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.

TDRefinement		93.16 1	95.57 1	90.36 1	88.79 48	93.57 1	97.27 1	78.23 20	95.55 2	93.00 1	93.98 17	96.01 49	87.53 1	97.69 1	96.81 1	97.33 1	95.34 4

XVS					91.28 23	91.23 8	96.89 2			87.14 27		94.53 84				95.84 15

X-MVStestdata					91.28 23	91.23 8	96.89 2			87.14 27		94.53 84				95.84 15

X-MVS		89.36 25	90.73 42	87.77 17	91.50 18	91.23 8	96.76 4	78.88 16	87.29 51	87.14 27	78.98 149	94.53 84	76.47 53	95.25 19	94.28 12	95.85 14	93.55 14

ACMMPR		91.30 4	92.88 10	89.46 4	91.92 11	91.61 5	96.60 5	79.46 12	90.08 29	88.53 14	89.54 76	95.57 61	84.25 7	95.24 20	94.27 13	95.97 11	93.85 7

MP-MVS		90.84 6	91.95 29	89.55 3	92.92 5	90.90 18	96.56 6	79.60 9	86.83 56	88.75 13	89.00 84	94.38 88	84.01 9	94.94 25	94.34 11	95.45 24	93.24 20

PGM-MVS		90.42 10	91.58 34	89.05 6	91.77 13	91.06 13	96.51 7	78.94 15	85.41 70	87.67 18	87.02 102	95.26 69	83.62 13	95.01 24	93.94 16	95.79 19	93.40 18

SteuartSystems-ACMMP		90.00 16	91.73 31	87.97 13	91.21 27	90.29 27	96.51 7	78.00 22	86.33 60	85.32 41	88.23 89	94.67 83	82.08 24	95.13 22	93.88 17	94.72 34	93.59 11
Skip Steuart: Steuart Systems R&D Blog.
ACMMP		90.63 8	92.40 18	88.56 9	91.24 26	91.60 6	96.49 9	77.53 25	87.89 44	86.87 31	87.24 99	96.46 32	82.87 19	95.59 15	94.50 9	96.35 6	93.51 15
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
CP-MVS		91.09 5	92.33 21	89.65 2	92.16 10	90.41 25	96.46 10	80.38 6	88.26 42	89.17 11	87.00 103	96.34 38	83.95 10	95.77 11	94.72 8	95.81 17	93.78 9

MPTG		90.38 11	91.35 37	89.25 5	93.08 3	86.59 58	96.45 11	79.00 14	90.23 26	89.30 10	85.87 113	94.97 79	82.54 21	95.05 23	94.83 7	95.14 27	91.94 32

ACMM	80.67 7	90.67 7	92.46 17	88.57 8	91.35 20	89.93 30	96.34 12	77.36 30	90.17 27	86.88 30	87.32 97	96.63 28	83.32 14	95.79 10	94.49 10	96.19 9	92.91 23
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
anonymousdsp		85.62 57	90.53 44	79.88 93	64.64 214	76.35 147	96.28 13	53.53 203	85.63 67	81.59 81	92.81 29	97.71 16	86.88 2	94.56 26	92.83 24	96.35 6	93.84 8

HSP-MVS		88.32 39	90.71 43	85.53 33	90.63 34	92.01 4	96.15 14	77.52 26	86.02 63	81.39 83	90.21 69	96.08 46	76.38 55	88.30 88	86.70 80	91.12 77	95.64 1

TSAR-MVS + MP.		89.67 22	92.25 23	86.65 24	91.53 16	90.98 16	96.15 14	73.30 52	87.88 45	81.83 75	92.92 28	95.15 73	82.23 22	93.58 33	92.25 33	94.87 30	93.01 22

APDe-MVS		89.85 19	92.91 9	86.29 26	90.47 36	91.34 7	96.04 16	76.41 37	91.11 15	78.50 99	93.44 20	95.82 53	81.55 27	93.16 36	91.90 37	94.77 33	93.58 13

ACMMP_Plus		89.86 18	91.96 28	87.42 19	91.00 28	90.08 28	96.00 17	76.61 34	89.28 33	87.73 17	90.04 70	91.80 120	78.71 38	94.36 29	93.82 18	94.48 35	94.32 5

CPTT-MVS		89.63 23	90.52 45	88.59 7	90.95 29	90.74 20	95.71 18	79.13 13	87.70 46	85.68 39	80.05 145	95.74 56	84.77 6	94.28 30	92.68 26	95.28 26	92.45 27

HFP-MVS		90.32 13	92.37 20	87.94 14	91.46 19	90.91 17	95.69 19	79.49 10	89.94 32	83.50 62	89.06 83	94.44 87	81.68 26	94.17 31	94.19 14	95.81 17	93.87 6

COLMAP_ROB	85.66 2	91.85 2	95.01 2	88.16 12	88.98 47	92.86 2	95.51 20	72.17 55	94.95 5	91.27 3	94.11 16	97.77 14	84.22 8	96.49 4	95.27 5	96.79 2	93.60 10
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
APD-MVS		89.14 27	91.25 39	86.67 23	91.73 14	91.02 15	95.50 21	77.74 23	84.04 82	79.47 94	91.48 49	94.85 80	81.14 28	92.94 39	92.20 35	94.47 36	92.24 28
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
LGP-MVS_train		90.56 9	92.38 19	88.43 10	90.88 30	91.15 11	95.35 22	77.65 24	86.26 62	87.23 24	90.45 68	97.35 20	83.20 15	95.44 16	93.41 20	96.28 8	92.63 24

HPM-MVS++		88.74 37	89.54 51	87.80 16	92.58 7	85.69 66	95.10 23	78.01 21	87.08 53	87.66 19	87.89 92	92.07 116	80.28 32	90.97 67	91.41 41	93.17 50	91.69 34

ESAPD		89.27 26	91.76 30	86.36 25	90.60 35	90.40 26	95.08 24	77.43 28	87.49 48	80.35 88	92.38 38	94.32 89	80.59 29	92.69 45	91.58 40	94.13 38	93.44 16

OPM-MVS		89.82 20	92.24 24	86.99 22	90.86 31	89.35 34	95.07 25	75.91 39	91.16 14	86.87 31	91.07 60	97.29 21	79.13 36	93.32 34	91.99 36	94.12 39	91.49 39

ACMP	80.00 8	90.12 15	92.30 22	87.58 18	90.83 32	91.10 12	94.96 26	76.06 38	87.47 49	85.33 40	88.91 86	97.65 18	82.13 23	95.31 17	93.44 19	96.14 10	92.22 29
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
UA-Net		89.02 31	91.44 36	86.20 27	94.88 1	89.84 31	94.76 27	77.45 27	85.41 70	74.79 114	88.83 87	88.90 142	78.67 40	96.06 7	95.45 4	96.66 3	95.58 2

DeepC-MVS	83.59 4	90.37 12	92.56 16	87.82 15	91.26 25	92.33 3	94.72 28	80.04 7	90.01 30	84.61 44	93.33 21	94.22 90	80.59 29	92.90 40	92.52 28	95.69 21	92.57 25
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
train_agg		86.67 49	87.73 68	85.43 34	91.51 17	82.72 81	94.47 29	74.22 49	81.71 104	81.54 82	89.20 82	92.87 104	78.33 42	90.12 74	88.47 63	92.51 61	89.04 58

SD-MVS		89.91 17	92.23 25	87.19 21	91.31 22	89.79 32	94.31 30	75.34 42	89.26 34	81.79 76	92.68 30	95.08 75	83.88 11	93.10 37	92.69 25	96.54 4	93.02 21

3Dnovator+	83.71 3	88.13 41	90.00 48	85.94 28	86.82 63	91.06 13	94.26 31	75.39 41	88.85 38	85.76 38	85.74 115	86.92 152	78.02 43	93.03 38	92.21 34	95.39 25	92.21 30

DeepPCF-MVS	81.61 6	87.95 44	90.29 47	85.22 37	87.48 59	90.01 29	93.79 32	73.54 50	88.93 36	83.89 53	89.40 78	90.84 129	80.26 33	90.62 71	90.19 49	92.36 62	92.03 31

PMVS	79.51 9	90.23 14	92.67 12	87.39 20	90.16 37	88.75 38	93.64 33	75.78 40	90.00 31	83.70 56	92.97 27	92.22 113	86.13 4	97.01 3	96.79 2	94.94 29	90.96 43
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
LTVRE_ROB	86.82 1	91.55 3	94.43 3	88.19 11	83.19 102	86.35 61	93.60 34	78.79 17	95.48 4	91.79 2	93.08 25	97.21 23	86.34 3	97.06 2	96.27 3	95.46 23	95.56 3
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
WR-MVS		89.79 21	93.66 4	85.27 36	91.32 21	88.27 42	93.49 35	79.86 8	92.75 8	75.37 110	96.86 1	98.38 6	75.10 66	95.93 8	94.07 15	96.46 5	89.39 55

NCCC		86.74 48	87.97 67	85.31 35	90.64 33	87.25 53	93.27 36	74.59 45	86.50 58	83.72 55	75.92 176	92.39 111	77.08 50	91.72 50	90.68 44	92.57 60	91.30 41

CNVR-MVS		86.93 47	88.98 55	84.54 41	90.11 38	87.41 52	93.23 37	73.47 51	86.31 61	82.25 70	82.96 131	92.15 114	76.04 58	91.69 51	90.69 43	92.17 64	91.64 37

OMC-MVS		88.16 40	91.34 38	84.46 43	86.85 62	90.63 22	93.01 38	67.00 89	90.35 25	87.40 21	86.86 105	96.35 37	77.66 46	92.63 46	90.84 42	94.84 31	91.68 35

CDPH-MVS		86.66 50	88.52 58	84.48 42	89.61 42	88.27 42	92.86 39	72.69 54	80.55 120	82.71 66	86.92 104	93.32 100	75.55 63	91.00 65	89.85 51	93.47 44	89.71 52

DeepC-MVS_fast	81.78 5	87.38 45	89.64 49	84.75 38	89.89 40	90.70 21	92.74 40	74.45 46	86.02 63	82.16 73	86.05 111	91.99 119	75.84 61	91.16 60	90.44 45	93.41 45	91.09 42
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
LS3D		89.02 31	91.69 32	85.91 29	89.72 41	90.81 19	92.56 41	71.69 57	90.83 19	87.24 22	89.71 74	92.07 116	78.37 41	94.43 28	92.59 27	95.86 13	91.35 40

ACMH+	79.05 11	89.62 24	93.08 7	85.58 31	88.58 50	89.26 35	92.18 42	74.23 48	93.55 7	82.66 67	92.32 40	98.35 8	80.29 31	95.28 18	92.34 31	95.52 22	90.43 46

WR-MVS_H		88.99 33	93.28 5	83.99 49	91.92 11	89.13 36	91.95 43	83.23 1	90.14 28	71.92 130	95.85 4	98.01 13	71.83 103	95.82 9	93.19 22	93.07 52	90.83 45

MVS_0304		84.73 69	86.19 81	83.02 53	88.32 52	86.71 57	91.55 44	70.87 61	73.79 162	82.88 65	85.13 120	93.35 99	72.55 98	88.62 84	87.69 69	91.93 66	88.05 67

ACMH	78.40 12	88.94 35	92.62 14	84.65 39	86.45 65	87.16 54	91.47 45	68.79 76	95.49 3	89.74 6	93.55 19	98.50 3	77.96 44	94.14 32	89.57 56	93.49 43	89.94 51
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
HQP-MVS		85.02 65	86.41 78	83.40 50	89.19 45	86.59 58	91.28 46	71.60 58	82.79 90	83.48 63	78.65 152	93.54 97	72.55 98	86.49 101	85.89 87	92.28 63	90.95 44

PCF-MVS	76.59 14	84.11 73	85.27 95	82.76 60	86.12 68	88.30 41	91.24 47	69.10 73	82.36 96	84.45 45	77.56 157	90.40 133	72.91 97	85.88 108	83.88 103	92.72 57	88.53 63
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
TSAR-MVS + ACMM		89.14 27	92.11 27	85.67 30	89.27 44	90.61 23	90.98 48	79.48 11	88.86 37	79.80 90	93.01 26	93.53 98	83.17 16	92.75 44	92.45 29	91.32 73	93.59 11

DTE-MVSNet		88.99 33	92.77 11	84.59 40	93.31 2	88.10 45	90.96 49	83.09 2	91.38 12	76.21 104	96.03 2	98.04 11	70.78 112	95.65 14	92.32 32	93.18 49	87.84 68

CNLPA		85.50 59	88.58 56	81.91 64	84.55 80	87.52 51	90.89 50	63.56 134	88.18 43	84.06 49	83.85 128	91.34 126	76.46 54	91.27 57	89.00 61	91.96 65	88.88 60

PEN-MVS		88.86 36	92.92 8	84.11 48	92.92 5	88.05 47	90.83 51	82.67 5	91.04 16	74.83 113	95.97 3	98.47 4	70.38 113	95.70 13	92.43 30	93.05 53	88.78 61

PHI-MVS		86.37 52	88.14 64	84.30 44	86.65 64	87.56 50	90.76 52	70.16 64	82.55 92	89.65 7	84.89 123	92.40 110	75.97 59	90.88 69	89.70 53	92.58 58	89.03 59

TAPA-MVS	78.00 13	85.88 56	88.37 60	82.96 56	84.69 77	88.62 39	90.62 53	64.22 125	89.15 35	88.05 15	78.83 150	93.71 93	76.20 57	90.11 75	88.22 66	94.00 40	89.97 50
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
Vis-MVSNet		83.32 83	88.12 65	77.71 114	77.91 162	83.44 78	90.58 54	69.49 69	81.11 116	67.10 153	89.85 72	91.48 124	71.71 104	91.34 56	89.37 57	89.48 92	90.26 47
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
CSCG		88.12 42	91.45 35	84.23 45	88.12 56	90.59 24	90.57 55	68.60 78	91.37 13	83.45 64	89.94 71	95.14 74	78.71 38	91.45 55	88.21 67	95.96 12	93.44 16

AdaColmap		84.15 72	85.14 98	83.00 55	89.08 46	87.14 55	90.56 56	70.90 60	82.40 95	80.41 86	73.82 188	84.69 160	75.19 65	91.58 53	89.90 50	91.87 67	86.48 76

v52		86.26 54	90.85 40	80.91 73	72.49 188	81.25 103	90.55 57	60.30 170	90.43 24	87.24 22	94.64 11	98.30 10	83.16 18	92.86 42	86.82 78	91.69 68	91.65 36

V4		86.26 54	90.85 40	80.91 73	72.49 188	81.25 103	90.55 57	60.31 169	90.44 23	87.23 24	94.64 11	98.31 9	83.17 16	92.87 41	86.82 78	91.69 68	91.64 37

PS-CasMVS		89.07 30	93.23 6	84.21 46	92.44 8	88.23 44	90.54 59	82.95 3	90.50 21	75.31 111	95.80 5	98.37 7	71.16 106	96.30 5	93.32 21	92.88 54	90.11 49

SixPastTwentyTwo		89.14 27	92.19 26	85.58 31	84.62 78	82.56 84	90.53 60	71.93 56	91.95 10	85.89 36	94.22 14	97.25 22	85.42 5	95.73 12	91.71 39	95.08 28	91.89 33

MVS_111021_HR		83.95 74	86.10 83	81.44 69	84.62 78	80.29 115	90.51 61	68.05 84	84.07 81	80.38 87	84.74 124	91.37 125	74.23 75	90.37 73	87.25 71	90.86 80	84.59 85

TSAR-MVS + COLMAP		85.51 58	88.36 61	82.19 62	86.05 69	87.69 49	90.50 62	70.60 63	86.40 59	82.33 68	89.69 75	92.52 108	74.01 81	87.53 92	86.84 77	89.63 89	87.80 69

abl_6				79.30 101	84.98 76	85.78 64	90.50 62	66.88 90	77.08 148	74.02 119	73.29 191	89.34 137	68.94 121			90.49 82	85.98 79

CP-MVSNet		88.71 38	92.63 13	84.13 47	92.39 9	88.09 46	90.47 64	82.86 4	88.79 39	75.16 112	94.87 7	97.68 17	71.05 108	96.16 6	93.18 23	92.85 55	89.64 53

MCST-MVS		84.79 68	86.48 76	82.83 59	87.30 60	87.03 56	90.46 65	69.33 72	83.14 86	82.21 72	81.69 139	92.14 115	75.09 67	87.27 95	84.78 96	92.58 58	89.30 56

gm-plane-assit		71.56 175	69.99 188	73.39 135	84.43 82	73.21 169	90.42 66	51.36 210	84.08 80	76.00 106	91.30 56	37.09 236	59.01 157	73.65 196	70.24 193	79.09 186	60.37 209

CANet		82.84 92	84.60 106	80.78 77	87.30 60	85.20 68	90.23 67	69.00 74	72.16 170	78.73 98	84.49 125	90.70 131	69.54 118	87.65 91	86.17 83	89.87 87	85.84 81

CLD-MVS		82.75 96	87.22 73	77.54 118	88.01 57	85.76 65	90.23 67	54.52 196	82.28 97	82.11 74	88.48 88	95.27 68	63.95 140	89.41 78	88.29 65	86.45 148	81.01 125
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
RPSCF		88.05 43	92.61 15	82.73 61	84.24 83	88.40 40	90.04 69	66.29 93	91.46 11	82.29 69	88.93 85	96.01 49	79.38 34	95.15 21	94.90 6	94.15 37	93.40 18

v7n		87.11 46	90.46 46	83.19 52	85.22 74	83.69 75	90.03 70	68.20 83	91.01 17	86.71 34	94.80 8	98.46 5	77.69 45	91.10 62	85.98 85	91.30 74	88.19 64

MVS_111021_LR		83.20 87	85.33 93	80.73 81	82.88 107	78.23 127	89.61 71	65.23 111	82.08 99	81.19 84	85.31 118	92.04 118	75.22 64	89.50 77	85.90 86	90.24 83	84.23 89

v1192		83.61 77	85.23 96	81.72 66	84.05 85	82.15 87	89.54 72	66.20 94	81.38 112	86.76 33	91.79 46	96.03 48	74.88 69	81.81 160	80.92 132	88.91 97	82.50 111

v1921920		83.49 80	84.94 102	81.80 65	83.78 90	81.20 106	89.50 73	65.91 100	81.64 106	87.18 26	91.70 47	95.39 67	75.85 60	81.56 163	80.27 140	88.60 101	82.80 107

v1240		83.57 79	84.94 102	81.97 63	84.05 85	81.27 102	89.46 74	66.06 97	81.31 114	87.50 20	91.88 45	95.46 66	76.25 56	81.16 165	80.51 137	88.52 103	82.98 105

v144192		83.43 81	84.97 101	81.63 68	83.43 94	81.23 105	89.42 75	66.04 98	81.45 111	86.40 35	91.46 51	95.70 60	75.76 62	82.14 156	80.23 141	88.74 98	82.57 110

MAR-MVS		81.98 106	82.92 131	80.88 76	85.18 75	85.85 63	89.13 76	69.52 67	71.21 174	82.25 70	71.28 198	88.89 143	69.69 115	88.71 83	86.96 73	89.52 91	87.57 71
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
TSAR-MVS + GP.		85.32 61	87.41 72	82.89 58	90.07 39	85.69 66	89.07 77	72.99 53	82.45 94	74.52 117	85.09 121	87.67 149	79.24 35	91.11 61	90.41 46	91.45 71	89.45 54

v1144		83.22 86	85.01 99	81.14 70	83.76 91	81.60 96	88.95 78	65.58 106	81.89 100	85.80 37	91.68 48	95.84 52	74.04 80	82.12 157	80.56 136	88.70 100	81.41 121

PLC	76.06 15	85.38 60	87.46 70	82.95 57	85.79 71	88.84 37	88.86 79	68.70 77	87.06 54	83.60 58	79.02 148	90.05 134	77.37 49	90.88 69	89.66 54	93.37 46	86.74 75
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
v7		82.76 94	84.65 105	80.55 86	83.27 99	81.77 91	88.66 80	65.10 112	79.23 136	83.60 58	91.47 50	95.47 64	74.12 76	82.61 150	80.66 133	88.52 103	81.35 122

v10		83.17 88	85.22 97	80.78 77	83.26 100	82.99 80	88.66 80	66.49 92	79.24 135	83.60 58	91.46 51	95.47 64	74.12 76	82.60 151	80.66 133	88.53 102	84.11 91

v13		83.75 76	86.20 80	80.89 75	83.38 96	81.93 89	88.58 82	66.09 96	83.55 83	84.28 46	92.67 31	96.79 26	74.67 71	84.42 130	79.72 145	88.36 105	84.31 88

v12		83.59 78	86.00 86	80.77 80	83.30 98	81.83 90	88.45 83	65.95 99	83.20 85	84.15 47	92.54 36	96.71 27	74.50 73	84.19 132	79.64 146	88.30 106	83.93 93

Effi-MVS+-dtu		82.04 105	83.39 129	80.48 88	85.48 73	86.57 60	88.40 84	68.28 82	69.04 182	73.13 124	76.26 168	91.11 128	74.74 70	88.40 86	87.76 68	92.84 56	84.57 86

EPNet		79.36 126	79.44 143	79.27 102	89.51 43	77.20 137	88.35 85	77.35 31	68.27 184	74.29 118	76.31 166	79.22 174	59.63 154	85.02 123	85.45 90	86.49 147	84.61 84
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
v11		83.30 84	85.58 91	80.64 83	83.53 93	81.74 92	88.30 86	65.46 108	82.75 91	84.63 43	92.49 37	96.17 44	73.90 82	82.69 149	79.59 147	88.04 114	83.66 95

V9		83.42 82	85.81 88	80.63 84	83.20 101	81.73 93	88.29 87	65.78 103	82.87 89	83.99 52	92.38 38	96.60 29	74.30 74	83.93 133	79.58 148	88.24 109	83.55 97

V14		83.23 85	85.59 90	80.48 88	83.09 104	81.63 95	88.13 88	65.61 105	82.53 93	83.81 54	92.17 41	96.50 30	74.07 79	83.66 135	79.51 150	88.17 111	83.16 101

UniMVSNet (Re)		84.95 66	88.53 57	80.78 77	87.82 58	84.21 71	88.03 89	76.50 35	81.18 115	69.29 140	92.63 34	96.83 25	69.07 120	91.23 59	89.60 55	93.97 41	84.00 92

PVSNet_Blended_VisFu		83.00 90	84.16 117	81.65 67	82.17 121	86.01 62	88.03 89	71.23 59	76.05 154	79.54 93	83.88 127	83.44 161	77.49 48	87.38 93	84.93 95	91.41 72	87.40 73

v15		83.06 89	85.39 92	80.35 91	83.01 105	81.53 97	87.98 91	65.47 107	82.19 98	83.66 57	92.00 42	96.40 36	73.87 83	83.39 137	79.44 151	88.10 113	82.76 108

Baseline_NR-MVSNet		82.79 93	86.51 75	78.44 111	88.30 53	75.62 158	87.81 92	74.97 43	81.53 108	66.84 154	94.71 10	96.46 32	66.90 131	91.79 48	83.37 111	85.83 160	82.09 116

Fast-Effi-MVS+		81.42 114	83.82 124	78.62 109	82.24 120	80.62 108	87.72 93	63.51 135	73.01 163	74.75 115	83.80 129	92.70 106	73.44 86	88.15 90	85.26 91	90.05 84	83.17 100

v2v482		82.20 102	84.26 114	79.81 97	82.67 111	80.18 117	87.67 94	63.96 131	81.69 105	84.73 42	91.27 57	96.33 39	72.05 102	81.94 159	79.56 149	87.79 120	78.84 142

v1141		82.26 99	84.32 110	79.85 95	82.86 108	80.31 113	87.58 95	63.48 136	81.86 103	84.03 51	91.33 54	96.28 42	73.23 93	82.39 152	79.08 162	87.93 117	78.97 140

divwei89l23v2f112		82.26 99	84.32 110	79.85 95	82.86 108	80.31 113	87.58 95	63.48 136	81.88 101	84.05 50	91.33 54	96.27 43	73.23 93	82.39 152	79.08 162	87.93 117	78.97 140

v1		82.27 98	84.32 110	79.87 94	82.86 108	80.32 112	87.57 97	63.47 138	81.87 102	84.13 48	91.34 53	96.29 41	73.23 93	82.39 152	79.08 162	87.94 116	78.98 139

Effi-MVS+		82.33 97	83.87 123	80.52 87	84.51 81	81.32 99	87.53 98	68.05 84	74.94 159	79.67 92	82.37 135	92.31 112	72.21 100	85.06 117	86.91 75	91.18 75	84.20 90

pmmvs-eth3d		79.64 123	82.06 136	76.83 119	80.05 137	72.64 171	87.47 99	66.59 91	80.83 117	73.50 121	89.32 80	93.20 101	67.78 126	80.78 168	81.64 126	85.58 163	76.01 151

TranMVSNet+NR-MVSNet		85.23 62	89.38 52	80.39 90	88.78 49	83.77 74	87.40 100	76.75 32	85.47 68	68.99 143	95.18 6	97.55 19	67.13 130	91.61 52	89.13 60	93.26 47	82.95 106

v1neww		81.76 109	83.95 121	79.21 104	82.41 113	80.46 109	87.26 101	62.93 144	79.28 133	81.62 79	91.06 61	95.72 58	73.31 89	82.83 143	79.22 156	87.73 122	79.07 136

v7new		81.76 109	83.95 121	79.21 104	82.41 113	80.46 109	87.26 101	62.93 144	79.28 133	81.62 79	91.06 61	95.72 58	73.31 89	82.83 143	79.22 156	87.73 122	79.07 136

v17		82.09 104	84.45 108	79.33 100	82.41 113	81.31 100	87.26 101	64.50 122	78.72 138	80.73 85	90.90 64	95.57 61	73.37 87	83.06 138	79.25 155	87.70 125	82.35 114

v8		82.20 102	84.56 107	79.45 98	82.42 112	81.65 94	87.26 101	64.27 123	79.36 131	81.70 77	91.04 63	95.75 55	73.30 91	82.82 145	79.18 159	87.74 121	82.09 116

v6		81.77 108	83.96 120	79.22 103	82.41 113	80.45 111	87.26 101	62.91 148	79.29 132	81.65 78	91.08 59	95.74 56	73.32 88	82.84 142	79.21 158	87.73 122	79.07 136

v16		81.92 107	84.32 110	79.12 106	82.31 118	81.29 101	87.20 106	64.51 121	78.16 142	79.76 91	90.86 65	95.23 70	73.29 92	83.05 139	79.29 154	87.63 126	82.34 115

3Dnovator	79.41 10	82.21 101	86.07 84	77.71 114	79.31 146	84.61 69	87.18 107	61.02 166	85.65 66	76.11 105	85.07 122	85.38 158	70.96 110	87.22 96	86.47 81	91.66 70	88.12 66

PM-MVS		80.42 120	83.63 126	76.67 120	78.04 157	72.37 173	87.14 108	60.18 172	80.13 123	71.75 131	86.12 110	93.92 92	77.08 50	86.56 100	85.12 93	85.83 160	81.18 123

NR-MVSNet		82.89 91	87.43 71	77.59 117	83.91 88	83.59 76	87.10 109	78.35 18	80.64 118	68.85 144	92.67 31	96.50 30	54.19 179	87.19 98	88.68 62	93.16 51	82.75 109

UniMVSNet_NR-MVSNet		84.62 70	88.00 66	80.68 82	88.18 55	83.83 73	87.06 110	76.47 36	81.46 110	70.49 136	93.24 22	95.56 63	68.13 124	90.43 72	88.47 63	93.78 42	83.02 103

DU-MVS		84.88 67	88.27 63	80.92 72	88.30 53	83.59 76	87.06 110	78.35 18	80.64 118	70.49 136	92.67 31	96.91 24	68.13 124	91.79 48	89.29 59	93.20 48	83.02 103

MSLP-MVS++		86.29 53	89.10 54	83.01 54	85.71 72	89.79 32	87.04 112	74.39 47	85.17 72	78.92 97	77.59 156	93.57 96	82.60 20	93.23 35	91.88 38	89.42 93	92.46 26

v18		81.62 112	83.99 119	78.86 107	82.08 122	81.12 107	86.93 113	64.24 124	77.44 144	79.47 94	90.53 66	94.99 78	72.99 96	82.72 148	79.18 159	87.48 129	81.91 119

EG-PatchMatch MVS		84.35 71	87.55 69	80.62 85	86.38 66	82.24 86	86.75 114	64.02 129	84.24 78	78.17 101	89.38 79	95.03 77	78.78 37	89.95 76	86.33 82	89.59 90	85.65 82

UGNet		79.62 124	85.91 87	72.28 140	73.52 182	83.91 72	86.64 115	69.51 68	79.85 126	62.57 166	85.82 114	89.63 135	53.18 186	88.39 87	87.35 70	88.28 108	86.43 77
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
V42		79.59 125	83.59 127	74.93 130	69.61 200	77.05 140	86.59 116	55.84 192	78.42 141	77.29 102	89.84 73	95.08 75	74.12 76	83.05 139	80.11 143	86.12 152	81.59 120

FPMVS		81.56 113	84.04 118	78.66 108	82.92 106	75.96 153	86.48 117	65.66 104	84.67 75	71.47 132	77.78 154	83.22 163	77.57 47	91.24 58	90.21 48	87.84 119	85.21 83

FC-MVSNet-train		79.20 128	86.29 79	70.94 147	84.06 84	77.67 131	85.68 118	64.11 128	82.90 88	52.22 195	92.57 35	93.69 94	49.52 203	88.30 88	86.93 74	90.03 85	81.95 118

IS_MVSNet		81.72 111	85.01 99	77.90 113	86.19 67	82.64 83	85.56 119	70.02 65	80.11 124	63.52 161	87.28 98	81.18 170	67.26 128	91.08 64	89.33 58	94.82 32	83.42 99

TinyColmap		83.79 75	86.12 82	81.07 71	83.42 95	81.44 98	85.42 120	68.55 79	88.71 40	89.46 8	87.60 94	92.72 105	70.34 114	89.29 79	81.94 124	89.20 94	81.12 124

EPP-MVSNet		82.76 94	86.47 77	78.45 110	86.00 70	84.47 70	85.39 121	68.42 80	84.17 79	62.97 164	89.26 81	76.84 185	72.13 101	92.56 47	90.40 47	95.76 20	87.56 72

Gipuma		86.47 51	89.25 53	83.23 51	83.88 89	78.78 123	85.35 122	68.42 80	92.69 9	89.03 12	91.94 43	96.32 40	81.80 25	94.45 27	86.86 76	90.91 78	83.69 94
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
FC-MVSNet-test		75.91 146	83.59 127	66.95 182	76.63 177	69.07 183	85.33 123	64.97 115	84.87 74	41.95 217	93.17 23	87.04 151	47.78 206	91.09 63	85.56 89	85.06 168	74.34 157

MSDG		81.39 115	84.23 116	78.09 112	82.40 117	82.47 85	85.31 124	60.91 167	79.73 127	80.26 89	86.30 108	88.27 147	69.67 116	87.20 97	84.98 94	89.97 86	80.67 126

v748		85.21 63	89.62 50	80.08 92	80.71 131	80.27 116	85.05 125	63.79 132	90.47 22	83.54 61	94.21 15	98.52 2	76.84 52	90.97 67	84.25 100	90.53 81	88.62 62

Fast-Effi-MVS+-dtu		76.92 137	77.18 158	76.62 121	79.55 143	79.17 120	84.80 126	77.40 29	64.46 204	68.75 146	70.81 204	86.57 153	63.36 146	81.74 161	81.76 125	85.86 159	75.78 153

DELS-MVS		79.71 122	83.74 125	75.01 128	79.31 146	82.68 82	84.79 127	60.06 173	75.43 157	69.09 142	86.13 109	89.38 136	67.16 129	85.12 116	83.87 104	89.65 88	83.57 96
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
MIMVSNet1		73.40 157	81.85 137	63.55 196	72.90 185	64.37 198	84.58 128	53.60 202	90.84 18	53.92 184	87.75 93	96.10 45	45.31 209	85.37 114	79.32 153	70.98 203	69.18 186

ambc			88.38 59		91.62 15	87.97 48	84.48 129		88.64 41	87.93 16	87.38 96	94.82 82	74.53 72	89.14 81	83.86 105	85.94 158	86.84 74

pmmvs4		75.92 145	77.48 156	74.10 133	78.21 156	70.94 175	84.06 130	64.78 116	75.13 158	68.47 148	84.12 126	83.32 162	64.74 139	75.93 187	79.14 161	84.31 171	73.77 165

MDA-MVSNet-bldmvs		76.51 139	82.87 132	69.09 163	50.71 232	74.72 163	84.05 131	60.27 171	81.62 107	71.16 134	88.21 90	91.58 121	69.62 117	92.78 43	77.48 171	78.75 187	73.69 167

CANet_DTU		75.04 149	78.45 146	71.07 143	77.27 167	77.96 128	83.88 132	58.00 184	64.11 205	68.67 147	75.65 178	88.37 146	53.92 181	82.05 158	81.11 129	84.67 169	79.88 133

Anonymous20231211		85.16 64	91.64 33	77.61 116	88.54 51	79.81 119	83.12 133	74.68 44	98.37 1	66.79 155	94.56 13	99.60 1	61.64 149	91.49 54	89.82 52	90.91 78	87.80 69

HyFIR lowres test		73.29 158	74.14 179	72.30 139	73.08 184	78.33 126	83.12 133	62.41 156	63.81 206	62.13 167	76.67 165	78.50 177	71.09 107	74.13 191	77.47 172	81.98 180	70.10 180

IB-MVS	71.28 17	75.21 148	77.00 161	73.12 138	76.76 171	77.45 133	83.05 135	58.92 179	63.01 209	64.31 160	59.99 227	87.57 150	68.64 122	86.26 105	82.34 122	87.05 140	82.36 113
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
CVMVSNet		75.65 147	77.62 155	73.35 137	71.95 192	69.89 179	83.04 136	60.84 168	69.12 180	68.76 145	79.92 146	78.93 176	73.64 85	81.02 166	81.01 131	81.86 181	83.43 98

USDC		81.39 115	83.07 130	79.43 99	81.48 127	78.95 122	82.62 137	66.17 95	87.45 50	90.73 4	82.40 134	93.65 95	66.57 133	83.63 136	77.97 166	89.00 96	77.45 149

EPNet_dtu		71.90 173	73.03 183	70.59 150	78.28 154	61.64 203	82.44 138	64.12 127	63.26 208	69.74 138	71.47 196	82.41 165	51.89 199	78.83 175	78.01 165	77.07 189	75.60 155
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
MDTV_nov1_ep13_2view		72.96 166	75.59 171	69.88 157	71.15 197	64.86 197	82.31 139	54.45 197	76.30 152	78.32 100	86.52 106	91.58 121	61.35 151	76.80 181	66.83 201	71.70 198	66.26 193

v148		79.33 127	82.32 135	75.84 124	80.14 136	75.74 155	81.98 140	57.06 188	81.51 109	79.36 96	89.42 77	96.42 34	71.32 105	81.54 164	75.29 180	85.20 166	76.32 150

EU-MVSNet		76.48 140	80.53 140	71.75 141	67.62 205	70.30 177	81.74 141	54.06 199	75.47 156	71.01 135	80.10 143	93.17 103	73.67 84	83.73 134	77.85 167	82.40 179	83.07 102

Vis-MVSNet (Re-imp)		76.15 143	80.84 139	70.68 149	83.66 92	74.80 162	81.66 142	69.59 66	80.48 121	46.94 211	87.44 95	80.63 172	53.14 187	86.87 99	84.56 98	89.12 95	71.12 176

DI_MVS_plusplus_trai		77.64 134	79.64 142	75.31 127	79.87 141	76.89 141	81.55 143	63.64 133	76.21 153	72.03 129	85.59 117	82.97 164	66.63 132	79.27 173	77.78 168	88.14 112	78.76 143

testgi		68.20 187	76.05 169	59.04 205	79.99 138	67.32 191	81.16 144	51.78 208	84.91 73	39.36 225	73.42 189	95.19 71	32.79 222	76.54 185	70.40 192	69.14 207	64.55 197

PatchMatch-RL		76.05 144	76.64 164	75.36 126	77.84 163	69.87 180	81.09 145	63.43 139	71.66 172	68.34 149	71.70 194	81.76 169	74.98 68	84.83 126	83.44 107	86.45 148	73.22 169

tpmp4_e23		68.32 186	66.04 201	70.98 145	77.52 166	69.23 182	80.99 146	65.46 108	68.09 185	69.25 141	70.77 206	54.03 225	59.35 155	69.01 211	63.02 208	73.34 195	68.15 188

IterMVS		73.62 156	76.53 165	70.23 154	71.83 193	77.18 138	80.69 147	53.22 204	72.23 169	66.62 156	85.21 119	78.96 175	69.54 118	76.28 186	71.63 189	79.45 184	74.25 160
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
canonicalmvs		81.22 117	86.04 85	75.60 125	83.17 103	83.18 79	80.29 148	65.82 102	85.97 65	67.98 151	77.74 155	91.51 123	65.17 136	88.62 84	86.15 84	91.17 76	89.09 57

GA-MVS		75.01 150	76.39 166	73.39 135	78.37 153	75.66 157	80.03 149	58.40 182	70.51 176	75.85 107	83.24 130	76.14 189	63.75 141	77.28 180	76.62 175	83.97 172	75.30 156

OpenMVS	75.38 16	78.44 131	81.39 138	74.99 129	80.46 133	79.85 118	79.99 150	58.31 183	77.34 146	73.85 120	77.19 161	82.33 168	68.60 123	84.67 127	81.95 123	88.72 99	86.40 78

MVS_Test		76.72 138	79.40 144	73.60 134	78.85 151	74.99 160	79.91 151	61.56 163	69.67 178	72.44 125	85.98 112	90.78 130	63.50 144	78.30 176	75.74 179	85.33 165	80.31 131

IterMVS-LS		79.79 121	82.56 133	76.56 122	81.83 125	77.85 130	79.90 152	69.42 71	78.93 137	71.21 133	90.47 67	85.20 159	70.86 111	80.54 170	80.57 135	86.15 151	84.36 87
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
conf0.05thres1000		77.12 136	82.38 134	70.98 145	82.30 119	77.95 129	79.86 153	64.74 117	86.63 57	53.93 183	85.74 115	75.63 195	56.85 163	88.98 82	84.10 101	88.20 110	77.61 148

QAPM		80.43 119	84.34 109	75.86 123	79.40 145	82.06 88	79.86 153	61.94 161	83.28 84	74.73 116	81.74 138	85.44 157	70.97 109	84.99 124	84.71 97	88.29 107	88.14 65

PVSNet_BlendedMVS		76.45 141	78.12 149	74.49 131	76.76 171	78.46 124	79.65 155	63.26 141	65.42 200	73.15 122	75.05 182	88.96 140	66.51 134	82.73 146	77.66 169	87.61 127	78.60 144

PVSNet_Blended		76.45 141	78.12 149	74.49 131	76.76 171	78.46 124	79.65 155	63.26 141	65.42 200	73.15 122	75.05 182	88.96 140	66.51 134	82.73 146	77.66 169	87.61 127	78.60 144

CDS-MVSNet		73.07 164	77.02 160	68.46 166	81.62 126	72.89 170	79.56 157	70.78 62	69.56 179	52.52 192	77.37 160	81.12 171	42.60 212	84.20 131	83.93 102	83.65 174	70.07 181
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
CostFormer		66.81 192	66.94 199	66.67 184	72.79 186	68.25 187	79.55 158	55.57 193	65.52 199	62.77 165	76.98 162	60.09 214	56.73 165	65.69 221	62.35 209	72.59 196	69.71 182

diffmvs		73.65 155	77.10 159	69.63 159	73.21 183	69.52 181	79.35 159	57.48 185	73.80 161	68.08 150	87.10 101	82.39 166	61.36 150	74.27 190	74.51 181	78.31 188	78.14 146

tfpn		72.99 165	75.25 174	70.36 153	81.87 124	77.09 139	79.28 160	64.16 126	79.58 129	53.14 187	76.97 163	48.75 229	56.35 167	87.31 94	82.75 116	87.35 134	74.31 158

CHOSEN 1792x2688		68.80 184	71.09 186	66.13 188	69.11 202	68.89 185	78.98 161	54.68 194	61.63 216	56.69 173	71.56 195	78.39 178	67.69 127	72.13 202	72.01 188	69.63 206	73.02 170

view800		74.68 151	78.74 145	69.94 156	81.12 129	76.59 142	78.94 162	63.24 143	78.56 140	53.06 188	75.61 179	76.26 188	56.07 168	86.32 103	83.75 106	87.18 139	74.10 161

thres600view7		74.34 153	78.43 147	69.56 160	80.47 132	76.28 148	78.65 163	62.56 153	77.39 145	52.53 191	74.03 187	76.78 186	55.90 170	85.06 117	85.19 92	87.25 137	74.29 159

view600		74.08 154	78.15 148	69.32 162	80.27 135	75.82 154	78.27 164	62.20 157	77.26 147	52.80 190	74.07 186	76.86 184	55.57 173	84.90 125	84.43 99	86.84 141	73.71 166

FMVSNet1		78.20 133	84.83 104	70.46 152	78.62 152	79.03 121	77.90 165	67.53 88	83.02 87	55.10 179	87.19 100	93.18 102	55.65 171	85.57 109	83.39 108	87.98 115	82.40 112

pmmvs6		80.46 118	88.34 62	71.26 142	81.96 123	77.51 132	77.54 166	68.83 75	93.72 6	55.92 176	93.94 18	98.03 12	55.94 169	89.21 80	85.61 88	87.36 133	80.38 127

TransMVSNet (Re)		79.05 129	86.66 74	70.18 155	83.32 97	75.99 152	77.54 166	63.98 130	90.68 20	55.84 177	94.80 8	96.06 47	53.73 185	86.27 104	83.22 112	86.65 142	79.61 134

thres400		73.13 163	76.99 162	68.62 165	79.46 144	74.93 161	77.23 168	61.23 164	75.54 155	52.31 194	72.20 193	77.10 183	54.89 175	82.92 141	82.62 121	86.57 145	73.66 168

MS-PatchMatch		71.18 178	73.99 180	67.89 176	77.16 168	71.76 174	77.18 169	56.38 191	67.35 186	55.04 180	74.63 184	75.70 190	62.38 148	76.62 183	75.97 178	79.22 185	75.90 152

tfpn111		71.60 174	74.66 177	68.04 170	77.97 158	76.44 144	77.04 170	62.68 149	66.81 190	50.69 202	62.10 222	75.67 191	52.46 194	85.06 117	82.64 117	87.42 130	73.87 162

conf0.01		69.59 181	71.01 187	67.95 172	77.74 164	76.09 150	77.04 170	62.58 152	66.81 190	50.54 204	63.00 220	51.78 228	52.46 194	84.53 128	82.64 117	87.32 135	72.19 173

conf0.002		68.60 185	69.17 192	67.92 175	77.66 165	76.01 151	77.04 170	62.56 153	66.81 190	50.51 205	61.21 225	44.01 233	52.46 194	84.44 129	80.29 139	87.31 136	71.44 175

conf200view11		72.00 172	75.40 172	68.04 170	77.97 158	76.44 144	77.04 170	62.68 149	66.81 190	50.69 202	67.30 211	75.67 191	52.46 194	85.06 117	82.64 117	87.42 130	73.87 162

tfpn200view9		72.01 171	75.40 172	68.06 169	77.97 158	76.44 144	77.04 170	62.67 151	66.81 190	50.82 200	67.30 211	75.67 191	52.46 194	85.06 117	82.64 117	87.41 132	73.86 164

DWT-MVSNet_training		63.07 200	60.04 222	66.61 185	71.64 194	65.27 196	76.80 175	53.82 200	55.90 225	63.07 163	62.23 221	41.87 235	62.54 147	64.32 224	63.71 206	71.78 197	66.97 190

no-one		78.59 130	85.28 94	70.79 148	59.01 221	68.77 186	76.62 176	46.06 214	80.25 122	75.75 108	81.85 137	97.75 15	83.63 12	90.99 66	87.20 72	83.67 173	90.14 48

tfpn1000		72.27 170	76.88 163	66.88 183	79.01 150	74.04 164	76.60 177	61.15 165	79.65 128	45.52 213	77.41 159	67.98 208	52.47 193	85.22 115	82.99 113	86.54 146	70.89 177

tfpnnormal		77.16 135	84.26 114	68.88 164	81.02 130	75.02 159	76.52 178	63.30 140	87.29 51	52.40 193	91.24 58	93.97 91	54.85 177	85.46 112	81.08 130	85.18 167	75.76 154

tfpn_n400		73.26 159	77.94 151	67.79 177	79.91 139	73.32 166	76.38 179	62.04 158	84.26 76	48.53 207	76.23 169	71.50 202	53.83 182	86.22 106	81.59 127	86.05 153	72.47 171

tfpnconf		73.26 159	77.94 151	67.79 177	79.91 139	73.32 166	76.38 179	62.04 158	84.26 76	48.53 207	76.23 169	71.50 202	53.83 182	86.22 106	81.59 127	86.05 153	72.47 171

pm-mvs1		78.21 132	85.68 89	69.50 161	80.38 134	75.73 156	76.25 181	65.04 113	87.59 47	54.47 182	93.16 24	95.99 51	54.20 178	86.37 102	82.98 114	86.64 143	77.96 147

tfpnview11		72.88 167	77.37 157	67.65 179	79.81 142	73.43 165	76.23 182	61.97 160	81.37 113	48.53 207	76.23 169	71.50 202	53.78 184	85.45 113	82.77 115	85.56 164	70.87 179

tpm cat1		64.79 199	62.74 215	67.17 180	74.61 181	65.91 194	76.18 183	59.32 176	64.88 203	66.41 157	71.21 199	53.56 226	59.17 156	61.53 227	58.16 220	67.33 210	63.95 198

RPMNet		67.02 191	63.99 209	70.56 151	71.55 195	67.63 188	75.81 184	69.44 70	59.93 219	63.24 162	64.32 216	47.51 230	59.68 153	70.37 208	69.64 195	83.64 175	68.49 187

MDTV_nov1_ep13		64.96 197	64.77 206	65.18 194	67.08 208	62.46 202	75.80 185	51.10 211	62.27 215	69.74 138	74.12 185	62.65 211	55.64 172	68.19 213	62.16 213	71.70 198	61.57 208

GBi-Net		73.17 161	77.64 153	67.95 172	76.76 171	77.36 134	75.77 186	64.57 118	62.99 210	51.83 196	76.05 172	77.76 180	52.73 190	85.57 109	83.39 108	86.04 155	80.37 128

test1		73.17 161	77.64 153	67.95 172	76.76 171	77.36 134	75.77 186	64.57 118	62.99 210	51.83 196	76.05 172	77.76 180	52.73 190	85.57 109	83.39 108	86.04 155	80.37 128

FMVSNet2		74.43 152	79.70 141	68.27 167	76.76 171	77.36 134	75.77 186	65.36 110	72.28 168	52.97 189	81.92 136	85.61 156	52.73 190	80.66 169	79.73 144	86.04 155	80.37 128

MVSTER		68.08 189	69.73 190	66.16 187	66.33 212	70.06 178	75.71 189	52.36 206	55.18 228	58.64 170	70.23 208	56.72 220	57.34 162	79.68 172	76.03 177	86.61 144	80.20 132

test20.03		69.91 179	76.20 168	62.58 198	84.01 87	67.34 190	75.67 190	65.88 101	79.98 125	40.28 222	82.65 132	89.31 138	39.63 216	77.41 179	73.28 184	69.98 204	63.40 201

thres200		72.41 169	76.00 170	68.21 168	78.28 154	76.28 148	74.94 191	62.56 153	72.14 171	51.35 199	69.59 209	76.51 187	54.89 175	85.06 117	80.51 137	87.25 137	71.92 174

FMVSNet3		71.40 177	75.20 176	66.97 181	75.00 179	76.59 142	74.29 192	64.57 118	62.99 210	51.83 196	76.05 172	77.76 180	51.49 200	76.58 184	77.03 174	84.62 170	79.43 135

Anonymous20231206		67.28 190	73.41 182	60.12 204	76.45 178	63.61 201	74.21 193	56.52 190	76.35 151	42.23 216	75.81 177	90.47 132	41.51 215	74.52 188	69.97 194	69.83 205	63.17 202

testmv		60.72 210	68.44 196	51.71 220	61.76 216	56.70 214	73.40 194	42.24 219	67.31 188	39.54 224	70.88 202	92.49 109	28.75 225	73.83 194	66.00 202	64.56 216	51.89 223

test1235678		60.73 209	68.46 195	51.71 220	61.76 216	56.73 213	73.40 194	42.24 219	67.34 187	39.55 223	70.90 201	92.54 107	28.75 225	73.84 193	66.00 202	64.57 215	51.90 222

thres100view900		69.86 180	72.97 184	66.24 186	77.97 158	72.49 172	73.29 196	59.12 177	66.81 190	50.82 200	67.30 211	75.67 191	50.54 202	78.24 177	79.40 152	85.71 162	70.88 178

PatchmatchNet		64.81 198	63.74 211	66.06 190	69.21 201	58.62 208	73.16 197	60.01 174	65.92 196	66.19 158	76.27 167	59.09 215	60.45 152	66.58 218	61.47 216	67.33 210	58.24 214
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
test-mter		59.39 213	61.59 217	56.82 209	53.21 227	54.82 215	73.12 198	26.57 232	53.19 229	56.31 174	64.71 214	60.47 213	56.36 166	68.69 212	64.27 205	75.38 191	65.00 194

tfpn_ndepth		68.20 187	72.18 185	63.55 196	74.64 180	73.24 168	72.41 199	59.76 175	70.54 175	41.93 218	60.96 226	68.69 207	46.23 208	82.16 155	80.14 142	86.34 150	69.56 183

testpf		55.64 222	50.84 231	61.24 201	67.03 209	54.45 216	72.29 200	65.04 113	37.23 234	54.99 181	53.99 229	43.12 234	44.34 210	55.22 232	51.59 230	63.76 217	60.25 210

pmmvs5		68.91 183	74.35 178	62.56 199	67.45 207	66.78 192	71.70 201	51.47 209	67.17 189	56.25 175	82.41 133	88.59 144	47.21 207	73.21 200	74.23 182	81.30 182	68.03 189

tpm		62.79 203	63.25 212	62.26 200	70.09 199	53.78 217	71.65 202	47.31 213	65.72 198	76.70 103	80.62 140	56.40 222	48.11 205	64.20 225	58.54 218	59.70 222	63.47 200

thresconf0.02		66.71 193	68.28 198	64.89 195	76.83 170	70.38 176	71.62 203	58.90 180	77.64 143	47.04 210	62.10 222	46.01 231	51.32 201	78.85 174	76.09 176	83.62 176	66.85 191

dps		65.14 196	64.50 207	65.89 191	71.41 196	65.81 195	71.44 204	61.59 162	58.56 222	61.43 168	75.45 180	52.70 227	58.06 160	69.57 210	64.65 204	71.39 201	64.77 195

gg-mvs-nofinetune		72.68 168	75.21 175	69.73 158	81.48 127	69.04 184	70.48 205	76.67 33	86.92 55	67.80 152	88.06 91	64.67 210	42.12 214	77.60 178	73.65 183	79.81 183	66.57 192

PatchT		66.25 194	66.76 200	65.67 192	55.87 226	60.75 205	70.17 206	59.00 178	59.80 221	72.30 126	78.68 151	54.12 224	65.04 137	71.64 204	72.91 185	71.63 200	69.40 184

pmmvs3		62.72 204	68.71 194	55.74 211	50.74 231	57.10 210	70.05 207	28.82 230	61.57 218	57.39 172	71.19 200	85.73 154	53.96 180	73.36 199	69.43 196	73.47 194	62.55 204

MIMVSNet		63.02 201	69.02 193	56.01 210	68.20 203	59.26 207	70.01 208	53.79 201	71.56 173	41.26 221	71.38 197	82.38 167	36.38 218	71.43 206	67.32 199	66.45 212	59.83 211

1111		55.38 223	59.51 225	50.57 222	72.41 190	48.16 227	69.76 209	57.08 186	76.79 149	32.10 230	80.12 141	35.41 237	25.87 227	67.23 214	57.74 221	46.17 232	51.09 225

.test1245		43.71 230	44.35 232	42.95 227	72.41 190	48.16 227	69.76 209	57.08 186	76.79 149	32.10 230	80.12 141	35.41 237	25.87 227	67.23 214	1.08 234	0.48 237	1.68 234

TAMVS		63.02 201	69.30 191	55.70 212	70.12 198	56.89 211	69.63 211	45.13 215	70.23 177	38.00 227	77.79 153	75.15 196	42.60 212	74.48 189	72.81 187	68.70 208	57.75 216

LP		65.71 195	69.91 189	60.81 203	56.75 225	61.37 204	69.55 212	56.80 189	73.01 163	60.48 169	79.76 147	70.57 205	55.47 174	72.77 201	67.19 200	65.81 213	64.71 196

PMMVS		61.98 207	65.61 203	57.74 207	45.03 233	51.76 224	69.54 213	35.05 225	55.49 227	55.32 178	68.23 210	78.39 178	58.09 159	70.21 209	71.56 190	83.42 178	63.66 199

test-LLR		62.15 206	59.46 226	65.29 193	79.07 148	52.66 220	69.46 214	62.93 144	50.76 232	53.81 185	63.11 218	58.91 216	52.87 188	66.54 219	62.34 210	73.59 192	61.87 206

TESTMET0.1,1		57.21 217	59.46 226	54.60 215	50.95 230	52.66 220	69.46 214	26.91 231	50.76 232	53.81 185	63.11 218	58.91 216	52.87 188	66.54 219	62.34 210	73.59 192	61.87 206

CHOSEN 280x420		56.32 221	58.85 228	53.36 216	51.63 229	39.91 234	69.12 216	38.61 224	56.29 224	36.79 228	48.84 232	62.59 212	63.39 145	73.61 197	67.66 198	60.61 220	63.07 203

testus		57.41 216	64.98 205	48.58 225	59.39 220	57.17 209	68.81 217	32.86 227	62.32 214	43.25 215	57.59 228	88.49 145	24.19 231	71.68 203	63.20 207	62.99 218	54.42 219

test0.0.03 1		61.79 208	65.33 204	57.65 208	79.07 148	64.09 199	68.51 218	62.93 144	61.59 217	33.71 229	61.58 224	71.58 201	33.43 221	70.95 207	68.68 197	68.26 209	58.82 212

tpmrst		59.42 212	60.02 223	58.71 206	67.56 206	53.10 219	66.99 219	51.88 207	63.80 207	57.68 171	76.73 164	56.49 221	48.73 204	56.47 231	55.55 223	59.43 223	58.02 215

FMVSNet5		56.37 220	60.14 221	51.98 219	60.83 218	59.58 206	66.85 220	42.37 218	52.68 230	41.33 220	47.09 233	54.68 223	35.28 219	73.88 192	70.77 191	65.24 214	62.26 205

test12356		54.63 225	63.78 210	43.96 226	51.77 228	51.90 223	65.92 221	30.12 228	62.44 213	30.38 232	64.65 215	89.07 139	30.62 223	73.53 198	62.11 214	54.92 226	42.78 230

CR-MVSNet		69.56 182	68.34 197	70.99 144	72.78 187	67.63 188	64.47 222	67.74 86	59.93 219	72.30 126	80.10 143	56.77 219	65.04 137	71.64 204	72.91 185	83.61 177	69.40 184

Patchmtry						56.88 212	64.47 222	67.74 86		72.30 126

test2356		51.28 228	53.40 230	48.80 224	58.53 223	52.10 222	63.63 224	40.83 222	51.94 231	39.35 226	53.46 230	45.22 232	28.78 224	64.39 223	60.77 217	61.70 219	45.92 228

CMPMVS	55.74 18	71.56 175	76.26 167	66.08 189	68.11 204	63.91 200	63.17 225	50.52 212	68.79 183	75.49 109	70.78 205	85.67 155	63.54 143	81.58 162	77.20 173	75.63 190	85.86 80
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
EMVS		58.97 215	62.63 216	54.70 214	66.26 213	48.71 226	61.74 226	42.71 217	72.80 167	46.00 212	73.01 192	71.66 199	57.91 161	80.41 171	50.68 231	53.55 229	41.11 232

EPMVS		56.62 219	59.77 224	52.94 217	62.41 215	50.55 225	60.66 227	52.83 205	65.15 202	41.80 219	77.46 158	57.28 218	42.68 211	59.81 229	54.82 224	57.23 225	53.35 220

E-PMN		59.07 214	62.79 214	54.72 213	67.01 210	47.81 230	60.44 228	43.40 216	72.95 165	44.63 214	70.42 207	73.17 198	58.73 158	80.97 167	51.98 228	54.14 228	42.26 231

ADS-MVSNet		56.89 218	61.09 218	52.00 218	59.48 219	48.10 229	58.02 229	54.37 198	72.82 166	49.19 206	75.32 181	65.97 209	37.96 217	59.34 230	54.66 225	52.99 230	51.42 224

N_pmnet		54.95 224	65.90 202	42.18 228	66.37 211	43.86 233	57.92 230	39.79 223	79.54 130	17.24 237	86.31 107	87.91 148	25.44 229	64.68 222	51.76 229	46.33 231	47.23 227

MVS-HIRNet		59.74 211	58.74 229	60.92 202	57.74 224	45.81 231	56.02 231	58.69 181	55.69 226	65.17 159	70.86 203	71.66 199	56.75 164	61.11 228	53.74 226	71.17 202	52.28 221

GG-mvs-BLEND		41.63 231	60.36 220	19.78 232	0.14 239	66.04 193	55.66 232	0.17 237	57.64 223	2.42 240	51.82 231	69.42 206	0.28 237	64.11 226	58.29 219	60.02 221	55.18 218

new-patchmatchnet		62.59 205	73.79 181	49.53 223	76.98 169	53.57 218	53.46 233	54.64 195	85.43 69	28.81 233	91.94 43	96.41 35	25.28 230	76.80 181	53.66 227	57.99 224	58.69 213

PMMVS2		48.13 229	64.06 208	29.55 231	44.06 234	36.69 235	51.95 234	29.97 229	74.75 160	8.90 239	76.02 175	91.24 127	7.53 233	73.78 195	55.91 222	34.87 234	40.01 233

new_pmnet		52.29 226	63.16 213	39.61 230	58.89 222	44.70 232	48.78 235	34.73 226	65.88 197	17.85 236	73.42 189	80.00 173	23.06 232	67.00 217	62.28 212	54.36 227	48.81 226

MVE	41.12 19	51.80 227	60.92 219	41.16 229	35.21 235	34.14 236	48.45 236	41.39 221	69.11 181	19.53 235	63.33 217	73.80 197	63.56 142	67.19 216	61.51 215	38.85 233	57.38 217
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
DeepMVS_CX						17.78 237	20.40 237	6.69 233	31.41 235	9.80 238	38.61 234	34.88 239	33.78 220	28.41 234		23.59 235	45.77 229

tmp_tt				13.54 233	16.73 236	6.42 238	8.49 238	2.36 234	28.69 236	27.44 234	18.40 235	13.51 240	3.70 234	33.23 233	36.26 232	22.54 236

Patchmatch-RL test							4.13 239

testmvs		0.93 233	1.37 234	0.41 235	0.36 238	0.36 240	0.62 240	0.39 235	1.48 237	0.18 242	2.41 236	1.31 242	0.41 236	1.25 236	1.08 234	0.48 237	1.68 234

test123		1.06 232	1.41 233	0.64 234	0.39 237	0.48 239	0.52 241	0.25 236	1.11 238	1.37 241	2.01 237	1.98 241	0.87 235	1.43 235	1.27 233	0.46 239	1.62 236

sosnet-low-res		0.00 234	0.00 235	0.00 236	0.00 240	0.00 241	0.00 242	0.00 238	0.00 239	0.00 243	0.00 238	0.00 243	0.00 238	0.00 237	0.00 236	0.00 240	0.00 237

sosnet		0.00 234	0.00 235	0.00 236	0.00 240	0.00 241	0.00 242	0.00 238	0.00 239	0.00 243	0.00 238	0.00 243	0.00 238	0.00 237	0.00 236	0.00 240	0.00 237

MTAPA										89.37 9		94.85 80

MTMP										90.54 5		95.16 72

mPP-MVS					93.05 4							95.77 54

NP-MVS									78.65 139

High-resolution multi-view benchmark