	92.76 113	92.09 122	95.02 202	92.76 184
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
MIMVSNet		82.97 192	84.00 156	81.77 208	82.23 224	92.25 181	87.40 191	72.73 224	81.48 159	69.55 175	68.79 167	72.42 136	81.82 190	92.23 127	92.25 117	96.89 160	88.61 211

CR-MVSNet		85.48 143	86.29 131	84.53 174	91.08 137	92.10 182	89.18 172	73.30 222	84.75 134	71.08 154	73.12 150	77.91 116	86.27 136	91.48 136	90.75 144	96.27 179	93.94 166

RPMNet		84.82 151	85.90 138	83.56 185	91.10 135	92.10 182	88.73 180	71.11 226	84.75 134	68.79 182	73.56 143	77.62 118	85.33 158	90.08 162	89.43 192	96.32 178	93.77 171

v748		81.57 206	80.68 209	82.60 201	85.55 214	92.07 184	83.57 211	82.06 169	74.64 216	69.97 171	63.11 213	61.46 215	78.09 204	87.30 202	89.88 183	96.37 177	96.32 108

test-LLR		86.88 124	88.28 102	85.24 159	91.22 133	92.07 184	87.41 189	83.62 148	84.58 136	69.33 177	83.00 86	82.79 87	84.24 167	92.26 124	89.81 184	95.64 190	93.44 173

TESTMET0.1,1		86.11 134	88.28 102	83.59 184	87.80 168	92.07 184	87.41 189	77.12 204	84.58 136	69.33 177	83.00 86	82.79 87	84.24 167	92.26 124	89.81 184	95.64 190	93.44 173

PEN-MVS		82.49 197	81.58 199	83.56 185	86.93 203	92.05 187	86.71 196	83.84 146	76.94 204	64.68 208	67.24 172	60.11 221	81.17 194	87.78 196	90.70 147	98.02 85	96.21 114

pmmvs4		86.00 136	84.28 153	88.00 119	87.80 168	92.01 188	89.94 158	84.91 134	86.79 112	80.98 101	73.41 147	66.34 179	88.12 114	89.31 179	88.90 197	96.24 180	93.20 178

tpmp4_e23		85.67 140	84.28 153	87.30 129	91.96 125	92.00 189	92.06 118	76.27 207	87.95 103	83.59 84	76.97 122	70.88 143	87.52 120	84.80 215	84.73 214	92.40 216	92.61 188

PMMVS		89.88 87	91.19 75	88.35 113	89.73 147	91.97 190	90.62 131	81.92 170	90.57 71	80.58 104	92.16 32	86.85 66	91.17 81	92.31 123	91.35 136	96.11 181	93.11 179

TinyColmap		84.04 171	82.01 195	86.42 139	90.87 138	91.84 191	88.89 178	84.07 144	82.11 157	69.89 173	71.08 154	60.81 218	89.04 106	90.52 154	89.19 194	95.76 185	88.50 212

CMPMVS	61.19 17	79.86 210	77.46 218	82.66 200	91.54 131	91.82 192	83.25 212	81.57 174	70.51 227	68.64 184	59.89 221	66.77 175	79.63 198	84.00 220	84.30 216	91.34 222	84.89 222
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
ADS-MVSNet		84.08 170	84.95 146	83.05 193	91.53 132	91.75 193	88.16 183	70.70 227	89.96 84	69.51 176	78.83 110	76.97 123	86.29 135	84.08 219	84.60 215	92.13 220	88.48 213

tpmrst		83.72 176	83.45 161	84.03 181	92.21 121	91.66 194	88.74 179	73.58 219	88.14 101	72.67 137	77.37 120	72.11 138	86.34 134	82.94 222	82.05 223	90.63 226	89.86 207

pmmvs6		80.90 207	78.77 212	83.38 188	85.84 211	91.61 195	86.01 201	82.54 162	64.17 232	70.43 161	54.14 230	67.06 171	80.73 196	90.50 155	89.17 195	94.74 204	94.75 156

PatchT		83.86 173	85.51 144	81.94 206	88.41 159	91.56 196	78.79 223	71.57 225	84.08 146	71.08 154	70.62 155	76.13 127	86.27 136	91.48 136	90.75 144	95.52 195	93.94 166

TransMVSNet (Re)		82.67 195	80.93 208	84.69 171	88.71 154	91.50 197	87.90 185	87.15 111	71.54 225	68.24 187	63.69 210	64.67 199	78.51 203	91.65 135	90.73 146	97.64 110	92.73 186

tpm cat1		84.13 168	81.99 196	86.63 137	91.74 127	91.50 197	90.68 130	75.69 210	86.12 119	85.44 76	72.39 151	70.72 144	85.16 159	80.89 228	81.56 226	91.07 224	90.71 200

DTE-MVSNet		81.76 204	81.04 206	82.60 201	86.63 207	91.48 199	85.97 202	83.70 147	76.45 208	62.44 213	67.16 173	59.98 222	78.98 201	87.15 203	89.93 182	97.88 94	95.12 153

tpm		83.16 188	83.64 158	82.60 201	90.75 139	91.05 200	88.49 181	73.99 215	82.36 155	67.08 196	78.10 115	68.79 156	84.17 170	85.95 209	85.96 207	91.09 223	93.23 177

CVMVSNet		83.83 174	85.53 143	81.85 207	89.60 148	90.92 201	87.81 187	83.21 154	80.11 175	60.16 219	76.47 125	78.57 111	76.79 208	89.76 165	90.13 174	93.51 206	92.75 185

MDTV_nov1_ep13_2view		80.43 208	80.94 207	79.84 210	84.82 218	90.87 202	84.23 209	73.80 216	80.28 174	64.33 209	70.05 162	68.77 157	79.67 197	84.83 214	83.50 219	92.17 218	88.25 215

testgi		81.94 202	84.09 155	79.43 212	89.53 150	90.83 203	82.49 215	81.75 173	80.59 167	59.46 221	82.82 89	65.75 187	67.97 220	90.10 161	89.52 191	95.39 196	89.03 208

Baseline_NR-MVSNet		85.28 145	83.42 163	87.46 128	87.77 170	90.80 204	89.90 161	87.69 104	83.93 147	74.16 130	64.72 206	66.43 176	87.48 122	90.14 158	90.83 140	97.73 102	97.11 77

IterMVS		85.25 146	86.49 129	83.80 182	90.42 143	90.77 205	90.02 155	78.04 201	84.10 145	66.27 201	77.28 121	78.41 112	83.01 179	90.88 145	89.72 188	95.04 201	94.24 162
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
our_test_3					86.93 203	89.77 206	81.61 217

v18		84.21 165	82.90 178	85.74 150	87.63 174	89.75 207	90.56 132	80.82 181	81.42 160	72.24 141	67.16 173	67.23 166	86.27 136	89.25 183	90.24 163	96.92 156	95.27 145

v17		84.10 169	82.83 181	85.57 155	87.58 176	89.72 208	90.30 144	80.70 183	81.00 164	71.72 146	67.01 175	67.24 165	86.19 140	89.32 177	90.25 160	96.95 144	95.29 143

v16		84.14 167	82.86 180	85.64 153	87.61 175	89.71 209	90.36 135	80.70 183	81.36 161	71.99 145	66.91 180	67.19 167	86.23 139	89.32 177	90.25 160	96.94 147	95.29 143

v15		83.67 178	82.37 186	85.19 161	87.39 186	89.63 210	90.19 148	80.43 185	79.49 186	70.27 162	66.37 187	66.33 180	85.88 150	89.34 176	90.23 166	96.96 142	95.22 150

V14		83.66 179	82.38 185	85.16 162	87.37 187	89.62 211	90.15 150	80.33 187	79.51 184	70.26 163	66.30 193	66.37 178	85.87 151	89.38 173	90.24 163	96.98 138	95.22 150

V9		83.61 180	82.33 188	85.11 163	87.34 188	89.59 212	90.10 153	80.25 188	79.38 188	70.17 165	66.15 194	66.33 180	85.82 153	89.41 172	90.24 163	96.99 137	95.23 149

v12		83.59 182	82.32 189	85.07 164	87.32 194	89.57 213	89.87 164	80.19 192	79.46 187	70.19 164	66.05 195	66.23 185	85.84 152	89.44 171	90.26 159	97.01 135	95.26 146

v11		83.72 176	82.61 184	85.02 165	87.34 188	89.56 214	89.89 162	79.92 194	79.55 183	69.21 181	66.36 189	65.48 190	86.84 129	91.43 139	90.51 152	96.92 156	95.37 142

v13		83.55 184	82.29 190	85.01 166	87.31 195	89.55 215	89.89 162	80.13 193	79.34 189	69.93 172	65.92 198	66.25 184	85.80 154	89.45 170	90.27 157	97.01 135	95.25 148

Anonymous20231206		78.09 214	78.11 215	78.07 216	85.19 217	89.17 216	80.99 218	81.24 179	75.46 214	58.25 223	54.78 229	59.90 223	66.73 224	88.94 192	88.26 198	96.01 182	90.25 204

MDA-MVSNet-bldmvs		73.81 221	72.56 226	75.28 219	72.52 238	88.87 217	74.95 227	82.67 160	71.57 223	55.02 226	65.96 197	42.84 240	76.11 210	70.61 237	81.47 227	90.38 228	86.59 217

FMVSNet5		84.47 159	84.72 149	84.18 179	83.30 223	88.43 218	88.09 184	79.42 197	84.25 142	74.14 131	73.15 149	78.74 110	83.65 176	91.19 141	91.19 138	96.46 175	86.07 219

PM-MVS		80.29 209	79.30 211	81.45 209	81.91 225	88.23 219	82.61 214	79.01 198	79.99 177	67.15 195	69.07 166	51.39 230	82.92 180	87.55 200	85.59 208	95.08 200	93.28 176

LP		77.28 217	76.57 220	78.12 215	88.17 162	88.06 220	80.85 220	68.35 234	80.78 166	61.49 217	57.59 223	61.80 213	77.59 205	81.45 227	82.34 222	92.25 217	83.96 225

MVS-HIRNet		78.16 213	77.57 217	78.83 213	85.83 212	87.76 221	76.67 224	70.22 228	75.82 213	67.39 192	55.61 225	70.52 145	81.96 189	86.67 207	85.06 213	90.93 225	81.58 228

test20.03		76.41 218	78.49 214	73.98 220	85.64 213	87.50 222	75.89 225	80.71 182	70.84 226	51.07 232	68.06 170	61.40 216	54.99 234	88.28 193	87.20 202	95.58 193	86.15 218

pmmvs-eth3d		79.78 211	77.58 216	82.34 204	81.57 226	87.46 223	82.92 213	81.28 177	75.33 215	71.34 150	61.88 215	52.41 229	81.59 193	87.56 199	86.90 203	95.36 198	91.48 192

N_pmnet		77.55 216	76.68 219	78.56 214	85.43 216	87.30 224	78.84 222	81.88 171	78.30 196	60.61 218	61.46 216	62.15 211	74.03 217	82.04 223	80.69 229	90.59 227	84.81 223

EU-MVSNet		78.43 212	80.25 210	76.30 218	83.81 222	87.27 225	80.99 218	79.52 196	76.01 211	54.12 228	70.44 158	64.87 196	67.40 223	86.23 208	85.54 210	91.95 221	91.41 193

MIMVSNet1		73.19 223	73.70 224	72.60 225	65.42 242	86.69 226	75.56 226	79.65 195	67.87 230	55.30 225	45.24 238	56.41 227	63.79 227	86.98 204	87.66 200	95.85 184	85.04 221

gm-plane-assit		77.65 215	78.50 213	76.66 217	87.96 166	85.43 227	64.70 236	74.50 213	64.15 233	51.26 231	61.32 218	58.17 226	84.11 172	95.16 51	93.83 72	97.45 116	91.41 193

new-patchmatchnet		72.32 224	71.09 227	73.74 221	81.17 228	84.86 228	72.21 233	77.48 203	68.32 229	54.89 227	55.10 227	49.31 234	63.68 228	79.30 229	76.46 234	93.03 209	84.32 224

new_pmnet		72.29 225	73.25 225	71.16 228	75.35 235	81.38 229	73.72 229	69.27 230	75.97 212	49.84 233	56.27 224	56.12 228	69.08 219	81.73 224	80.86 228	89.72 232	80.44 229

testus		73.65 222	74.92 222	72.17 226	80.93 229	81.11 230	73.02 232	75.23 212	73.23 218	48.77 234	69.38 165	46.10 239	62.28 229	84.84 213	86.01 206	92.77 211	83.75 226

testpf		74.66 219	76.34 221	72.71 223	87.34 188	80.91 231	73.15 231	60.30 241	78.73 193	61.68 216	69.83 164	62.22 210	67.48 221	76.83 232	78.17 233	86.28 235	87.68 216

test2356		73.82 220	74.82 223	72.66 224	81.25 227	80.70 232	73.47 230	75.91 208	72.55 220	48.73 235	68.14 169	50.74 231	63.96 226	84.44 217	85.57 209	92.63 213	81.60 227

pmmvs3		71.13 226	71.06 228	71.21 227	73.54 237	80.19 233	71.69 234	64.86 235	62.04 235	52.10 230	54.92 228	48.00 237	75.03 213	83.75 221	83.24 220	90.04 231	85.27 220

ambc			67.96 229		73.69 236	79.79 234	73.82 228		71.61 222	59.80 220	46.00 235	20.79 246	66.15 225	86.92 205	80.11 231	89.13 233	90.50 201

FPMVS		69.87 227	67.10 230	73.10 222	84.09 221	78.35 235	79.40 221	76.41 206	71.92 221	57.71 224	54.06 231	50.04 232	56.72 232	71.19 236	68.70 237	84.25 237	75.43 233

1111		66.22 228	66.42 231	65.98 229	75.69 232	76.42 236	58.90 237	63.25 236	57.86 236	48.33 236	45.46 236	49.13 235	61.32 230	81.57 225	82.80 221	88.38 234	71.69 238

.test1245		48.95 237	46.78 239	51.48 235	75.69 232	76.42 236	58.90 237	63.25 236	57.86 236	48.33 236	45.46 236	49.13 235	61.32 230	81.57 225	5.58 243	1.40 247	11.42 244

testmv		65.29 229	65.25 232	65.34 230	77.73 230	75.55 238	58.75 239	73.56 220	53.22 239	38.47 241	49.33 232	38.30 241	53.38 235	79.13 230	81.65 224	90.15 229	79.58 230

test1235678		65.29 229	65.24 233	65.34 230	77.73 230	75.54 239	58.75 239	73.56 220	53.19 240	38.47 241	49.32 233	38.28 242	53.38 235	79.13 230	81.65 224	90.15 229	79.57 231

DeepMVS_CX						71.82 240	68.37 235	48.05 243	77.38 200	46.88 238	65.77 199	47.03 238	67.48 221	64.27 240		76.89 242	76.72 232

test12356		60.37 233	61.08 234	59.53 234	72.42 239	70.09 241	57.72 241	69.53 229	51.31 241	36.05 243	47.32 234	32.04 243	36.19 240	74.15 235	80.35 230	85.27 236	72.29 236

no-one		49.70 236	49.06 238	50.46 237	65.32 243	67.46 242	38.16 246	68.73 232	34.38 245	22.88 245	24.40 241	22.99 245	28.55 243	51.41 241	70.93 235	79.08 241	71.81 237

PMMVS2		53.68 235	55.72 237	51.30 236	58.84 244	67.02 243	54.23 242	60.97 240	47.50 242	19.42 246	34.81 240	31.97 244	30.88 242	65.84 239	69.99 236	83.47 238	72.92 235

Gipuma		58.52 234	56.17 236	61.27 233	67.14 241	58.06 244	52.16 244	68.40 233	69.00 228	45.02 239	22.79 242	20.57 247	55.11 233	76.27 233	79.33 232	79.80 240	67.16 239
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
PMVS	56.77 18	61.27 232	58.64 235	64.35 232	75.66 234	54.60 245	53.62 243	74.23 214	53.69 238	58.37 222	44.27 239	49.38 233	44.16 239	69.51 238	65.35 239	80.07 239	73.66 234
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
MVE	39.81 19	39.52 239	41.58 240	37.11 241	33.93 247	49.06 246	26.45 249	54.22 242	29.46 246	24.15 244	20.77 244	10.60 250	34.42 241	51.12 242	65.27 240	49.49 246	64.81 240
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
tmp_tt				50.24 238	68.55 240	46.86 247	48.90 245	18.28 244	86.51 117	68.32 186	70.19 160	65.33 191	26.69 244	74.37 234	66.80 238	70.72 243

EMVS		39.04 240	34.32 242	44.54 240	58.25 245	39.35 248	27.61 248	62.55 239	35.99 243	16.40 248	20.04 245	14.77 248	44.80 237	33.12 244	44.10 242	57.61 245	52.89 242

E-PMN		40.00 238	35.74 241	44.98 239	57.69 246	39.15 249	28.05 247	62.70 238	35.52 244	17.78 247	20.90 243	14.36 249	44.47 238	35.89 243	47.86 241	59.15 244	56.47 241

testmvs		4.35 241	6.54 243	1.79 243	0.60 248	1.82 250	3.06 251	0.95 245	7.22 247	0.88 250	12.38 246	1.25 251	3.87 246	6.09 245	5.58 243	1.40 247	11.42 244

test123		3.48 242	5.31 244	1.34 244	0.20 250	1.52 251	2.17 252	0.58 246	6.13 248	0.31 251	9.85 247	0.31 252	3.90 245	2.65 246	5.28 245	0.87 249	11.46 243

sosnet-low-res		0.00 243	0.00 245	0.00 245	0.00 251	0.00 252	0.00 253	0.00 248	0.00 249	0.00 252	0.00 248	0.00 253	0.00 248	0.00 247	0.00 246	0.00 250	0.00 246

sosnet		0.00 243	0.00 245	0.00 245	0.00 251	0.00 252	0.00 253	0.00 248	0.00 249	0.00 252	0.00 248	0.00 253	0.00 248	0.00 247	0.00 246	0.00 250	0.00 246

MTAPA										95.36 2		97.46 17

MTMP										95.70 1		96.90 22

Patchmatch-RL test							18.47 250

mPP-MVS					98.76 20							95.49 34

NP-MVS									91.63 61

High-resolution multi-view benchmark