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.

MCST-MVS		93.81 13	94.06 15	93.53 13	96.79 21	96.85 16	95.95 10	91.69 12	92.20 22	87.17 28	90.83 23	93.41 4	91.96 12	94.49 20	93.50 25	97.61 1	97.12 18

ESAPD		95.35 1	95.97 1	94.63 2	97.35 6	97.95 1	97.09 2	93.48 1	93.91 9	90.13 11	96.41 2	95.14 1	92.88 4	95.64 3	94.53 8	96.86 2	98.21 2

TSAR-MVS + GP.		92.71 24	93.91 17	91.30 30	91.96 66	96.00 37	93.43 36	87.94 35	92.53 18	86.27 36	93.57 11	91.94 13	91.44 21	93.29 33	92.89 37	96.78 3	97.15 17

APDe-MVS		95.23 2	95.69 2	94.70 1	97.12 11	97.81 3	97.19 1	92.83 2	95.06 2	90.98 5	96.47 1	92.77 8	93.38 1	95.34 6	94.21 12	96.68 4	98.17 3

MVS_0304		90.88 35	91.35 33	90.34 37	93.91 46	96.79 19	94.49 29	86.54 44	86.57 50	82.85 49	81.68 47	89.70 27	87.57 48	94.64 17	93.93 17	96.67 5	96.15 37

3Dnovator	85.17 5	90.48 37	89.90 41	91.16 32	94.88 37	95.74 41	93.82 32	85.36 51	89.28 40	87.81 24	74.34 76	87.40 42	88.56 37	93.07 36	93.74 21	96.53 6	95.71 43

TSAR-MVS + MP.		94.48 7	94.97 6	93.90 8	95.53 32	97.01 11	96.69 3	90.71 18	94.24 5	90.92 6	94.97 4	92.19 11	93.03 2	94.83 12	93.60 22	96.51 7	97.97 5

HSP-MVS		94.83 3	95.37 3	94.21 5	96.82 20	97.94 2	96.69 3	92.37 7	93.97 8	90.29 9	96.16 3	93.71 3	92.70 5	94.80 13	93.13 32	96.37 8	97.90 6

QAPM		89.49 43	89.58 43	89.38 47	94.73 39	95.94 38	92.35 44	85.00 54	85.69 55	80.03 62	76.97 66	87.81 40	87.87 43	92.18 50	92.10 41	96.33 9	96.40 34

MVS_111021_HR		90.56 36	91.29 34	89.70 44	94.71 40	95.63 42	91.81 52	86.38 45	87.53 47	81.29 55	87.96 28	85.43 46	87.69 45	93.90 27	92.93 35	96.33 9	95.69 44

CANet		91.33 33	91.46 31	91.18 31	95.01 34	96.71 20	93.77 33	87.39 40	87.72 46	87.26 27	81.77 45	89.73 26	87.32 51	94.43 21	93.86 18	96.31 11	96.02 39

APD-MVS		94.37 8	94.47 12	94.26 3	97.18 9	96.99 12	96.53 5	92.68 3	92.45 20	89.96 12	94.53 7	91.63 15	92.89 3	94.58 18	93.82 19	96.31 11	97.26 14
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
SteuartSystems-ACMMP		94.06 10	94.65 8	93.38 15	96.97 16	97.36 6	96.12 7	91.78 10	92.05 24	87.34 26	94.42 8	90.87 19	91.87 14	95.47 5	94.59 7	96.21 13	97.77 8
Skip Steuart: Steuart Systems R&D Blog.
PHI-MVS		92.05 28	93.74 18	90.08 39	94.96 35	97.06 9	93.11 40	87.71 38	90.71 32	80.78 58	92.40 17	91.03 17	87.68 46	94.32 22	94.48 9	96.21 13	96.16 36

CNVR-MVS		94.37 8	94.65 8	94.04 7	97.29 7	97.11 8	96.00 8	92.43 6	93.45 12	89.85 14	90.92 21	93.04 6	92.59 7	95.77 2	94.82 4	96.11 15	97.42 13

MSLP-MVS++		92.02 30	91.40 32	92.75 20	96.01 27	95.88 40	93.73 35	89.00 28	89.89 39	90.31 8	81.28 49	88.85 33	91.45 19	92.88 40	94.24 11	96.00 16	96.76 26

NCCC		93.69 16	93.66 19	93.72 12	97.37 5	96.66 25	95.93 12	92.50 5	93.40 15	88.35 21	87.36 31	92.33 10	92.18 10	94.89 11	94.09 14	96.00 16	96.91 22

ACMMP		92.03 29	92.16 28	91.87 29	95.88 29	96.55 27	94.47 30	89.49 27	91.71 27	85.26 37	91.52 20	84.48 49	90.21 28	92.82 41	91.63 45	95.92 18	96.42 32
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
zzz-MVS		93.80 14	93.45 22	94.20 6	97.53 3	96.43 32	95.88 14	91.12 16	94.09 7	92.74 3	87.68 29	90.77 20	92.04 11	94.74 15	93.56 24	95.91 19	96.85 23

ACMMPR		93.72 15	93.94 16	93.48 14	97.07 12	96.93 13	95.78 16	90.66 20	93.88 10	89.24 17	93.53 12	89.08 32	92.24 9	93.89 28	93.50 25	95.88 20	96.73 27

HFP-MVS		94.02 11	94.22 14	93.78 10	97.25 8	96.85 16	95.81 15	90.94 17	94.12 6	90.29 9	94.09 10	89.98 25	92.52 8	93.94 26	93.49 27	95.87 21	97.10 19

PVSNet_BlendedMVS		88.19 55	88.00 54	88.42 56	92.71 62	94.82 53	89.08 75	83.81 62	84.91 57	86.38 33	79.14 55	78.11 78	82.66 68	93.05 37	91.10 48	95.86 22	94.86 54

PVSNet_Blended		88.19 55	88.00 54	88.42 56	92.71 62	94.82 53	89.08 75	83.81 62	84.91 57	86.38 33	79.14 55	78.11 78	82.66 68	93.05 37	91.10 48	95.86 22	94.86 54

DeepC-MVS_fast	88.76 1	93.10 19	93.02 26	93.19 18	97.13 10	96.51 29	95.35 21	91.19 15	93.14 17	88.14 22	85.26 37	89.49 29	91.45 19	95.17 7	95.07 2	95.85 24	96.48 30
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
PGM-MVS		92.76 22	93.03 25	92.45 24	97.03 14	96.67 24	95.73 18	87.92 36	90.15 38	86.53 32	92.97 15	88.33 38	91.69 17	93.62 31	93.03 33	95.83 25	96.41 33

SD-MVS		94.53 6	95.22 5	93.73 11	95.69 31	97.03 10	95.77 17	91.95 9	94.41 4	91.35 4	94.97 4	93.34 5	91.80 15	94.72 16	93.99 16	95.82 26	98.07 4

SMA-MVS		94.57 5	95.23 4	93.80 9	97.56 2	97.61 5	95.92 13	92.02 8	94.43 3	89.74 15	92.16 18	92.63 9	91.78 16	95.98 1	95.57 1	95.80 27	98.25 1

DeepC-MVS	87.86 3	92.26 27	91.86 30	92.73 21	96.18 24	96.87 15	95.19 23	91.76 11	92.17 23	86.58 31	81.79 44	85.85 44	90.88 25	94.57 19	94.61 6	95.80 27	97.18 15
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
XVS					93.11 54	96.70 21	91.91 49			83.95 43		88.82 34				95.79 29

X-MVStestdata					93.11 54	96.70 21	91.91 49			83.95 43		88.82 34				95.79 29

DELS-MVS		89.71 41	89.68 42	89.74 42	93.75 48	96.22 34	93.76 34	85.84 47	82.53 63	85.05 39	78.96 57	84.24 50	84.25 65	94.91 10	94.91 3	95.78 31	96.02 39
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
X-MVS		92.36 26	92.75 27	91.90 28	96.89 17	96.70 21	95.25 22	90.48 23	91.50 29	83.95 43	88.20 27	88.82 34	89.11 32	93.75 29	93.43 28	95.75 32	96.83 25

CP-MVS		93.25 18	93.26 23	93.24 17	96.84 19	96.51 29	95.52 19	90.61 21	92.37 21	88.88 18	90.91 22	89.52 28	91.91 13	93.64 30	92.78 38	95.69 33	97.09 20

MVSTER		86.03 66	86.12 68	85.93 66	88.62 106	89.93 126	89.33 71	79.91 116	81.87 67	81.35 54	81.07 50	74.91 88	80.66 85	92.13 51	90.10 66	95.68 34	92.80 85

3Dnovator+	86.06 4	91.60 31	90.86 37	92.47 23	96.00 28	96.50 31	94.70 27	87.83 37	90.49 34	89.92 13	74.68 74	89.35 30	90.66 26	94.02 24	94.14 13	95.67 35	96.85 23

OpenMVS	82.53 11	87.71 57	86.84 63	88.73 52	94.42 42	95.06 49	91.02 58	83.49 69	82.50 64	82.24 53	67.62 119	85.48 45	85.56 60	91.19 59	91.30 47	95.67 35	94.75 56

CDPH-MVS		91.14 34	92.01 29	90.11 38	96.18 24	96.18 35	94.89 26	88.80 32	88.76 43	77.88 73	89.18 26	87.71 41	87.29 52	93.13 35	93.31 30	95.62 37	95.84 41

MP-MVS		93.35 17	93.59 20	93.08 19	97.39 4	96.82 18	95.38 20	90.71 18	90.82 31	88.07 23	92.83 16	90.29 23	91.32 23	94.03 23	93.19 31	95.61 38	97.16 16

FMVSNet3		84.44 75	84.64 76	84.21 83	84.32 149	90.13 121	89.85 65	80.37 107	81.17 71	75.50 77	69.63 103	79.69 72	79.62 117	89.72 92	90.52 62	95.59 39	91.58 128

HPM-MVS++		94.60 4	94.91 7	94.24 4	97.86 1	96.53 28	96.14 6	92.51 4	93.87 11	90.76 7	93.45 13	93.84 2	92.62 6	95.11 9	94.08 15	95.58 40	97.48 11

CLD-MVS		88.66 47	88.52 48	88.82 51	91.37 68	94.22 58	92.82 43	82.08 94	88.27 45	85.14 38	81.86 43	78.53 77	85.93 59	91.17 60	90.61 59	95.55 41	95.00 51
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
ACMMP_Plus		93.94 12	94.49 11	93.30 16	97.03 14	97.31 7	95.96 9	91.30 14	93.41 14	88.55 20	93.00 14	90.33 22	91.43 22	95.53 4	94.41 10	95.53 42	97.47 12

GBi-Net		84.51 73	84.80 74	84.17 84	84.20 150	89.95 123	89.70 66	80.37 107	81.17 71	75.50 77	69.63 103	79.69 72	79.75 114	90.73 78	90.72 55	95.52 43	91.71 122

test1		84.51 73	84.80 74	84.17 84	84.20 150	89.95 123	89.70 66	80.37 107	81.17 71	75.50 77	69.63 103	79.69 72	79.75 114	90.73 78	90.72 55	95.52 43	91.71 122

FMVSNet2		83.87 77	83.73 80	84.05 89	84.20 150	89.95 123	89.70 66	80.21 112	79.17 91	74.89 81	65.91 125	77.49 81	79.75 114	90.87 75	91.00 52	95.52 43	91.71 122

OPM-MVS		87.56 59	85.80 70	89.62 45	93.90 47	94.09 60	94.12 31	88.18 33	75.40 113	77.30 76	76.41 67	77.93 80	88.79 34	92.20 48	90.82 54	95.40 46	93.72 69

abl_6				90.66 35	94.65 41	96.27 33	92.21 45	86.94 42	90.23 36	86.38 33	85.50 36	92.96 7	88.37 39			95.40 46	95.46 49

CSCG		92.76 22	93.16 24	92.29 25	96.30 23	97.74 4	94.67 28	88.98 30	92.46 19	89.73 16	86.67 33	92.15 12	88.69 36	92.26 46	92.92 36	95.40 46	97.89 7

canonicalmvs		89.36 44	89.92 39	88.70 53	91.38 67	95.92 39	91.81 52	82.61 86	90.37 35	82.73 51	82.09 42	79.28 75	88.30 40	91.17 60	93.59 23	95.36 49	97.04 21

IB-MVS	79.09 12	82.60 90	82.19 87	83.07 100	91.08 70	93.55 67	80.90 185	81.35 101	76.56 104	80.87 57	64.81 141	69.97 107	68.87 186	85.64 145	90.06 68	95.36 49	94.74 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
gg-mvs-nofinetune		75.64 192	77.26 167	73.76 197	87.92 110	92.20 88	87.32 102	64.67 216	51.92 222	35.35 229	46.44 215	77.05 83	71.97 171	92.64 43	91.02 51	95.34 51	89.53 151

Vis-MVSNet		84.38 76	86.68 66	81.70 119	87.65 115	94.89 51	88.14 89	80.90 105	74.48 123	68.23 115	77.53 63	80.72 64	69.98 182	92.68 42	91.90 42	95.33 52	94.58 60
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
TAPA-MVS	84.37 7	88.91 46	88.93 46	88.89 50	93.00 57	94.85 52	92.00 48	84.84 55	91.68 28	80.05 61	79.77 53	84.56 48	88.17 41	90.11 87	89.00 97	95.30 53	92.57 95
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
MVS_Test		86.93 61	87.24 61	86.56 65	90.10 95	93.47 68	90.31 61	80.12 113	83.55 61	78.12 69	79.58 54	79.80 70	85.45 61	90.17 86	90.59 60	95.29 54	93.53 71

LGP-MVS_train		88.25 54	88.55 47	87.89 59	92.84 60	93.66 65	93.35 37	85.22 53	85.77 53	74.03 84	86.60 34	76.29 84	86.62 56	91.20 58	90.58 61	95.29 54	95.75 42

UniMVSNet_NR-MVSNet		81.87 99	81.33 97	82.50 104	85.31 137	91.30 98	85.70 139	84.25 58	75.89 108	64.21 150	66.95 122	64.65 126	80.22 99	87.07 119	89.18 92	95.27 56	94.29 61

IS_MVSNet		86.18 64	88.18 52	83.85 91	91.02 71	94.72 55	87.48 98	82.46 88	81.05 75	70.28 96	76.98 65	82.20 60	76.65 144	93.97 25	93.38 29	95.18 57	94.97 52

train_agg		92.87 21	93.53 21	92.09 26	96.88 18	95.38 44	95.94 11	90.59 22	90.65 33	83.65 46	94.31 9	91.87 14	90.30 27	93.38 32	92.42 39	95.17 58	96.73 27

NR-MVSNet		80.25 123	79.98 121	80.56 145	85.20 139	90.94 104	85.65 141	83.58 67	75.74 110	61.36 178	65.30 133	56.75 195	72.38 170	88.46 107	88.80 99	95.16 59	93.87 67

PVSNet_Blended_VisFu		87.40 60	87.80 56	86.92 64	92.86 58	95.40 43	88.56 87	83.45 72	79.55 88	82.26 52	74.49 75	84.03 51	79.24 129	92.97 39	91.53 46	95.15 60	96.65 29

tfpn		81.79 102	80.06 118	83.82 92	90.61 76	92.91 83	87.62 96	82.34 90	73.66 136	67.46 119	64.99 138	55.50 201	79.77 113	91.12 72	89.62 80	95.14 61	92.59 92

EPP-MVSNet		86.55 62	87.76 58	85.15 69	90.52 80	94.41 56	87.24 106	82.32 91	81.79 68	73.60 86	78.57 59	82.41 58	82.07 72	91.23 56	90.39 63	95.14 61	95.48 48

TranMVSNet+NR-MVSNet		80.52 119	79.84 123	81.33 129	84.92 146	90.39 111	85.53 144	84.22 60	74.27 126	60.68 183	64.93 140	59.96 174	77.48 140	86.75 128	89.28 88	95.12 63	93.29 72

conf0.002		82.54 92	80.83 110	84.54 73	90.28 92	93.24 72	89.05 77	82.75 76	75.14 114	69.75 99	67.99 116	57.12 193	80.38 92	91.16 63	89.79 75	95.02 64	91.36 131

view800		82.38 95	80.93 107	84.06 87	90.59 78	92.96 81	88.11 90	82.44 89	73.92 130	68.10 116	65.07 137	61.64 155	80.10 103	91.17 60	89.24 90	95.01 65	92.56 96

conf0.01		82.64 89	81.02 102	84.53 75	90.30 87	93.22 73	89.05 77	82.75 76	75.14 114	69.69 100	67.15 121	59.19 182	80.38 92	91.16 63	89.51 81	95.00 66	91.76 120

UniMVSNet (Re)		81.22 109	81.08 101	81.39 126	85.35 136	91.76 93	84.93 152	82.88 74	76.13 107	65.02 146	64.94 139	63.09 137	75.17 151	87.71 113	89.04 95	94.97 67	94.88 53

FC-MVSNet-train		85.18 72	85.31 72	85.03 70	90.67 75	91.62 95	87.66 95	83.61 64	79.75 85	74.37 83	78.69 58	71.21 104	78.91 131	91.23 56	89.96 71	94.96 68	94.69 59

tfpn1000		81.03 112	81.70 90	80.25 152	90.18 93	91.35 96	83.96 160	81.15 104	78.00 98	62.11 167	73.37 87	65.75 120	69.17 185	88.68 104	87.44 115	94.93 69	87.29 170

tfpn111		83.51 81	82.68 84	84.47 77	90.30 87	93.09 75	89.05 77	82.72 78	75.14 114	69.49 104	74.24 77	63.13 133	80.38 92	91.15 65	89.51 81	94.91 70	92.50 101

tfpnview11		80.84 115	81.10 100	80.54 146	90.10 95	90.96 102	85.44 146	81.84 96	75.77 109	59.27 189	73.54 83	64.40 127	71.69 174	89.16 97	87.97 108	94.91 70	85.92 184

conf200view11		82.85 86	81.46 93	84.47 77	90.30 87	93.09 75	89.05 77	82.72 78	75.14 114	69.49 104	65.72 127	63.13 133	80.38 92	91.15 65	89.51 81	94.91 70	92.50 101

tfpn200view9		82.86 85	81.46 93	84.48 76	90.30 87	93.09 75	89.05 77	82.71 80	75.14 114	69.56 101	65.72 127	63.13 133	80.38 92	91.15 65	89.51 81	94.91 70	92.50 101

thres600view7		82.53 93	81.02 102	84.28 81	90.61 76	93.05 78	88.57 85	82.67 82	74.12 129	68.56 113	65.09 136	62.13 153	80.40 91	91.15 65	89.02 96	94.88 74	92.59 92

thres200		82.77 87	81.25 98	84.54 73	90.38 84	93.05 78	89.13 73	82.67 82	74.40 124	69.53 103	65.69 130	63.03 138	80.63 86	91.15 65	89.42 86	94.88 74	92.04 111

view600		82.51 94	81.00 106	84.27 82	90.56 79	92.95 82	88.57 85	82.57 87	74.16 128	68.70 112	65.13 135	62.15 152	80.36 97	91.15 65	88.98 98	94.87 76	92.48 104

Effi-MVS+		85.33 71	85.08 73	85.63 68	89.69 102	93.42 69	89.90 64	80.31 111	79.32 89	72.48 92	73.52 86	74.03 91	86.55 57	90.99 73	89.98 70	94.83 77	94.27 63

thresconf0.02		81.14 111	80.93 107	81.39 126	90.01 99	91.31 97	86.79 126	82.28 92	76.97 103	61.46 176	74.24 77	62.08 154	72.98 169	88.70 103	87.90 109	94.81 78	85.28 187

thres400		82.68 88	81.15 99	84.47 77	90.52 80	92.89 84	88.95 83	82.71 80	74.33 125	69.22 108	65.31 132	62.61 142	80.63 86	90.96 74	89.50 85	94.79 79	92.45 106

ACMP	83.90 8	88.32 53	88.06 53	88.62 54	92.18 64	93.98 62	91.28 57	85.24 52	86.69 49	81.23 56	85.62 35	75.13 87	87.01 54	89.83 90	89.77 77	94.79 79	95.43 50
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
DU-MVS		81.20 110	80.30 116	82.25 106	84.98 144	90.94 104	85.70 139	83.58 67	75.74 110	64.21 150	65.30 133	59.60 179	80.22 99	86.89 123	89.31 87	94.77 81	94.29 61

UA-Net		86.07 65	87.78 57	84.06 87	92.85 59	95.11 48	87.73 94	84.38 57	73.22 139	73.18 88	79.99 52	89.22 31	71.47 177	93.22 34	93.03 33	94.76 82	90.69 143

MVS_111021_LR		90.14 40	90.89 36	89.26 48	93.23 53	94.05 61	90.43 60	84.65 56	90.16 37	84.52 42	90.14 24	83.80 53	87.99 42	92.50 44	90.92 53	94.74 83	94.70 58

thres100view900		82.55 91	81.01 105	84.34 80	90.30 87	92.27 87	89.04 82	82.77 75	75.14 114	69.56 101	65.72 127	63.13 133	79.62 117	89.97 89	89.26 89	94.73 84	91.61 127

AdaColmap		90.29 38	88.38 50	92.53 22	96.10 26	95.19 47	92.98 41	91.40 13	89.08 42	88.65 19	78.35 60	81.44 62	91.30 24	90.81 77	90.21 65	94.72 85	93.59 70

PCF-MVS	84.60 6	88.66 47	87.75 59	89.73 43	93.06 56	96.02 36	93.22 39	90.00 25	82.44 65	80.02 63	77.96 61	85.16 47	87.36 50	88.54 106	88.54 102	94.72 85	95.61 46
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
DI_MVS_plusplus_trai		86.41 63	85.54 71	87.42 62	89.24 103	93.13 74	92.16 47	82.65 84	82.30 66	80.75 59	68.30 115	80.41 65	85.01 62	90.56 83	90.07 67	94.70 87	94.01 65

conf0.05thres1000		81.00 113	79.12 130	83.20 98	90.14 94	92.15 90	87.05 118	82.09 93	68.11 181	66.19 129	59.67 181	61.10 166	79.05 130	90.47 85	89.11 93	94.68 88	93.22 73

Fast-Effi-MVS+		83.77 79	82.98 82	84.69 72	87.98 109	91.87 92	88.10 91	77.70 146	78.10 97	73.04 89	69.13 110	68.51 113	86.66 55	90.49 84	89.85 74	94.67 89	92.88 81

WR-MVS_H		75.84 190	76.93 173	74.57 196	82.86 185	89.50 137	78.34 198	79.36 129	66.90 185	52.51 203	60.20 178	59.71 176	59.73 205	83.61 179	85.77 160	94.65 90	92.84 83

OMC-MVS		90.23 39	90.40 38	90.03 40	93.45 51	95.29 45	91.89 51	86.34 46	93.25 16	84.94 40	81.72 46	86.65 43	88.90 33	91.69 53	90.27 64	94.65 90	93.95 66

tfpn_ndepth		81.77 104	82.29 86	81.15 134	89.79 101	91.71 94	85.49 145	81.63 100	79.17 91	64.76 148	73.04 88	68.14 117	70.62 180	88.72 102	87.88 111	94.63 92	87.38 168

HQP-MVS		89.13 45	89.58 43	88.60 55	93.53 50	93.67 64	93.29 38	87.58 39	88.53 44	75.50 77	87.60 30	80.32 66	87.07 53	90.66 82	89.95 72	94.62 93	96.35 35

PEN-MVS		76.02 187	76.07 179	75.95 186	83.17 179	87.97 158	79.65 189	80.07 115	66.57 187	51.45 206	60.94 162	55.47 202	66.81 195	82.72 186	86.80 124	94.59 94	92.03 112

ACMM	83.27 10	87.68 58	86.09 69	89.54 46	93.26 52	92.19 89	91.43 55	86.74 43	86.02 52	82.85 49	75.63 70	75.14 86	88.41 38	90.68 81	89.99 69	94.59 94	92.97 79
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
CANet_DTU		85.43 70	87.72 60	82.76 103	90.95 74	93.01 80	89.99 63	75.46 172	82.67 62	64.91 147	83.14 40	80.09 67	80.68 84	92.03 52	91.03 50	94.57 96	92.08 109

tfpn_n400		80.63 117	80.79 111	80.43 149	90.02 97	91.08 100	85.34 148	81.79 98	72.93 142	59.27 189	73.54 83	64.40 127	71.61 175	89.05 99	88.21 105	94.56 97	86.32 179

tfpnconf		80.63 117	80.79 111	80.43 149	90.02 97	91.08 100	85.34 148	81.79 98	72.93 142	59.27 189	73.54 83	64.40 127	71.61 175	89.05 99	88.21 105	94.56 97	86.32 179

CP-MVSNet		76.36 183	76.41 176	76.32 183	82.73 188	88.64 149	79.39 191	79.62 124	67.21 182	53.70 200	60.72 167	55.22 203	67.91 191	83.52 180	86.34 137	94.55 99	93.19 74

EG-PatchMatch MVS		76.40 182	75.47 190	77.48 170	85.86 130	90.22 119	82.45 169	73.96 179	59.64 212	59.60 187	52.75 206	62.20 151	68.44 188	88.23 109	87.50 114	94.55 99	87.78 165

tfpnnormal		77.46 168	74.86 195	80.49 147	86.34 127	88.92 148	84.33 158	81.26 102	61.39 207	61.70 173	51.99 208	53.66 208	74.84 154	88.63 105	87.38 118	94.50 101	92.08 109

Vis-MVSNet (Re-imp)		83.65 80	86.81 65	79.96 154	90.46 83	92.71 85	84.84 153	82.00 95	80.93 77	62.44 164	76.29 68	82.32 59	65.54 199	92.29 45	91.66 44	94.49 102	91.47 129

PS-CasMVS		75.90 189	75.86 186	75.96 185	82.59 189	88.46 154	79.23 194	79.56 126	66.00 190	52.77 202	59.48 183	54.35 206	67.14 194	83.37 183	86.23 138	94.47 103	93.10 76

Baseline_NR-MVSNet		79.84 130	78.37 142	81.55 124	84.98 144	86.66 169	85.06 150	83.49 69	75.57 112	63.31 158	58.22 191	60.97 167	78.00 137	86.89 123	87.13 120	94.47 103	93.15 75

FMVSNet1		81.64 106	80.61 114	82.84 102	82.36 191	89.20 142	88.67 84	79.58 125	70.79 156	72.63 91	58.95 187	72.26 101	79.34 128	90.73 78	90.72 55	94.47 103	91.62 126

DTE-MVSNet		75.14 194	75.44 191	74.80 193	83.18 178	87.19 164	78.25 200	80.11 114	66.05 189	48.31 212	60.88 166	54.67 204	64.54 201	82.57 188	86.17 150	94.43 106	90.53 147

PLC	83.76 9	88.61 49	86.83 64	90.70 34	94.22 43	92.63 86	91.50 54	87.19 41	89.16 41	86.87 29	75.51 71	80.87 63	89.98 30	90.01 88	89.20 91	94.41 107	90.45 148
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
UGNet		85.90 68	88.23 51	83.18 99	88.96 104	94.10 59	87.52 97	83.60 65	81.66 69	77.90 72	80.76 51	83.19 55	66.70 196	91.13 71	90.71 58	94.39 108	96.06 38
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
TransMVSNet (Re)		76.57 175	75.16 193	78.22 167	85.60 133	87.24 163	82.46 168	81.23 103	59.80 211	59.05 194	57.07 193	59.14 184	66.60 197	88.09 110	86.82 123	94.37 109	87.95 164

gm-plane-assit		70.29 204	70.65 205	69.88 205	85.03 142	78.50 214	58.41 225	65.47 212	50.39 224	40.88 221	49.60 210	50.11 214	75.14 152	91.43 55	89.78 76	94.32 110	84.73 193

CNLPA		88.40 50	87.00 62	90.03 40	93.73 49	94.28 57	89.56 69	85.81 48	91.87 25	87.55 25	69.53 108	81.49 61	89.23 31	89.45 95	88.59 101	94.31 111	93.82 68

MAR-MVS		88.39 52	88.44 49	88.33 58	94.90 36	95.06 49	90.51 59	83.59 66	85.27 56	79.07 66	77.13 64	82.89 57	87.70 44	92.19 49	92.32 40	94.23 112	94.20 64
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
MSDG		83.87 77	81.02 102	87.19 63	92.17 65	89.80 129	89.15 72	85.72 49	80.61 80	79.24 65	66.66 123	68.75 112	82.69 67	87.95 112	87.44 115	94.19 113	85.92 184

pm-mvs1		78.51 160	77.75 159	79.40 157	84.83 147	89.30 139	83.55 164	79.38 128	62.64 203	63.68 156	58.73 189	64.68 125	70.78 179	89.79 91	87.84 112	94.17 114	91.28 132

v13		78.99 152	77.25 168	81.02 138	83.87 169	86.47 178	86.60 129	75.96 169	69.87 170	66.07 135	60.25 177	61.41 163	79.49 126	85.72 142	86.22 139	94.14 115	90.84 138

v12		79.03 150	77.28 166	81.06 137	83.88 168	86.49 176	86.62 128	76.02 167	69.99 169	66.18 131	60.34 176	61.44 162	79.54 121	85.70 143	86.21 141	94.11 116	90.82 140

V9		79.08 148	77.32 165	81.14 135	83.89 164	86.52 175	86.85 124	76.06 166	70.02 168	66.42 127	60.44 173	61.52 159	79.54 121	85.68 144	86.21 141	94.08 117	90.83 139

CPTT-MVS		91.39 32	90.95 35	91.91 27	95.06 33	95.24 46	95.02 25	88.98 30	91.02 30	86.71 30	84.89 39	88.58 37	91.60 18	90.82 76	89.67 79	94.08 117	96.45 31

v144192		78.81 154	77.22 169	80.67 143	82.95 182	89.79 130	86.40 132	77.42 147	68.26 180	63.13 159	59.50 182	58.13 189	80.08 104	85.93 139	86.08 155	94.06 119	92.83 84

V14		79.11 147	77.35 163	81.16 133	83.90 161	86.54 174	86.94 120	76.10 165	70.14 167	66.41 128	60.59 170	61.54 158	79.59 120	85.64 145	86.20 144	94.04 120	90.82 140

v1921920		78.57 159	76.99 172	80.41 151	82.93 183	89.63 135	86.38 133	77.14 150	68.31 179	61.80 171	58.89 188	56.79 194	80.19 101	86.50 134	86.05 157	94.02 121	92.76 87

ACMH	78.52 14	81.86 100	80.45 115	83.51 96	90.51 82	91.22 99	85.62 142	84.23 59	70.29 162	62.21 165	69.04 112	64.05 131	84.48 64	87.57 114	88.45 104	94.01 122	92.54 99
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
WR-MVS		76.63 174	78.02 152	75.02 191	84.14 153	89.76 131	78.34 198	80.64 106	69.56 172	52.32 204	61.26 153	61.24 165	60.66 204	84.45 174	87.07 121	93.99 123	92.77 86

v15		79.13 146	77.37 161	81.19 131	83.90 161	86.56 173	87.01 119	76.15 163	70.20 165	66.48 126	60.71 168	61.55 157	79.60 119	85.59 148	86.19 146	93.98 124	90.80 142

v11		79.02 151	77.36 162	80.95 139	83.89 164	86.48 177	86.53 130	75.77 171	69.69 171	65.21 145	60.36 175	60.24 172	80.32 98	87.20 117	86.54 132	93.96 125	91.02 134

v17		79.59 140	77.88 156	81.60 123	84.03 155	86.66 169	87.13 117	76.31 159	72.09 147	68.29 114	61.15 159	62.57 143	79.90 107	85.55 150	86.20 144	93.93 126	90.93 135

v10		79.62 139	78.19 144	81.28 130	83.73 171	89.69 133	87.27 104	76.86 152	70.50 159	65.46 140	60.58 172	60.47 169	80.44 90	86.91 121	86.63 129	93.93 126	92.55 98

anonymousdsp		77.94 163	79.00 131	76.71 179	79.03 205	87.83 159	79.58 190	72.87 183	65.80 192	58.86 195	65.82 126	62.48 146	75.99 147	86.77 127	88.66 100	93.92 128	95.68 45

v1240		78.15 161	76.53 175	80.04 153	82.85 186	89.48 138	85.61 143	76.77 153	67.05 183	61.18 181	58.37 190	56.16 199	79.89 108	86.11 138	86.08 155	93.92 128	92.47 105

v16		79.65 138	77.91 155	81.69 120	84.04 154	86.65 171	87.20 108	76.32 158	72.41 146	68.71 111	61.13 160	62.52 144	79.93 106	85.55 150	86.22 139	93.92 128	90.91 137

v7		79.79 133	78.28 143	81.54 125	83.73 171	90.34 116	87.27 104	78.27 139	70.50 159	65.59 139	60.59 170	60.47 169	80.46 89	86.90 122	86.63 129	93.92 128	92.56 96

v8		79.90 129	78.39 141	81.66 121	83.97 157	89.81 128	87.16 115	77.40 148	71.49 150	67.71 117	61.24 155	62.49 145	79.83 109	85.48 158	86.17 150	93.89 132	92.02 113

v6		80.11 124	78.47 136	82.01 109	83.97 157	90.49 108	87.19 111	79.67 121	71.59 149	67.51 118	61.26 153	62.46 147	79.81 112	85.49 154	86.18 149	93.89 132	91.86 114

v1neww		80.09 125	78.45 138	82.00 110	83.97 157	90.49 108	87.18 112	79.67 121	71.49 150	67.44 120	61.24 155	62.41 148	79.83 109	85.49 154	86.19 146	93.88 134	91.86 114

v7new		80.09 125	78.45 138	82.00 110	83.97 157	90.49 108	87.18 112	79.67 121	71.49 150	67.44 120	61.24 155	62.41 148	79.83 109	85.49 154	86.19 146	93.88 134	91.86 114

v2v482		79.84 130	78.07 146	81.90 113	83.75 170	90.21 120	87.17 114	79.85 120	70.65 157	65.93 137	61.93 150	60.07 173	80.82 80	85.25 163	86.71 126	93.88 134	91.70 125

v18		79.71 137	77.98 153	81.73 118	84.02 156	86.67 168	87.37 101	76.35 156	72.61 145	68.86 110	61.35 152	62.65 141	79.94 105	85.49 154	86.21 141	93.85 137	90.92 136

IterMVS-LS		83.28 84	82.95 83	83.65 93	88.39 108	88.63 150	86.80 125	78.64 135	76.56 104	73.43 87	72.52 93	75.35 85	80.81 82	86.43 135	88.51 103	93.84 138	92.66 89
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
TSAR-MVS + ACMM		92.97 20	94.51 10	91.16 32	95.88 29	96.59 26	95.09 24	90.45 24	93.42 13	83.01 48	94.68 6	90.74 21	88.74 35	94.75 14	93.78 20	93.82 139	97.63 9

v1144		79.38 144	77.83 157	81.18 132	83.62 173	90.23 118	87.15 116	78.35 138	69.13 174	64.02 154	60.20 178	59.41 180	80.14 102	86.78 126	86.57 131	93.81 140	92.53 100

v1141		79.75 135	78.04 150	81.75 115	83.89 164	90.37 113	87.20 108	79.89 118	70.23 163	66.18 131	60.92 163	61.48 160	79.54 121	85.36 160	86.17 150	93.81 140	91.76 120

divwei89l23v2f112		79.75 135	78.04 150	81.75 115	83.90 161	90.37 113	87.21 107	79.90 117	70.20 165	66.18 131	60.92 163	61.48 160	79.52 124	85.36 160	86.17 150	93.81 140	91.77 118

v1		79.76 134	78.06 148	81.74 117	83.89 164	90.38 112	87.20 108	79.88 119	70.23 163	66.17 134	60.92 163	61.56 156	79.50 125	85.37 159	86.17 150	93.81 140	91.77 118

v1192		78.94 153	77.33 164	80.82 141	83.25 177	89.90 127	86.91 122	77.72 145	68.63 178	62.61 163	59.17 184	57.53 191	80.62 88	86.89 123	86.47 134	93.79 144	92.75 88

TSAR-MVS + COLMAP		88.40 50	89.09 45	87.60 61	92.72 61	93.92 63	92.21 45	85.57 50	91.73 26	73.72 85	91.75 19	73.22 98	87.64 47	91.49 54	89.71 78	93.73 145	91.82 117

EPNet		89.60 42	89.91 40	89.24 49	96.45 22	93.61 66	92.95 42	88.03 34	85.74 54	83.36 47	87.29 32	83.05 56	80.98 79	92.22 47	91.85 43	93.69 146	95.58 47
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
HyFIR lowres test		81.62 108	79.45 129	84.14 86	91.00 72	93.38 70	88.27 88	78.19 140	76.28 106	70.18 97	48.78 211	73.69 94	83.52 66	87.05 120	87.83 113	93.68 147	89.15 153

V42		79.59 140	78.43 140	80.94 140	82.79 187	89.71 132	86.66 127	76.73 154	71.38 153	67.42 122	61.01 161	62.30 150	78.39 134	85.56 149	86.48 133	93.65 148	92.60 91

v7n		77.22 169	76.23 178	78.38 166	81.89 194	89.10 146	82.24 174	76.36 155	65.96 191	61.21 180	56.56 194	55.79 200	75.07 153	86.55 131	86.68 127	93.52 149	92.95 80

ACMH+	79.08 13	81.84 101	80.06 118	83.91 90	89.92 100	90.62 106	86.21 134	83.48 71	73.88 132	65.75 138	66.38 124	65.30 123	84.63 63	85.90 140	87.25 119	93.45 150	91.13 133

pmmvs5		76.93 171	76.33 177	77.62 169	81.97 193	88.40 155	81.32 179	74.35 177	65.42 197	61.42 177	63.07 147	57.95 190	73.23 167	85.60 147	85.35 168	93.41 151	88.55 156

v52		76.55 176	75.89 184	77.31 173	79.94 204	88.49 153	81.07 183	73.62 180	65.49 195	61.66 174	56.29 197	58.90 186	74.30 160	83.47 182	85.62 164	93.28 152	92.99 77

V4		76.55 176	75.89 184	77.32 172	79.95 203	88.50 152	81.07 183	73.62 180	65.47 196	61.71 172	56.31 196	58.87 188	74.28 161	83.48 181	85.62 164	93.28 152	92.98 78

USDC		80.69 116	79.89 122	81.62 122	86.48 125	89.11 145	86.53 130	78.86 132	81.15 74	63.48 157	72.98 90	59.12 185	81.16 77	87.10 118	85.01 170	93.23 154	84.77 192

LS3D		85.96 67	84.37 77	87.81 60	94.13 44	93.27 71	90.26 62	89.00 28	84.91 57	72.84 90	71.74 94	72.47 100	87.45 49	89.53 94	89.09 94	93.20 155	89.60 150

MS-PatchMatch		81.79 102	81.44 95	82.19 108	90.35 85	89.29 140	88.08 92	75.36 173	77.60 99	69.00 109	64.37 144	78.87 76	77.14 143	88.03 111	85.70 162	93.19 156	86.24 181

Fast-Effi-MVS+-dtu		79.95 128	80.69 113	79.08 159	86.36 126	89.14 144	85.85 137	72.28 184	72.85 144	59.32 188	70.43 101	68.42 114	77.57 139	86.14 137	86.44 135	93.11 157	91.39 130

v748		76.17 185	75.10 194	77.43 171	81.60 195	88.01 157	79.02 195	76.28 160	64.47 199	64.14 152	56.55 195	56.26 198	70.40 181	82.50 189	85.77 160	93.11 157	92.15 108

GA-MVS		79.52 142	79.71 126	79.30 158	85.68 132	90.36 115	84.55 155	78.44 137	70.47 161	57.87 196	68.52 114	61.38 164	76.21 146	89.40 96	87.89 110	93.04 159	89.96 149

DeepPCF-MVS	88.51 2	92.64 25	94.42 13	90.56 36	94.84 38	96.92 14	91.31 56	89.61 26	95.16 1	84.55 41	89.91 25	91.45 16	90.15 29	95.12 8	94.81 5	92.90 160	97.58 10

CDS-MVSNet		81.63 107	82.09 88	81.09 136	87.21 120	90.28 117	87.46 100	80.33 110	69.06 175	70.66 93	71.30 95	73.87 92	67.99 189	89.58 93	89.87 73	92.87 161	90.69 143
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
Effi-MVS+-dtu		82.05 97	81.76 89	82.38 105	87.72 113	90.56 107	86.90 123	78.05 142	73.85 133	66.85 124	71.29 96	71.90 102	82.00 73	86.64 130	85.48 166	92.76 162	92.58 94

v148		78.59 158	76.84 174	80.62 144	83.61 174	89.16 143	83.65 163	79.24 130	69.38 173	69.34 107	59.88 180	60.41 171	75.19 150	83.81 178	84.63 175	92.70 163	90.63 145

pmmvs4		79.99 127	78.08 145	82.22 107	83.04 181	87.16 165	84.95 151	78.80 134	78.64 94	74.53 82	64.61 142	59.41 180	79.45 127	84.13 176	84.54 176	92.53 164	88.08 161

RPMNet		77.07 170	77.63 160	76.42 181	85.56 134	85.15 189	81.37 177	65.27 213	74.71 121	60.29 184	63.71 146	66.59 119	73.64 163	82.71 187	82.12 190	92.38 165	88.39 157

CR-MVSNet		78.71 156	78.86 132	78.55 164	85.85 131	85.15 189	82.30 172	68.23 199	74.71 121	65.37 142	64.39 143	69.59 109	77.18 141	85.10 168	84.87 171	92.34 166	88.21 159

diffmvs		85.70 69	86.35 67	84.95 71	87.75 111	90.96 102	89.09 74	78.56 136	86.50 51	80.44 60	77.86 62	83.93 52	81.64 74	85.52 152	86.79 125	92.21 167	92.87 82

COLMAP_ROB	76.78 15	80.50 121	78.49 135	82.85 101	90.96 73	89.65 134	86.20 135	83.40 73	77.15 101	66.54 125	62.27 149	65.62 122	77.89 138	85.23 164	84.70 174	92.11 168	84.83 191
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
PatchMatch-RL		83.34 83	81.36 96	85.65 67	90.33 86	89.52 136	84.36 157	81.82 97	80.87 79	79.29 64	74.04 79	62.85 140	86.05 58	88.40 108	87.04 122	92.04 169	86.77 175

PMMVS		81.65 105	84.05 78	78.86 161	78.56 207	82.63 201	83.10 165	67.22 204	81.39 70	70.11 98	84.91 38	79.74 71	82.12 71	87.31 115	85.70 162	92.03 170	86.67 178

MIMVSNet		74.69 196	75.60 189	73.62 198	76.02 215	85.31 188	81.21 182	67.43 202	71.02 155	59.07 193	54.48 199	64.07 130	66.14 198	86.52 133	86.64 128	91.83 171	81.17 204

FC-MVSNet-test		76.53 179	81.62 92	70.58 204	84.99 143	85.73 183	74.81 205	78.85 133	77.00 102	39.13 228	75.90 69	73.50 95	54.08 211	86.54 132	85.99 158	91.65 172	86.68 176

pmmvs-eth3d		74.32 198	71.96 203	77.08 176	77.33 211	82.71 200	78.41 197	76.02 167	66.65 186	65.98 136	54.23 202	49.02 217	73.14 168	82.37 191	82.69 187	91.61 173	86.05 183

SixPastTwentyTwo		76.02 187	75.72 187	76.36 182	83.38 175	87.54 160	75.50 204	76.22 161	65.50 194	57.05 197	70.64 98	53.97 207	74.54 156	80.96 195	82.12 190	91.44 174	89.35 152

pmmvs6		74.83 195	72.89 201	77.09 175	82.11 192	87.50 161	80.88 186	76.97 151	52.79 221	61.91 170	46.66 214	60.49 168	69.28 184	86.74 129	85.46 167	91.39 175	90.56 146

test-mter		77.79 164	80.02 120	75.18 190	81.18 199	82.85 199	80.52 188	62.03 221	73.62 137	62.16 166	73.55 82	73.83 93	73.81 162	84.67 171	83.34 182	91.37 176	88.31 158

TDRefinement		79.05 149	77.05 171	81.39 126	88.45 107	89.00 147	86.92 121	82.65 84	74.21 127	64.41 149	59.17 184	59.16 183	74.52 157	85.23 164	85.09 169	91.37 176	87.51 167

test-LLR		79.47 143	79.84 123	79.03 160	87.47 117	82.40 204	81.24 180	78.05 142	73.72 134	62.69 161	73.76 80	74.42 89	73.49 164	84.61 172	82.99 185	91.25 178	87.01 173

TESTMET0.1,1		77.78 165	79.84 123	75.38 189	80.86 200	82.40 204	81.24 180	62.72 220	73.72 134	62.69 161	73.76 80	74.42 89	73.49 164	84.61 172	82.99 185	91.25 178	87.01 173

TinyColmap		76.73 172	73.95 198	79.96 154	85.16 141	85.64 185	82.34 171	78.19 140	70.63 158	62.06 168	60.69 169	49.61 215	80.81 82	85.12 167	83.69 181	91.22 180	82.27 199

FMVSNet5		75.50 193	76.07 179	74.83 192	76.16 213	81.19 207	81.34 178	70.21 193	73.20 140	61.59 175	58.97 186	68.33 115	68.50 187	85.87 141	85.85 159	91.18 181	79.11 210

IterMVS		78.79 155	79.71 126	77.71 168	85.26 138	85.91 181	84.54 156	69.84 196	73.38 138	61.25 179	70.53 100	70.35 105	74.43 158	85.21 166	83.80 180	90.95 182	88.77 155
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
PatchT		76.42 180	77.81 158	74.80 193	78.46 208	84.30 194	71.82 211	65.03 215	73.89 131	65.37 142	61.58 151	66.70 118	77.18 141	85.10 168	84.87 171	90.94 183	88.21 159

EPNet_dtu		81.98 98	83.82 79	79.83 156	94.10 45	85.97 180	87.29 103	84.08 61	80.61 80	59.96 185	81.62 48	77.19 82	62.91 203	87.21 116	86.38 136	90.66 184	87.77 166
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
PM-MVS		74.17 199	73.10 199	75.41 188	76.07 214	82.53 202	77.56 201	71.69 186	71.04 154	61.92 169	61.23 158	47.30 218	74.82 155	81.78 193	79.80 195	90.42 185	88.05 162

test0.0.03 1		76.03 186	78.51 134	73.12 201	87.47 117	85.13 191	76.32 202	78.05 142	73.19 141	50.98 209	70.64 98	69.28 110	55.53 207	85.33 162	84.38 177	90.39 186	81.63 202

Anonymous20231206		70.80 203	70.59 206	71.04 203	81.60 195	82.49 203	74.64 206	75.87 170	64.17 200	49.27 210	44.85 218	53.59 209	54.68 210	83.07 184	82.34 189	90.17 187	83.65 195

CHOSEN 1792x2688		82.16 96	80.91 109	83.61 94	91.14 69	92.01 91	89.55 70	79.15 131	79.87 84	70.29 95	52.51 207	72.56 99	81.39 75	88.87 101	88.17 107	90.15 188	92.37 107

MIMVSNet1		65.00 211	66.24 211	63.55 216	58.41 232	80.01 210	69.00 214	74.03 178	55.81 219	41.88 220	36.81 227	49.48 216	47.89 219	81.32 194	82.40 188	90.08 189	77.88 212

LTVRE_ROB	74.41 16	75.78 191	74.72 196	77.02 177	85.88 129	89.22 141	82.44 170	77.17 149	50.57 223	45.45 216	65.44 131	52.29 211	81.25 76	85.50 153	87.42 117	89.94 190	92.62 90
Andreas Kuhn, Heiko HirschmÃ¼ller, Daniel Scharstein, Helmut Mayer: A TV Prior for High-Quality Scalable Multi-View Stereo Reconstruction. International Journal of Computer Vision 2016
DWT-MVSNet_training		80.51 120	78.05 149	83.39 97	88.64 105	88.33 156	86.11 136	76.33 157	79.65 86	78.64 68	69.62 106	58.89 187	80.82 80	80.50 197	82.03 192	89.77 191	87.36 169

GG-mvs-BLEND		57.56 220	82.61 85	28.34 233	0.22 238	90.10 122	79.37 192	0.14 237	79.56 87	0.40 241	71.25 97	83.40 54	0.30 238	86.27 136	83.87 178	89.59 192	83.83 194

CVMVSNet		76.70 173	78.46 137	74.64 195	83.34 176	84.48 193	81.83 176	74.58 174	68.88 176	51.23 208	69.77 102	70.05 106	67.49 192	84.27 175	83.81 179	89.38 193	87.96 163

TAMVS		76.42 180	77.16 170	75.56 187	83.05 180	85.55 186	80.58 187	71.43 187	65.40 198	61.04 182	67.27 120	69.22 111	67.99 189	84.88 170	84.78 173	89.28 194	83.01 198

CostFormer		80.94 114	80.21 117	81.79 114	87.69 114	88.58 151	87.47 99	70.66 190	80.02 82	77.88 73	73.03 89	71.40 103	78.24 135	79.96 200	79.63 196	88.82 195	88.84 154

RPSCF		83.46 82	83.36 81	83.59 95	87.75 111	87.35 162	84.82 154	79.46 127	83.84 60	78.12 69	82.69 41	79.87 68	82.60 70	82.47 190	81.13 194	88.78 196	86.13 182

test20.03		68.31 208	70.05 207	66.28 212	82.41 190	80.84 208	67.35 216	76.11 164	58.44 214	40.80 222	53.77 203	54.54 205	42.28 224	83.07 184	81.96 193	88.73 197	77.76 213

testgi		71.92 202	74.20 197	69.27 207	84.58 148	83.06 196	73.40 207	74.39 176	64.04 201	46.17 215	68.90 113	57.15 192	48.89 218	84.07 177	83.08 184	88.18 198	79.09 211

dps		78.02 162	75.94 183	80.44 148	86.06 128	86.62 172	82.58 167	69.98 194	75.14 114	77.76 75	69.08 111	59.93 175	78.47 133	79.47 202	77.96 203	87.78 199	83.40 196

PatchmatchNet		78.67 157	78.85 133	78.46 165	86.85 124	86.03 179	83.77 162	68.11 201	80.88 78	66.19 129	72.90 91	73.40 96	78.06 136	79.25 204	77.71 206	87.75 200	81.75 201
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
tpmp4_e23		79.82 132	77.96 154	82.00 110	87.59 116	86.93 166	87.81 93	72.21 185	79.99 83	78.02 71	67.83 118	64.77 124	78.74 132	79.99 199	78.90 199	87.65 201	87.29 170

MDTV_nov1_ep13_2view		73.21 201	72.91 200	73.56 199	80.01 201	84.28 195	78.62 196	66.43 208	68.64 177	59.12 192	60.39 174	59.69 178	69.81 183	78.82 206	77.43 208	87.36 202	81.11 205

MDTV_nov1_ep13		79.14 145	79.49 128	78.74 162	85.40 135	86.89 167	84.32 159	70.29 192	78.85 93	69.42 106	75.37 72	73.29 97	75.64 149	80.61 196	79.48 198	87.36 202	81.91 200

Anonymous20231211		62.95 217	60.42 220	65.89 213	74.22 218	78.37 215	67.66 215	74.47 175	40.37 231	39.59 226	27.51 230	38.26 229	52.13 212	75.39 215	77.89 205	87.28 204	85.16 188

EPMVS		77.53 167	78.07 146	76.90 178	86.89 123	84.91 192	82.18 175	66.64 207	81.00 76	64.11 153	72.75 92	69.68 108	74.42 159	79.36 203	78.13 202	87.14 205	80.68 206

EU-MVSNet		69.98 205	72.30 202	67.28 210	75.67 216	79.39 211	73.12 208	69.94 195	63.59 202	42.80 219	62.93 148	56.71 196	55.07 209	79.13 205	78.55 201	87.06 206	85.82 186

new-patchmatchnet		63.80 214	63.31 216	64.37 214	76.49 212	75.99 216	63.73 220	70.99 189	57.27 215	43.08 218	45.86 216	43.80 219	45.13 223	73.20 219	70.68 224	86.80 207	76.34 215

CMPMVS	56.49 17	73.84 200	71.73 204	76.31 184	85.20 139	85.67 184	75.80 203	73.23 182	62.26 204	65.40 141	53.40 205	59.70 177	71.77 173	80.25 198	79.56 197	86.45 208	81.28 203
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
MDA-MVSNet-bldmvs		66.22 210	64.49 213	68.24 208	61.67 228	82.11 206	70.07 213	76.16 162	59.14 213	47.94 213	54.35 201	35.82 230	67.33 193	64.94 228	75.68 212	86.30 209	79.36 208

MVS-HIRNet		68.83 206	66.39 210	71.68 202	77.58 209	75.52 217	66.45 217	65.05 214	62.16 205	62.84 160	44.76 219	56.60 197	71.96 172	78.04 207	75.06 215	86.18 210	72.56 219

LP		68.35 207	67.23 208	69.67 206	77.49 210	79.38 212	72.84 210	61.37 222	66.94 184	55.08 198	47.00 213	50.35 213	65.16 200	75.61 213	76.03 210	86.08 211	75.28 216

CHOSEN 280x420		80.28 122	81.66 91	78.67 163	82.92 184	79.24 213	85.36 147	66.79 206	78.11 96	70.32 94	75.03 73	79.87 68	81.09 78	89.07 98	83.16 183	85.54 212	87.17 172

tpm cat1		77.78 165	75.28 192	80.70 142	87.14 121	85.84 182	85.81 138	70.40 191	77.44 100	78.80 67	63.72 145	64.01 132	76.55 145	75.60 214	75.21 214	85.51 213	85.12 189

testus		63.31 216	64.48 214	61.94 219	73.99 219	71.99 220	63.56 222	63.25 218	57.01 217	39.41 227	54.38 200	38.73 228	46.24 222	77.01 209	77.93 204	85.20 214	74.29 218

test2356		63.96 212	64.10 215	63.78 215	74.71 217	71.55 221	65.83 218	67.38 203	57.11 216	40.41 223	53.58 204	41.13 224	49.35 217	77.00 210	77.57 207	85.01 215	70.79 220

tpm		76.30 184	76.05 181	76.59 180	86.97 122	83.01 198	83.83 161	67.06 205	71.83 148	63.87 155	69.56 107	62.88 139	73.41 166	79.79 201	78.59 200	84.41 216	86.68 176

tpmrst		76.55 176	75.99 182	77.20 174	87.32 119	83.05 197	82.86 166	65.62 211	78.61 95	67.22 123	69.19 109	65.71 121	75.87 148	76.75 211	75.33 213	84.31 217	83.28 197

pmmvs3		61.89 218	61.74 218	62.06 218	64.30 226	70.83 223	64.22 219	52.14 231	48.78 225	44.47 217	41.67 221	41.70 223	63.03 202	76.06 212	76.02 211	84.18 218	77.14 214

ADS-MVSNet		74.53 197	75.69 188	73.17 200	81.57 197	80.71 209	79.27 193	63.03 219	79.27 90	59.94 186	67.86 117	68.32 116	71.08 178	77.33 208	76.83 209	84.12 219	79.53 207

ambc			61.92 217		70.98 225	73.54 219	63.64 221		60.06 210	52.23 205	38.44 222	19.17 237	57.12 206	82.33 192	75.03 216	83.21 220	84.89 190

testmv		56.62 222	56.41 223	56.86 222	71.92 221	67.58 226	52.17 228	65.69 209	40.60 229	28.53 232	37.90 223	31.52 231	40.10 226	72.64 220	74.73 217	82.78 221	69.91 221

test1235678		56.61 223	56.40 224	56.86 222	71.92 221	67.58 226	52.17 228	65.69 209	40.58 230	28.52 233	37.89 224	31.49 232	40.10 226	72.64 220	74.72 218	82.78 221	69.90 222

FPMVS		63.63 215	60.08 221	67.78 209	80.01 201	71.50 222	72.88 209	69.41 198	61.82 206	53.11 201	45.12 217	42.11 222	50.86 215	66.69 225	63.84 227	80.41 223	69.46 224

N_pmnet		66.85 209	66.63 209	67.11 211	78.73 206	74.66 218	70.53 212	71.07 188	66.46 188	46.54 214	51.68 209	51.91 212	55.48 208	74.68 216	72.38 220	80.29 224	74.65 217

new_pmnet		59.28 219	61.47 219	56.73 224	61.66 229	68.29 225	59.57 224	54.91 227	60.83 209	34.38 230	44.66 220	43.65 220	49.90 216	71.66 222	71.56 223	79.94 225	69.67 223

PMVS	50.48 18	55.81 224	51.93 225	60.33 220	72.90 220	49.34 233	48.78 230	69.51 197	43.49 228	54.25 199	36.26 228	41.04 225	39.71 228	65.07 227	60.70 228	76.85 226	67.58 225
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
testpf		63.91 213	65.23 212	62.38 217	81.32 198	69.95 224	62.71 223	54.16 229	61.29 208	48.73 211	57.31 192	52.50 210	50.97 214	67.50 224	68.86 225	76.36 227	79.21 209

1111		57.32 221	57.20 222	57.46 221	71.89 223	67.50 228	52.34 226	58.78 224	46.57 226	39.69 224	37.38 225	38.78 226	46.37 220	74.15 217	74.36 219	75.70 228	61.66 227

test12356		50.02 225	51.22 226	48.61 226	63.00 227	60.15 231	47.60 232	56.49 226	38.02 232	24.74 235	36.14 229	25.93 234	24.79 231	66.19 226	71.68 222	75.07 229	60.44 229

no-one		44.14 227	43.91 229	44.40 228	59.91 230	61.10 230	34.07 235	60.09 223	27.71 234	14.44 237	19.11 232	19.28 236	23.90 233	47.36 232	66.69 226	73.98 230	66.11 226

Gipuma		49.17 226	47.05 227	51.65 225	59.67 231	48.39 234	41.98 233	63.47 217	55.64 220	33.33 231	14.90 233	13.78 238	41.34 225	69.31 223	72.30 221	70.11 231	55.00 231
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
PMMVS2		41.68 228	44.74 228	38.10 229	46.97 235	52.32 232	40.63 234	48.08 232	35.51 233	7.36 240	26.86 231	24.64 235	16.72 234	55.24 230	59.03 229	68.85 232	59.59 230

DeepMVS_CX						48.31 235	48.03 231	26.08 234	56.42 218	25.77 234	47.51 212	31.31 233	51.30 213	48.49 231		53.61 233	61.52 228

tmp_tt				32.73 232	43.96 236	21.15 239	26.71 236	8.99 235	65.67 193	51.39 207	56.01 198	42.64 221	11.76 235	56.60 229	50.81 231	53.55 234

E-PMN		31.40 230	26.80 232	36.78 230	51.39 234	29.96 237	20.20 237	54.17 228	25.93 236	12.75 238	14.73 234	8.58 240	34.10 230	27.36 234	37.83 232	48.07 235	43.18 233

EMVS		30.49 232	25.44 233	36.39 231	51.47 233	29.89 238	20.17 238	54.00 230	26.49 235	12.02 239	13.94 236	8.84 239	34.37 229	25.04 235	34.37 233	46.29 236	39.53 234

MVE	30.17 19	30.88 231	33.52 231	27.80 234	23.78 237	39.16 236	18.69 239	46.90 233	21.88 237	15.39 236	14.37 235	7.31 241	24.41 232	41.63 233	56.22 230	37.64 237	54.07 232
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
.test1245		41.43 229	38.48 230	44.88 227	71.89 223	67.50 228	52.34 226	58.78 224	46.57 226	39.69 224	37.38 225	38.78 226	46.37 220	74.15 217	1.18 234	0.20 238	3.76 236

testmvs		1.03 233	1.63 234	0.34 235	0.09 239	0.35 240	0.61 241	0.16 236	1.49 238	0.10 242	3.15 237	0.15 242	0.86 237	1.32 236	1.18 234	0.20 238	3.76 236

test123		0.87 234	1.40 235	0.25 236	0.03 240	0.25 241	0.35 242	0.08 238	1.21 239	0.05 243	2.84 238	0.03 243	0.89 236	0.43 237	1.16 236	0.13 240	3.87 235

sosnet-low-res		0.00 235	0.00 236	0.00 237	0.00 241	0.00 242	0.00 243	0.00 239	0.00 240	0.00 244	0.00 239	0.00 244	0.00 239	0.00 238	0.00 237	0.00 241	0.00 238

sosnet		0.00 235	0.00 236	0.00 237	0.00 241	0.00 242	0.00 243	0.00 239	0.00 240	0.00 244	0.00 239	0.00 244	0.00 239	0.00 238	0.00 237	0.00 241	0.00 238

MTAPA										92.97 2		91.03 17

MTMP										93.14 1		90.21 24

Patchmatch-RL test							8.55 240

mPP-MVS					97.06 13							88.08 39

NP-MVS									87.47 48

Patchmtry						85.54 187	82.30 172	68.23 199		65.37 142

High-resolution multi-view benchmark