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	botani.	boulde.	bridge	door	exhibi.	lectur.	living.	lounge	observ.	old co.	statue	terrac.

OPU-MVS				96.21 1	98.00 42	90.85 1	97.13 9				97.08 37	92.59 1	98.94 80	92.25 42	98.99 10	98.84 8

SED-MVS		95.91 1	96.28 1	94.80 33	98.77 4	85.99 54	97.13 9	97.44 13	90.31 26	97.71 1	98.07 4	92.31 2	99.58 5	95.66 2	99.13 3	98.84 8

test_241102_ONE					98.77 4	85.99 54		97.44 13	90.26 30	97.71 1	97.96 8	92.31 2	99.38 29

test_0728_THIRD									90.75 19	97.04 8	98.05 6	92.09 4	99.55 12	95.64 4	99.13 3	99.13 1

DPE-MVS		95.57 3	95.67 3	95.25 7	98.36 25	87.28 15	95.56 71	97.51 4	89.13 55	97.14 7	97.91 9	91.64 5	99.62 1	94.61 11	99.17 2	98.86 7

MSP-MVS		95.67 2	96.02 2	94.64 40	98.78 2	85.93 57	97.09 11	96.73 73	90.27 28	97.04 8	98.05 6	91.47 6	99.55 12	95.62 5	99.08 7	98.45 31

test0726					98.78 2	85.93 57	97.19 6	97.47 8	90.27 28	97.64 4	98.13 1	91.47 6

test_241102_TWO								97.44 13	90.31 26	97.62 5	98.07 4	91.46 8	99.58 5	95.66 2	99.12 6	98.98 6

DVP-MVS		95.42 5	95.56 5	94.98 19	98.49 16	86.52 37	96.91 20	97.47 8	91.73 8	96.10 13	96.69 54	89.90 9	99.30 39	94.70 9	98.04 63	99.13 1

DeepPCF-MVS	89.96 1	94.20 35	94.77 13	92.49 108	96.52 86	80.00 201	94.00 174	97.08 44	90.05 32	95.65 17	97.29 23	89.66 10	98.97 76	93.95 16	98.71 30	98.50 21

SD-MVS		94.96 11	95.33 7	93.88 62	97.25 69	86.69 30	96.19 39	97.11 43	90.42 25	96.95 10	97.27 24	89.53 11	96.91 228	94.38 13	98.85 15	98.03 66
Zhenlong Yuan, Jiakai Cao, Zhaoxin Li, Hao Jiang and Zhaoqi Wang: SD-MVS: Segmentation-driven Deformation Multi-View Stereo with Spherical Refinement and EM optimization. AAAI2024
CNVR-MVS		95.40 6	95.37 6	95.50 5	98.11 36	88.51 5	95.29 82	96.96 52	92.09 3	95.32 19	97.08 37	89.49 12	99.33 36	95.10 8	98.85 15	98.66 14

APDe-MVS		95.46 4	95.64 4	94.91 22	98.26 28	86.29 48	97.46 2	97.40 18	89.03 58	96.20 12	98.10 2	89.39 13	99.34 33	95.88 1	99.03 9	99.10 3

MCST-MVS		94.45 20	94.20 28	95.19 9	98.46 18	87.50 13	95.00 102	97.12 41	87.13 104	92.51 69	96.30 70	89.24 14	99.34 33	93.46 21	98.62 44	98.73 11

TSAR-MVS + MP.		94.85 12	94.94 10	94.58 43	98.25 29	86.33 44	96.11 45	96.62 84	88.14 84	96.10 13	96.96 43	89.09 15	98.94 80	94.48 12	98.68 35	98.48 23
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
SteuartSystems-ACMMP		95.20 7	95.32 8	94.85 27	96.99 72	86.33 44	97.33 3	97.30 27	91.38 11	95.39 18	97.46 17	88.98 16	99.40 28	94.12 15	98.89 14	98.82 10
Skip Steuart: Steuart Systems R&D Blog.
SMA-MVS		95.20 7	95.07 9	95.59 3	98.14 35	88.48 6	96.26 37	97.28 29	85.90 130	97.67 3	98.10 2	88.41 17	99.56 7	94.66 10	99.19 1	98.71 12

9.14			94.47 17		97.79 48		96.08 46	97.44 13	86.13 128	95.10 22	97.40 18	88.34 18	99.22 45	93.25 27	98.70 32

xxxxxxxxxxxxxcwj		94.65 15	94.70 14	94.48 47	97.85 44	85.63 67	95.21 88	95.47 164	89.44 44	95.71 15	97.70 11	88.28 19	99.35 31	93.89 18	98.78 21	98.48 23

SF-MVS		94.97 10	94.90 12	95.20 8	97.84 46	87.76 8	96.65 28	97.48 7	87.76 93	95.71 15	97.70 11	88.28 19	99.35 31	93.89 18	98.78 21	98.48 23

HPM-MVS++		95.14 9	94.91 11	95.83 2	98.25 29	89.65 2	95.92 55	96.96 52	91.75 7	94.02 33	96.83 47	88.12 21	99.55 12	93.41 24	98.94 12	98.28 44

agg_prior1		93.29 55	92.97 58	94.26 55	97.38 60	85.92 59	93.92 177	96.72 75	81.96 213	92.16 74	96.23 74	87.85 22	98.97 76	91.95 53	98.55 49	97.90 75

CSCG		93.23 58	93.05 55	93.76 68	98.04 40	84.07 95	96.22 38	97.37 19	84.15 167	90.05 108	95.66 96	87.77 23	99.15 52	89.91 84	98.27 57	98.07 62

NCCC		94.81 13	94.69 15	95.17 10	97.83 47	87.46 14	95.66 66	96.93 55	92.34 2	93.94 34	96.58 61	87.74 24	99.44 27	92.83 32	98.40 52	98.62 16

ETH3D-3000-0.1		94.61 16	94.44 18	95.12 11	97.70 51	87.71 9	95.98 52	97.44 13	86.67 117	95.25 21	97.31 22	87.73 25	99.24 43	93.11 30	98.76 26	98.40 34

TEST9					97.53 54	86.49 38	94.07 167	96.78 69	81.61 226	92.77 59	96.20 76	87.71 26	99.12 54

train_agg		93.44 51	93.08 54	94.52 45	97.53 54	86.49 38	94.07 167	96.78 69	81.86 219	92.77 59	96.20 76	87.63 27	99.12 54	92.14 47	98.69 33	97.94 71

test_8					97.49 57	86.30 47	94.02 172	96.76 72	81.86 219	92.70 63	96.20 76	87.63 27	99.02 67

TSAR-MVS + GP.		93.66 46	93.41 48	94.41 52	96.59 82	86.78 26	94.40 142	93.93 233	89.77 38	94.21 29	95.59 98	87.35 29	98.61 102	92.72 33	96.15 99	97.83 78

APD-MVS		94.24 30	94.07 34	94.75 36	98.06 39	86.90 21	95.88 56	96.94 54	85.68 136	95.05 23	97.18 32	87.31 30	99.07 57	91.90 57	98.61 45	98.28 44
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
ETH3 D test6400		93.64 47	93.22 51	94.92 20	97.79 48	86.84 22	95.31 77	97.26 30	82.67 200	93.81 37	96.29 71	87.29 31	99.27 42	89.87 85	98.67 37	98.65 15

ETH3D cwj APD-0.16		93.91 40	93.53 46	95.06 13	96.76 77	87.78 7	94.92 107	97.21 35	84.33 165	93.89 36	97.09 36	87.20 32	99.29 41	91.90 57	98.44 51	98.12 58

Regformer-2		94.33 27	94.22 25	94.68 38	95.54 119	86.75 29	94.57 130	96.70 77	91.84 6	94.41 25	96.56 63	87.19 33	99.13 53	93.50 20	97.65 73	98.16 54

segment_acmp												87.16 34

Regformer-1		94.22 32	94.13 32	94.51 46	95.54 119	86.36 43	94.57 130	96.44 92	91.69 9	94.32 28	96.56 63	87.05 35	99.03 63	93.35 25	97.65 73	98.15 55

testtj		94.39 25	94.18 29	95.00 16	98.24 31	86.77 28	96.16 40	97.23 33	87.28 102	94.85 24	97.04 39	86.99 36	99.52 20	91.54 63	98.33 55	98.71 12

旧先验1					96.79 76	81.81 152		95.67 148			96.81 49	86.69 37			97.66 72	96.97 111

test_prior3		93.60 48	93.53 46	93.82 64	97.29 65	84.49 82	94.12 160	96.88 59	87.67 96	92.63 64	96.39 68	86.62 38	98.87 83	91.50 64	98.67 37	98.11 60

test_prior2							94.12 160		87.67 96	92.63 64	96.39 68	86.62 38		91.50 64	98.67 37

CDPH-MVS		92.83 62	92.30 68	94.44 48	97.79 48	86.11 51	94.06 169	96.66 81	80.09 245	92.77 59	96.63 58	86.62 38	99.04 62	87.40 112	98.66 40	98.17 53

DPM-MVS		92.58 67	91.74 74	95.08 12	96.19 94	89.31 3	92.66 222	96.56 89	83.44 182	91.68 88	95.04 112	86.60 41	98.99 73	85.60 132	97.92 67	96.93 113

DELS-MVS		93.43 52	93.25 50	93.97 59	95.42 122	85.04 73	93.06 213	97.13 40	90.74 20	91.84 82	95.09 111	86.32 42	99.21 46	91.22 69	98.45 50	97.65 82
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
ZNCC-MVS		94.47 18	94.28 22	95.03 14	98.52 14	86.96 17	96.85 23	97.32 25	88.24 79	93.15 51	97.04 39	86.17 43	99.62 1	92.40 38	98.81 18	98.52 20

HFP-MVS		94.52 17	94.40 19	94.86 25	98.61 9	86.81 24	96.94 15	97.34 20	88.63 68	93.65 39	97.21 29	86.10 44	99.49 23	92.35 40	98.77 24	98.30 40

#test#		94.32 28	94.14 31	94.86 25	98.61 9	86.81 24	96.43 31	97.34 20	87.51 99	93.65 39	97.21 29	86.10 44	99.49 23	91.68 61	98.77 24	98.30 40

MVS_111021_HR		93.45 50	93.31 49	93.84 63	96.99 72	84.84 74	93.24 206	97.24 31	88.76 64	91.60 89	95.85 90	86.07 46	98.66 97	91.91 54	98.16 59	98.03 66

Regformer-4		93.91 40	93.81 39	94.19 57	95.36 123	85.47 69	94.68 122	96.41 95	91.60 10	93.75 38	96.71 52	85.95 47	99.10 56	93.21 28	96.65 90	98.01 68

ACMMP_NAP		94.74 14	94.56 16	95.28 6	98.02 41	87.70 10	95.68 64	97.34 20	88.28 78	95.30 20	97.67 13	85.90 48	99.54 16	93.91 17	98.95 11	98.60 17

Regformer-3		93.68 45	93.64 45	93.81 66	95.36 123	84.61 78	94.68 122	95.83 137	91.27 12	93.60 42	96.71 52	85.75 49	98.86 86	92.87 31	96.65 90	97.96 70

PHI-MVS		93.89 42	93.65 44	94.62 42	96.84 75	86.43 40	96.69 27	97.49 5	85.15 151	93.56 45	96.28 72	85.60 50	99.31 38	92.45 35	98.79 19	98.12 58

MP-MVS-pluss		94.21 33	94.00 36	94.85 27	98.17 34	86.65 33	94.82 114	97.17 39	86.26 124	92.83 57	97.87 10	85.57 51	99.56 7	94.37 14	98.92 13	98.34 37
MP-MVS-pluss: MP-MVS-pluss. MP-MVS-pluss
GST-MVS		94.21 33	93.97 37	94.90 24	98.41 22	86.82 23	96.54 30	97.19 36	88.24 79	93.26 47	96.83 47	85.48 52	99.59 4	91.43 67	98.40 52	98.30 40

MP-MVS		94.25 29	94.07 34	94.77 35	98.47 17	86.31 46	96.71 26	96.98 48	89.04 57	91.98 78	97.19 31	85.43 53	99.56 7	92.06 49	98.79 19	98.44 32
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
DeepC-MVS_fast	89.43 2	94.04 36	93.79 40	94.80 33	97.48 58	86.78 26	95.65 68	96.89 58	89.40 47	92.81 58	96.97 42	85.37 54	99.24 43	90.87 77	98.69 33	98.38 36
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
region2R		94.43 22	94.27 24	94.92 20	98.65 7	86.67 32	96.92 19	97.23 33	88.60 70	93.58 43	97.27 24	85.22 55	99.54 16	92.21 43	98.74 29	98.56 19

CP-MVS		94.34 26	94.21 27	94.74 37	98.39 23	86.64 34	97.60 1	97.24 31	88.53 72	92.73 62	97.23 27	85.20 56	99.32 37	92.15 46	98.83 17	98.25 49

test12				94.34 53	97.13 70	86.15 50		96.29 101		91.04 99		85.08 57	99.01 69		98.13 60	97.86 76

ACMMPR		94.43 22	94.28 22	94.91 22	98.63 8	86.69 30	96.94 15	97.32 25	88.63 68	93.53 46	97.26 26	85.04 58	99.54 16	92.35 40	98.78 21	98.50 21

XVS		94.45 20	94.32 20	94.85 27	98.54 12	86.60 35	96.93 17	97.19 36	90.66 23	92.85 55	97.16 34	85.02 59	99.49 23	91.99 50	98.56 47	98.47 27

X-MVStestdata		88.31 157	86.13 197	94.85 27	98.54 12	86.60 35	96.93 17	97.19 36	90.66 23	92.85 55	23.41 347	85.02 59	99.49 23	91.99 50	98.56 47	98.47 27

MSLP-MVS++		93.72 44	94.08 33	92.65 101	97.31 63	83.43 111	95.79 59	97.33 23	90.03 33	93.58 43	96.96 43	84.87 61	97.76 159	92.19 45	98.66 40	96.76 117

HPM-MVS		94.02 37	93.88 38	94.43 50	98.39 23	85.78 64	97.25 5	97.07 45	86.90 112	92.62 66	96.80 51	84.85 62	99.17 49	92.43 36	98.65 42	98.33 38
Chunlin Ren, Qingshan Xu, Shikun Zhang, Jiaqi Yang: Hierarchical Prior Mining for Non-local Multi-View Stereo. ICCV 2023
SR-MVS		94.23 31	94.17 30	94.43 50	98.21 33	85.78 64	96.40 33	96.90 57	88.20 82	94.33 27	97.40 18	84.75 63	99.03 63	93.35 25	97.99 64	98.48 23

PGM-MVS		93.96 39	93.72 43	94.68 38	98.43 19	86.22 49	95.30 80	97.78 1	87.45 100	93.26 47	97.33 21	84.62 64	99.51 21	90.75 79	98.57 46	98.32 39

EI-MVSNet-Vis-set		93.01 61	92.92 59	93.29 72	95.01 135	83.51 110	94.48 134	95.77 141	90.87 15	92.52 68	96.67 56	84.50 65	99.00 72	91.99 50	94.44 127	97.36 93

zzz-MVS		94.47 18	94.30 21	95.00 16	98.42 20	86.95 18	95.06 100	96.97 49	91.07 13	93.14 52	97.56 14	84.30 66	99.56 7	93.43 22	98.75 27	98.47 27

MTAPA		94.42 24	94.22 25	95.00 16	98.42 20	86.95 18	94.36 150	96.97 49	91.07 13	93.14 52	97.56 14	84.30 66	99.56 7	93.43 22	98.75 27	98.47 27

CS-MVS		92.60 66	92.56 64	92.73 96	95.55 118	82.35 143	96.14 42	96.85 62	88.71 65	91.44 92	91.51 236	84.13 68	98.48 108	91.27 68	97.47 76	97.34 94

ETV-MVS		92.74 64	92.66 62	92.97 86	95.20 131	84.04 97	95.07 97	96.51 90	90.73 21	92.96 54	91.19 243	84.06 69	98.34 121	91.72 60	96.54 93	96.54 125

EI-MVSNet-UG-set		92.74 64	92.62 63	93.12 78	94.86 146	83.20 116	94.40 142	95.74 144	90.71 22	92.05 77	96.60 60	84.00 70	98.99 73	91.55 62	93.63 135	97.17 102

mPP-MVS		93.99 38	93.78 41	94.63 41	98.50 15	85.90 62	96.87 21	96.91 56	88.70 66	91.83 84	97.17 33	83.96 71	99.55 12	91.44 66	98.64 43	98.43 33

APD-MVS_3200maxsize		93.78 43	93.77 42	93.80 67	97.92 43	84.19 93	96.30 35	96.87 61	86.96 108	93.92 35	97.47 16	83.88 72	98.96 79	92.71 34	97.87 68	98.26 48

EIA-MVS		91.95 74	91.94 71	91.98 128	95.16 132	80.01 200	95.36 74	96.73 73	88.44 73	89.34 115	92.16 209	83.82 73	98.45 114	89.35 89	97.06 82	97.48 90

EPP-MVSNet		91.70 80	91.56 76	92.13 124	95.88 107	80.50 187	97.33 3	95.25 180	86.15 126	89.76 110	95.60 97	83.42 74	98.32 124	87.37 114	93.25 145	97.56 88

UA-Net		92.83 62	92.54 65	93.68 69	96.10 99	84.71 77	95.66 66	96.39 97	91.92 4	93.22 49	96.49 65	83.16 75	98.87 83	84.47 145	95.47 107	97.45 92

UniMVSNet_NR-MVSNet		89.92 115	89.29 115	91.81 140	93.39 200	83.72 103	94.43 140	97.12 41	89.80 37	86.46 161	93.32 170	83.16 75	97.23 206	84.92 138	81.02 287	94.49 201

1121		90.42 104	89.49 109	93.20 75	97.27 67	84.46 85	92.63 223	95.51 162	71.01 324	91.20 97	96.21 75	82.92 77	99.05 59	80.56 207	98.07 62	96.10 139

新几何1				93.10 79	97.30 64	84.35 91		95.56 157	71.09 323	91.26 96	96.24 73	82.87 78	98.86 86	79.19 226	98.10 61	96.07 141

原ACMM1				92.01 125	97.34 62	81.05 171		96.81 67	78.89 257	90.45 103	95.92 87	82.65 79	98.84 91	80.68 205	98.26 58	96.14 134

casdiffmvs		92.51 68	92.43 67	92.74 95	94.41 165	81.98 149	94.54 132	96.23 107	89.57 42	91.96 79	96.17 80	82.58 80	98.01 146	90.95 75	95.45 109	98.23 50

DeepC-MVS	88.79 3	93.31 54	92.99 57	94.26 55	96.07 101	85.83 63	94.89 109	96.99 47	89.02 59	89.56 111	97.37 20	82.51 81	99.38 29	92.20 44	98.30 56	97.57 87
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
HPM-MVS_fast		93.40 53	93.22 51	93.94 61	98.36 25	84.83 75	97.15 8	96.80 68	85.77 133	92.47 70	97.13 35	82.38 82	99.07 57	90.51 81	98.40 52	97.92 74

baseline		92.39 71	92.29 69	92.69 100	94.46 162	81.77 153	94.14 159	96.27 102	89.22 51	91.88 80	96.00 84	82.35 83	97.99 148	91.05 71	95.27 114	98.30 40

canonicalmvs		93.27 56	92.75 61	94.85 27	95.70 114	87.66 11	96.33 34	96.41 95	90.00 34	94.09 31	94.60 129	82.33 84	98.62 101	92.40 38	92.86 153	98.27 46

DP-MVS Recon		91.95 74	91.28 79	93.96 60	98.33 27	85.92 59	94.66 125	96.66 81	82.69 199	90.03 109	95.82 91	82.30 85	99.03 63	84.57 144	96.48 96	96.91 114

PAPR		90.02 110	89.27 117	92.29 120	95.78 110	80.95 174	92.68 221	96.22 108	81.91 216	86.66 159	93.75 165	82.23 86	98.44 115	79.40 225	94.79 117	97.48 90

MVS_Test		91.31 86	91.11 82	91.93 132	94.37 166	80.14 192	93.46 194	95.80 139	86.46 120	91.35 95	93.77 163	82.21 87	98.09 140	87.57 110	94.95 116	97.55 89

nrg030		91.08 91	90.39 92	93.17 77	93.07 209	86.91 20	96.41 32	96.26 103	88.30 77	88.37 128	94.85 120	82.19 88	97.64 167	91.09 70	82.95 259	94.96 176

UniMVSNet (Re)		89.80 117	89.07 120	92.01 125	93.60 195	84.52 81	94.78 117	97.47 8	89.26 50	86.44 164	92.32 204	82.10 89	97.39 193	84.81 141	80.84 291	94.12 213

testdata				90.49 184	96.40 88	77.89 247		95.37 176	72.51 316	93.63 41	96.69 54	82.08 90	97.65 165	83.08 160	97.39 77	95.94 145

PAPM_NR		91.22 88	90.78 90	92.52 107	97.60 53	81.46 161	94.37 149	96.24 106	86.39 122	87.41 144	94.80 122	82.06 91	98.48 108	82.80 168	95.37 110	97.61 84

MG-MVS		91.77 77	91.70 75	92.00 127	97.08 71	80.03 199	93.60 189	95.18 184	87.85 90	90.89 100	96.47 66	82.06 91	98.36 118	85.07 136	97.04 83	97.62 83

CANet		93.54 49	93.20 53	94.55 44	95.65 115	85.73 66	94.94 105	96.69 79	91.89 5	90.69 101	95.88 89	81.99 93	99.54 16	93.14 29	97.95 66	98.39 35

FC-MVSNet-test		90.27 106	90.18 97	90.53 180	93.71 191	79.85 205	95.77 60	97.59 2	89.31 49	86.27 167	94.67 126	81.93 94	97.01 222	84.26 147	88.09 215	94.71 187

FIs		90.51 103	90.35 93	90.99 169	93.99 181	80.98 172	95.73 61	97.54 3	89.15 54	86.72 158	94.68 125	81.83 95	97.24 204	85.18 135	88.31 211	94.76 186

ACMMP		93.24 57	92.88 60	94.30 54	98.09 38	85.33 71	96.86 22	97.45 11	88.33 76	90.15 107	97.03 41	81.44 96	99.51 21	90.85 78	95.74 102	98.04 65
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
Effi-MVS+		91.59 82	91.11 82	93.01 84	94.35 169	83.39 113	94.60 127	95.10 188	87.10 105	90.57 102	93.10 181	81.43 97	98.07 142	89.29 90	94.48 125	97.59 86

MVS_111021_LR		92.47 69	92.29 69	92.98 85	95.99 104	84.43 89	93.08 211	96.09 116	88.20 82	91.12 98	95.72 95	81.33 98	97.76 159	91.74 59	97.37 78	96.75 118

mvs_anonymous		89.37 132	89.32 114	89.51 227	93.47 198	74.22 287	91.65 253	94.83 205	82.91 195	85.45 187	93.79 161	81.23 99	96.36 260	86.47 127	94.09 129	97.94 71

PVSNet_BlendedMVS		89.98 111	89.70 106	90.82 173	96.12 96	81.25 166	93.92 177	96.83 64	83.49 181	89.10 118	92.26 207	81.04 100	98.85 89	86.72 125	87.86 219	92.35 288

PVSNet_Blended		90.73 95	90.32 94	91.98 128	96.12 96	81.25 166	92.55 227	96.83 64	82.04 211	89.10 118	92.56 197	81.04 100	98.85 89	86.72 125	95.91 100	95.84 150

alignmvs		93.08 60	92.50 66	94.81 32	95.62 117	87.61 12	95.99 50	96.07 118	89.77 38	94.12 30	94.87 117	80.56 102	98.66 97	92.42 37	93.10 148	98.15 55

abl_6		93.18 59	93.05 55	93.57 71	97.52 56	84.27 92	95.53 72	96.67 80	87.85 90	93.20 50	97.22 28	80.35 103	99.18 48	91.91 54	97.21 79	97.26 97

API-MVS		90.66 98	90.07 99	92.45 110	96.36 90	84.57 80	96.06 48	95.22 183	82.39 202	89.13 117	94.27 141	80.32 104	98.46 111	80.16 214	96.71 88	94.33 206

PVSNet_Blended_VisFu		91.38 84	90.91 87	92.80 92	96.39 89	83.17 117	94.87 111	96.66 81	83.29 186	89.27 116	94.46 133	80.29 105	99.17 49	87.57 110	95.37 110	96.05 143

test222					96.55 84	81.70 154	92.22 237	95.01 191	68.36 329	90.20 106	96.14 81	80.26 106			97.80 70	96.05 143

diffmvs		91.37 85	91.23 80	91.77 141	93.09 208	80.27 189	92.36 232	95.52 161	87.03 107	91.40 94	94.93 114	80.08 107	97.44 181	92.13 48	94.56 123	97.61 84

Test By Simon												80.02 108

IterMVS-LS		88.36 156	87.91 149	89.70 220	93.80 188	78.29 238	93.73 183	95.08 190	85.73 134	84.75 205	91.90 223	79.88 109	96.92 227	83.83 152	82.51 265	93.89 225
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
EI-MVSNet		89.10 136	88.86 127	89.80 216	91.84 241	78.30 237	93.70 186	95.01 191	85.73 134	87.15 148	95.28 103	79.87 110	97.21 208	83.81 153	87.36 224	93.88 227

TAPA-MVS	84.62 6	88.16 161	87.01 168	91.62 145	96.64 80	80.65 181	94.39 144	96.21 111	76.38 281	86.19 169	95.44 99	79.75 111	98.08 141	62.75 324	95.29 112	96.13 135
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
Fast-Effi-MVS+		89.41 129	88.64 129	91.71 143	94.74 148	80.81 178	93.54 190	95.10 188	83.11 189	86.82 157	90.67 260	79.74 112	97.75 162	80.51 209	93.55 136	96.57 123

pcd_1.5k_mvsjas		6.64 325	8.86 327	0.00 337	0.00 356	0.00 357	0.00 348	0.00 358	0.00 352	0.00 353	0.00 353	79.70 113	0.00 354	0.00 351	0.00 351	0.00 350

PS-MVSNAJss		89.97 112	89.62 107	91.02 166	91.90 239	80.85 177	95.26 85	95.98 123	86.26 124	86.21 168	94.29 138	79.70 113	97.65 165	88.87 95	88.10 213	94.57 194

PS-MVSNAJ		91.18 89	90.92 86	91.96 130	95.26 129	82.60 138	92.09 242	95.70 146	86.27 123	91.84 82	92.46 199	79.70 113	98.99 73	89.08 92	95.86 101	94.29 207

xiu_mvs_v2_base		91.13 90	90.89 88	91.86 135	94.97 138	82.42 139	92.24 236	95.64 153	86.11 129	91.74 87	93.14 179	79.67 116	98.89 82	89.06 93	95.46 108	94.28 208

WR-MVS_H		87.80 170	87.37 159	89.10 235	93.23 204	78.12 241	95.61 69	97.30 27	87.90 88	83.72 234	92.01 220	79.65 117	96.01 273	76.36 251	80.54 295	93.16 262

EPNet		91.79 76	91.02 85	94.10 58	90.10 300	85.25 72	96.03 49	92.05 267	92.83 1	87.39 147	95.78 92	79.39 118	99.01 69	88.13 104	97.48 75	98.05 64
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
miper_ehance_all_eth		87.22 197	86.62 182	89.02 238	92.13 231	77.40 261	90.91 264	94.81 207	81.28 232	84.32 220	90.08 271	79.26 119	96.62 238	83.81 153	82.94 260	93.04 267

miper_enhance_ethall		86.90 205	86.18 196	89.06 236	91.66 249	77.58 258	90.22 275	94.82 206	79.16 254	84.48 211	89.10 285	79.19 120	96.66 236	84.06 149	82.94 260	92.94 270

NR-MVSNet		88.58 152	87.47 157	91.93 132	93.04 211	84.16 94	94.77 118	96.25 105	89.05 56	80.04 286	93.29 173	79.02 121	97.05 220	81.71 190	80.05 301	94.59 192

TAMVS		89.21 134	88.29 140	91.96 130	93.71 191	82.62 137	93.30 200	94.19 225	82.22 206	87.78 139	93.94 152	78.83 122	96.95 225	77.70 239	92.98 151	96.32 128

cl_fuxian		87.14 201	86.50 186	89.04 237	92.20 228	77.26 262	91.22 261	94.70 211	82.01 212	84.34 219	90.43 264	78.81 123	96.61 241	83.70 155	81.09 284	93.25 257

1112_ss		88.42 153	87.33 160	91.72 142	94.92 142	80.98 172	92.97 216	94.54 213	78.16 270	83.82 232	93.88 157	78.78 124	97.91 154	79.45 221	89.41 191	96.26 131

CDS-MVSNet		89.45 126	88.51 131	92.29 120	93.62 194	83.61 108	93.01 214	94.68 212	81.95 214	87.82 138	93.24 175	78.69 125	96.99 223	80.34 211	93.23 146	96.28 130
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
WTY-MVS		89.60 120	88.92 124	91.67 144	95.47 121	81.15 170	92.38 231	94.78 209	83.11 189	89.06 120	94.32 136	78.67 126	96.61 241	81.57 191	90.89 173	97.24 98

CPTT-MVS		91.99 73	91.80 73	92.55 105	98.24 31	81.98 149	96.76 25	96.49 91	81.89 218	90.24 105	96.44 67	78.59 127	98.61 102	89.68 86	97.85 69	97.06 106

IS-MVSNet		91.43 83	91.09 84	92.46 109	95.87 109	81.38 164	96.95 14	93.69 239	89.72 40	89.50 113	95.98 85	78.57 128	97.77 158	83.02 162	96.50 95	98.22 51

OMC-MVS		91.23 87	90.62 91	93.08 80	96.27 92	84.07 95	93.52 191	95.93 127	86.95 109	89.51 112	96.13 82	78.50 129	98.35 120	85.84 130	92.90 152	96.83 116

PCF-MVS	84.11 10	87.74 171	86.08 201	92.70 99	94.02 176	84.43 89	89.27 289	95.87 134	73.62 307	84.43 214	94.33 135	78.48 130	98.86 86	70.27 287	94.45 126	94.81 184
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
LCM-MVSNet-Re		88.30 158	88.32 139	88.27 256	94.71 151	72.41 306	93.15 207	90.98 296	87.77 92	79.25 293	91.96 221	78.35 131	95.75 285	83.04 161	95.62 103	96.65 121

HY-MVS	83.01 12	89.03 139	87.94 148	92.29 120	94.86 146	82.77 128	92.08 243	94.49 214	81.52 228	86.93 153	92.79 193	78.32 132	98.23 127	79.93 216	90.55 174	95.88 148

MVS		87.44 187	86.10 200	91.44 150	92.61 222	83.62 107	92.63 223	95.66 150	67.26 330	81.47 264	92.15 210	77.95 133	98.22 129	79.71 218	95.48 106	92.47 283

MVSFormer		91.68 81	91.30 78	92.80 92	93.86 185	83.88 100	95.96 53	95.90 131	84.66 161	91.76 85	94.91 115	77.92 134	97.30 196	89.64 87	97.11 80	97.24 98

lupinMVS		90.92 92	90.21 95	93.03 83	93.86 185	83.88 100	92.81 219	93.86 234	79.84 248	91.76 85	94.29 138	77.92 134	98.04 144	90.48 82	97.11 80	97.17 102

Test_1112_low_res		87.65 174	86.51 185	91.08 162	94.94 141	79.28 219	91.77 247	94.30 221	76.04 286	83.51 241	92.37 202	77.86 136	97.73 163	78.69 230	89.13 198	96.22 132

VNet		92.24 72	91.91 72	93.24 74	96.59 82	83.43 111	94.84 113	96.44 92	89.19 53	94.08 32	95.90 88	77.85 137	98.17 131	88.90 94	93.38 142	98.13 57

DU-MVS		89.34 133	88.50 132	91.85 137	93.04 211	83.72 103	94.47 137	96.59 86	89.50 43	86.46 161	93.29 173	77.25 138	97.23 206	84.92 138	81.02 287	94.59 192

Baseline_NR-MVSNet		87.07 202	86.63 181	88.40 252	91.44 252	77.87 248	94.23 156	92.57 257	84.12 168	85.74 175	92.08 216	77.25 138	96.04 270	82.29 176	79.94 302	91.30 305

jason		90.80 93	90.10 98	92.90 89	93.04 211	83.53 109	93.08 211	94.15 227	80.22 242	91.41 93	94.91 115	76.87 140	97.93 153	90.28 83	96.90 84	97.24 98
jason: jason.
PAPM		86.68 213	85.39 220	90.53 180	93.05 210	79.33 218	89.79 282	94.77 210	78.82 259	81.95 261	93.24 175	76.81 141	97.30 196	66.94 307	93.16 147	94.95 179

Vis-MVSNet (Re-imp)		89.59 121	89.44 111	90.03 205	95.74 111	75.85 278	95.61 69	90.80 302	87.66 98	87.83 137	95.40 102	76.79 142	96.46 254	78.37 231	96.73 87	97.80 79

baseline1		88.10 162	87.28 162	90.57 177	94.96 139	80.07 195	94.27 153	91.29 289	86.74 114	87.41 144	94.00 149	76.77 143	96.20 265	80.77 202	79.31 309	95.44 161

114514_t		89.51 123	88.50 132	92.54 106	98.11 36	81.99 148	95.16 93	96.36 99	70.19 326	85.81 173	95.25 105	76.70 144	98.63 100	82.07 179	96.86 86	97.00 110

PLC	84.53 7	89.06 138	88.03 145	92.15 123	97.27 67	82.69 135	94.29 152	95.44 170	79.71 250	84.01 228	94.18 143	76.68 145	98.75 95	77.28 243	93.41 141	95.02 172
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
TranMVSNet+NR-MVSNet		88.84 144	87.95 147	91.49 148	92.68 221	83.01 123	94.92 107	96.31 100	89.88 36	85.53 181	93.85 159	76.63 146	96.96 224	81.91 183	79.87 304	94.50 199

MAR-MVS		90.30 105	89.37 113	93.07 82	96.61 81	84.48 84	95.68 64	95.67 148	82.36 204	87.85 136	92.85 187	76.63 146	98.80 93	80.01 215	96.68 89	95.91 146
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
WR-MVS		88.38 154	87.67 153	90.52 182	93.30 203	80.18 190	93.26 203	95.96 125	88.57 71	85.47 186	92.81 191	76.12 148	96.91 228	81.24 194	82.29 267	94.47 204

v8		87.50 186	86.71 176	89.89 210	91.37 258	79.40 212	94.50 133	95.38 174	84.81 158	83.60 239	91.33 238	76.05 149	97.42 183	82.84 166	80.51 298	92.84 274

v148		87.04 203	86.32 192	89.21 231	90.94 276	77.26 262	93.71 185	94.43 216	84.84 157	84.36 218	90.80 257	76.04 150	97.05 220	82.12 178	79.60 306	93.31 254

eth_miper_zixun_eth		86.50 219	85.77 213	88.68 246	91.94 238	75.81 279	90.47 270	94.89 200	82.05 209	84.05 226	90.46 263	75.96 151	96.77 232	82.76 169	79.36 308	93.46 251

3Dnovator+	87.14 4	92.42 70	91.37 77	95.55 4	95.63 116	88.73 4	97.07 13	96.77 71	90.84 16	84.02 227	96.62 59	75.95 152	99.34 33	87.77 107	97.68 71	98.59 18

BH-untuned		88.60 151	88.13 144	90.01 207	95.24 130	78.50 232	93.29 201	94.15 227	84.75 159	84.46 212	93.40 167	75.76 153	97.40 190	77.59 240	94.52 124	94.12 213

cl-mvsnet1		86.53 217	85.78 211	88.75 243	92.02 236	76.45 272	90.74 266	94.30 221	81.83 221	83.34 245	90.82 256	75.75 154	96.57 244	81.73 189	81.52 280	93.24 258

BH-w/o		87.57 182	87.05 167	89.12 234	94.90 144	77.90 246	92.41 229	93.51 241	82.89 196	83.70 235	91.34 237	75.75 154	97.07 218	75.49 259	93.49 138	92.39 286

cl-mvsnet_		86.52 218	85.78 211	88.75 243	92.03 235	76.46 271	90.74 266	94.30 221	81.83 221	83.34 245	90.78 258	75.74 156	96.57 244	81.74 188	81.54 279	93.22 259

cdsmvs_eth3d_5k		22.14 320	29.52 322	0.00 337	0.00 356	0.00 357	0.00 348	95.76 142	0.00 352	0.00 353	94.29 138	75.66 157	0.00 354	0.00 351	0.00 351	0.00 350

CNLPA		89.07 137	87.98 146	92.34 116	96.87 74	84.78 76	94.08 166	93.24 244	81.41 229	84.46 212	95.13 110	75.57 158	96.62 238	77.21 244	93.84 133	95.61 157

CHOSEN 1792x2688		88.84 144	87.69 151	92.30 119	96.14 95	81.42 163	90.01 279	95.86 135	74.52 300	87.41 144	93.94 152	75.46 159	98.36 118	80.36 210	95.53 104	97.12 105

CP-MVSNet		87.63 177	87.26 164	88.74 245	93.12 207	76.59 270	95.29 82	96.58 87	88.43 74	83.49 242	92.98 184	75.28 160	95.83 281	78.97 227	81.15 283	93.79 233

v10		87.25 194	86.38 187	89.85 211	91.19 264	79.50 209	94.48 134	95.45 168	83.79 174	83.62 238	91.19 243	75.13 161	97.42 183	81.94 182	80.60 293	92.63 279

Vis-MVSNet		91.75 78	91.23 80	93.29 72	95.32 126	83.78 102	96.14 42	95.98 123	89.89 35	90.45 103	96.58 61	75.09 162	98.31 125	84.75 142	96.90 84	97.78 81
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
sss		88.93 142	88.26 142	90.94 172	94.05 175	80.78 179	91.71 250	95.38 174	81.55 227	88.63 123	93.91 156	75.04 163	95.47 297	82.47 172	91.61 164	96.57 123

v1144		87.61 180	86.79 174	90.06 204	91.01 271	79.34 215	93.95 176	95.42 173	83.36 185	85.66 177	91.31 241	74.98 164	97.42 183	83.37 157	82.06 270	93.42 252

miper_lstm_enhance		85.27 241	84.59 238	87.31 277	91.28 262	74.63 284	87.69 309	94.09 231	81.20 235	81.36 267	89.85 277	74.97 165	94.30 310	81.03 198	79.84 305	93.01 268

test_yl		90.69 96	90.02 103	92.71 97	95.72 112	82.41 141	94.11 162	95.12 186	85.63 137	91.49 90	94.70 123	74.75 166	98.42 116	86.13 128	92.53 157	97.31 95

DCV-MVSNet		90.69 96	90.02 103	92.71 97	95.72 112	82.41 141	94.11 162	95.12 186	85.63 137	91.49 90	94.70 123	74.75 166	98.42 116	86.13 128	92.53 157	97.31 95

V42		87.68 172	86.86 170	90.15 198	90.58 291	80.14 192	94.24 155	95.28 179	83.66 176	85.67 176	91.33 238	74.73 168	97.41 188	84.43 146	81.83 274	92.89 272

XVG-OURS-SEG-HR		89.95 113	89.45 110	91.47 149	94.00 180	81.21 169	91.87 245	96.06 120	85.78 132	88.55 124	95.73 94	74.67 169	97.27 200	88.71 97	89.64 189	95.91 146

v2v482		87.84 168	87.06 166	90.17 196	90.99 272	79.23 222	94.00 174	95.13 185	84.87 156	85.53 181	92.07 218	74.45 170	97.45 179	84.71 143	81.75 276	93.85 231

CLD-MVS		89.47 125	88.90 125	91.18 157	94.22 170	82.07 147	92.13 240	96.09 116	87.90 88	85.37 196	92.45 200	74.38 171	97.56 171	87.15 117	90.43 175	93.93 223
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
XXY-MVS		87.65 174	86.85 171	90.03 205	92.14 230	80.60 184	93.76 182	95.23 181	82.94 193	84.60 207	94.02 147	74.27 172	95.49 296	81.04 196	83.68 252	94.01 221

HQP_MVS		90.60 102	90.19 96	91.82 138	94.70 152	82.73 132	95.85 57	96.22 108	90.81 17	86.91 154	94.86 118	74.23 173	98.12 132	88.15 102	89.99 180	94.63 188

plane_prior6					94.52 158	82.75 129						74.23 173

v144192		87.19 199	86.35 190	89.74 217	90.64 289	78.24 239	93.92 177	95.43 171	81.93 215	85.51 183	91.05 251	74.21 175	97.45 179	82.86 165	81.56 278	93.53 246

VPA-MVSNet		89.62 119	88.96 122	91.60 146	93.86 185	82.89 127	95.46 73	97.33 23	87.91 87	88.43 127	93.31 171	74.17 176	97.40 190	87.32 115	82.86 264	94.52 197

ab-mvs		89.41 129	88.35 136	92.60 102	95.15 133	82.65 136	92.20 238	95.60 155	83.97 171	88.55 124	93.70 166	74.16 177	98.21 130	82.46 173	89.37 192	96.94 112

1314		87.51 184	86.57 184	90.34 193	92.42 225	79.74 207	92.63 223	95.35 178	78.35 266	80.14 283	91.62 232	74.05 178	97.15 210	81.05 195	93.53 137	94.12 213

test_djsdf		89.03 139	88.64 129	90.21 195	90.74 286	79.28 219	95.96 53	95.90 131	84.66 161	85.33 198	92.94 185	74.02 179	97.30 196	89.64 87	88.53 204	94.05 219

cl-mvsnet2		86.78 209	85.98 204	89.18 233	92.34 226	77.62 257	90.84 265	94.13 229	81.33 231	83.97 229	90.15 269	73.96 180	96.60 243	84.19 148	82.94 260	93.33 253

AdaColmap		89.89 116	89.07 120	92.37 115	97.41 59	83.03 121	94.42 141	95.92 128	82.81 197	86.34 166	94.65 127	73.89 181	99.02 67	80.69 204	95.51 105	95.05 171

HyFIR lowres test		88.09 163	86.81 172	91.93 132	96.00 103	80.63 182	90.01 279	95.79 140	73.42 308	87.68 141	92.10 215	73.86 182	97.96 150	80.75 203	91.70 163	97.19 101

HQP2-MVS												73.83 183

HQP-MVS		89.80 117	89.28 116	91.34 153	94.17 171	81.56 155	94.39 144	96.04 121	88.81 61	85.43 190	93.97 151	73.83 183	97.96 150	87.11 119	89.77 187	94.50 199

3Dnovator	86.66 5	91.73 79	90.82 89	94.44 48	94.59 156	86.37 42	97.18 7	97.02 46	89.20 52	84.31 222	96.66 57	73.74 185	99.17 49	86.74 122	97.96 65	97.79 80

EPNet_dtu		86.49 221	85.94 207	88.14 261	90.24 298	72.82 299	94.11 162	92.20 263	86.66 118	79.42 292	92.36 203	73.52 186	95.81 283	71.26 282	93.66 134	95.80 153
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
TransMVSNet (Re)		84.43 254	83.06 256	88.54 249	91.72 245	78.44 233	95.18 91	92.82 251	82.73 198	79.67 289	92.12 212	73.49 187	95.96 275	71.10 286	68.73 332	91.21 308

Effi-MVS+-dtu		88.65 149	88.35 136	89.54 224	93.33 201	76.39 273	94.47 137	94.36 218	87.70 94	85.43 190	89.56 282	73.45 188	97.26 202	85.57 133	91.28 166	94.97 173

mvs-test1		89.45 126	89.14 118	90.38 190	93.33 201	77.63 256	94.95 104	94.36 218	87.70 94	87.10 151	92.81 191	73.45 188	98.03 145	85.57 133	93.04 149	95.48 159

baseline2		86.50 219	85.39 220	89.84 212	91.12 268	76.70 268	91.88 244	88.58 326	82.35 205	79.95 287	90.95 253	73.42 190	97.63 168	80.27 213	89.95 183	95.19 168

PEN-MVS		86.80 208	86.27 194	88.40 252	92.32 227	75.71 280	95.18 91	96.38 98	87.97 85	82.82 251	93.15 178	73.39 191	95.92 276	76.15 255	79.03 311	93.59 244

v1192		87.25 194	86.33 191	90.00 208	90.76 285	79.04 223	93.80 180	95.48 163	82.57 201	85.48 185	91.18 245	73.38 192	97.42 183	82.30 175	82.06 270	93.53 246

QAPM		89.51 123	88.15 143	93.59 70	94.92 142	84.58 79	96.82 24	96.70 77	78.43 265	83.41 243	96.19 79	73.18 193	99.30 39	77.11 246	96.54 93	96.89 115

tpmrst		85.35 238	84.99 227	86.43 294	90.88 281	67.88 329	88.71 297	91.43 286	80.13 244	86.08 171	88.80 291	73.05 194	96.02 272	82.48 171	83.40 258	95.40 163

PS-CasMVS		87.32 191	86.88 169	88.63 248	92.99 215	76.33 275	95.33 76	96.61 85	88.22 81	83.30 247	93.07 182	73.03 195	95.79 284	78.36 232	81.00 289	93.75 239

DTE-MVSNet		86.11 225	85.48 218	87.98 264	91.65 250	74.92 283	94.93 106	95.75 143	87.36 101	82.26 256	93.04 183	72.85 196	95.82 282	74.04 271	77.46 316	93.20 260

MVSTER		88.84 144	88.29 140	90.51 183	92.95 216	80.44 188	93.73 183	95.01 191	84.66 161	87.15 148	93.12 180	72.79 197	97.21 208	87.86 106	87.36 224	93.87 228

v1921920		86.97 204	86.06 202	89.69 221	90.53 294	78.11 242	93.80 180	95.43 171	81.90 217	85.33 198	91.05 251	72.66 198	97.41 188	82.05 180	81.80 275	93.53 246

DP-MVS		87.25 194	85.36 222	92.90 89	97.65 52	83.24 115	94.81 115	92.00 269	74.99 295	81.92 262	95.00 113	72.66 198	99.05 59	66.92 309	92.33 160	96.40 126

v7n		86.81 207	85.76 214	89.95 209	90.72 287	79.25 221	95.07 97	95.92 128	84.45 164	82.29 255	90.86 254	72.60 200	97.53 173	79.42 224	80.52 297	93.08 266

OPM-MVS		90.12 108	89.56 108	91.82 138	93.14 206	83.90 99	94.16 158	95.74 144	88.96 60	87.86 135	95.43 101	72.48 201	97.91 154	88.10 105	90.18 179	93.65 243

LS3D		87.89 167	86.32 192	92.59 103	96.07 101	82.92 126	95.23 86	94.92 199	75.66 288	82.89 250	95.98 85	72.48 201	99.21 46	68.43 301	95.23 115	95.64 156

pm-mvs1		86.61 214	85.54 216	89.82 213	91.44 252	80.18 190	95.28 84	94.85 203	83.84 173	81.66 263	92.62 196	72.45 203	96.48 251	79.67 219	78.06 312	92.82 275

PMMVS		85.71 233	84.96 229	87.95 265	88.90 313	77.09 264	88.68 298	90.06 313	72.32 317	86.47 160	90.76 259	72.15 204	94.40 308	81.78 187	93.49 138	92.36 287

PatchmatchNet		85.85 230	84.70 235	89.29 230	91.76 244	75.54 281	88.49 300	91.30 288	81.63 225	85.05 201	88.70 293	71.71 205	96.24 264	74.61 269	89.05 199	96.08 140
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
sam_mvs1												71.70 206				96.12 136

patchmatchnet-post											83.76 324	71.53 207	96.48 251

v1240		86.78 209	85.85 209	89.56 223	90.45 295	77.79 251	93.61 188	95.37 176	81.65 223	85.43 190	91.15 247	71.50 208	97.43 182	81.47 193	82.05 272	93.47 250

anonymousdsp		87.84 168	87.09 165	90.12 200	89.13 310	80.54 185	94.67 124	95.55 158	82.05 209	83.82 232	92.12 212	71.47 209	97.15 210	87.15 117	87.80 220	92.67 277

Patchmatch-test		81.37 283	79.30 288	87.58 271	90.92 278	74.16 289	80.99 336	87.68 331	70.52 325	76.63 306	88.81 289	71.21 210	92.76 325	60.01 332	86.93 230	95.83 151

F-COLMAP		87.95 166	86.80 173	91.40 151	96.35 91	80.88 176	94.73 120	95.45 168	79.65 251	82.04 260	94.61 128	71.13 211	98.50 107	76.24 254	91.05 171	94.80 185

pmmvs4		85.43 236	83.86 246	90.16 197	90.02 303	82.97 125	90.27 272	92.67 255	75.93 287	80.73 273	91.74 227	71.05 212	95.73 286	78.85 228	83.46 256	91.78 296

CR-MVSNet		85.35 238	83.76 247	90.12 200	90.58 291	79.34 215	85.24 323	91.96 273	78.27 267	85.55 179	87.87 306	71.03 213	95.61 287	73.96 273	89.36 193	95.40 163

Patchmtry		82.71 268	80.93 272	88.06 263	90.05 302	76.37 274	84.74 325	91.96 273	72.28 318	81.32 268	87.87 306	71.03 213	95.50 295	68.97 297	80.15 300	92.32 289

RPMNet		83.18 266	80.87 273	90.12 200	90.58 291	79.34 215	85.24 323	90.78 303	71.44 321	85.55 179	82.97 328	70.87 215	95.61 287	61.01 328	89.36 193	95.40 163

Patchmatch-RL test		81.67 277	79.96 282	86.81 292	85.42 330	71.23 312	82.17 334	87.50 332	78.47 264	77.19 304	82.50 329	70.81 216	93.48 318	82.66 170	72.89 324	95.71 155

CostFormer		85.77 232	84.94 230	88.26 257	91.16 267	72.58 305	89.47 287	91.04 295	76.26 284	86.45 163	89.97 274	70.74 217	96.86 231	82.35 174	87.07 229	95.34 166

sam_mvs												70.60 218

xiu_mvs_v1_base_debu		90.64 99	90.05 100	92.40 111	93.97 182	84.46 85	93.32 196	95.46 165	85.17 148	92.25 71	94.03 144	70.59 219	98.57 104	90.97 72	94.67 118	94.18 209

xiu_mvs_v1_base		90.64 99	90.05 100	92.40 111	93.97 182	84.46 85	93.32 196	95.46 165	85.17 148	92.25 71	94.03 144	70.59 219	98.57 104	90.97 72	94.67 118	94.18 209

xiu_mvs_v1_base_debi		90.64 99	90.05 100	92.40 111	93.97 182	84.46 85	93.32 196	95.46 165	85.17 148	92.25 71	94.03 144	70.59 219	98.57 104	90.97 72	94.67 118	94.18 209

test_post											10.29 348	70.57 222	95.91 278

CANet_DTU		90.26 107	89.41 112	92.81 91	93.46 199	83.01 123	93.48 192	94.47 215	89.43 46	87.76 140	94.23 142	70.54 223	99.03 63	84.97 137	96.39 97	96.38 127

BH-RMVSNet		88.37 155	87.48 156	91.02 166	95.28 127	79.45 211	92.89 218	93.07 247	85.45 143	86.91 154	94.84 121	70.35 224	97.76 159	73.97 272	94.59 122	95.85 149

Fast-Effi-MVS+-dtu		87.44 187	86.72 175	89.63 222	92.04 234	77.68 255	94.03 171	93.94 232	85.81 131	82.42 254	91.32 240	70.33 225	97.06 219	80.33 212	90.23 178	94.14 212

MDTV_nov1_ep13_2view						55.91 345	87.62 311		73.32 309	84.59 208		70.33 225		74.65 268		95.50 158

ACMM	84.12 9	89.14 135	88.48 135	91.12 158	94.65 155	81.22 168	95.31 77	96.12 115	85.31 147	85.92 172	94.34 134	70.19 227	98.06 143	85.65 131	88.86 201	94.08 217
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
ET-MVSNet_ETH3D		87.51 184	85.91 208	92.32 117	93.70 193	83.93 98	92.33 233	90.94 298	84.16 166	72.09 325	92.52 198	69.90 228	95.85 280	89.20 91	88.36 210	97.17 102

LPG-MVS_test		89.45 126	88.90 125	91.12 158	94.47 160	81.49 159	95.30 80	96.14 113	86.73 115	85.45 187	95.16 108	69.89 229	98.10 134	87.70 108	89.23 196	93.77 237

LGP-MVS_train				91.12 158	94.47 160	81.49 159		96.14 113	86.73 115	85.45 187	95.16 108	69.89 229	98.10 134	87.70 108	89.23 196	93.77 237

CHOSEN 280x420		85.15 243	83.99 244	88.65 247	92.47 223	78.40 235	79.68 338	92.76 252	74.90 297	81.41 266	89.59 280	69.85 231	95.51 293	79.92 217	95.29 112	92.03 293

LTVRE_ROB	82.13 13	86.26 224	84.90 231	90.34 193	94.44 164	81.50 157	92.31 235	94.89 200	83.03 191	79.63 290	92.67 194	69.69 232	97.79 157	71.20 283	86.26 232	91.72 297
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
OpenMVS	83.78 11	88.74 147	87.29 161	93.08 80	92.70 220	85.39 70	96.57 29	96.43 94	78.74 262	80.85 272	96.07 83	69.64 233	99.01 69	78.01 237	96.65 90	94.83 183

MDTV_nov1_ep13			83.56 251		91.69 248	69.93 323	87.75 308	91.54 282	78.60 263	84.86 204	88.90 288	69.54 234	96.03 271	70.25 288	88.93 200

PatchT		82.68 269	81.27 268	86.89 290	90.09 301	70.94 317	84.06 327	90.15 310	74.91 296	85.63 178	83.57 325	69.37 235	94.87 306	65.19 314	88.50 206	94.84 182

VPNet		88.20 160	87.47 157	90.39 188	93.56 196	79.46 210	94.04 170	95.54 160	88.67 67	86.96 152	94.58 131	69.33 236	97.15 210	84.05 150	80.53 296	94.56 195

ACMP	84.23 8	89.01 141	88.35 136	90.99 169	94.73 149	81.27 165	95.07 97	95.89 133	86.48 119	83.67 236	94.30 137	69.33 236	97.99 148	87.10 121	88.55 203	93.72 241
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
test_post1							88.00 305				9.81 349	69.31 238	95.53 291	76.65 249

tpmvs		83.35 265	82.07 262	87.20 284	91.07 270	71.00 316	88.31 303	91.70 277	78.91 256	80.49 278	87.18 313	69.30 239	97.08 217	68.12 305	83.56 254	93.51 249

thres200		87.21 198	86.24 195	90.12 200	95.36 123	78.53 230	93.26 203	92.10 265	86.42 121	88.00 134	91.11 249	69.24 240	98.00 147	69.58 295	91.04 172	93.83 232

tfpn200view9		87.58 181	86.64 179	90.41 187	95.99 104	78.64 227	94.58 128	91.98 271	86.94 110	88.09 129	91.77 225	69.18 241	98.10 134	70.13 291	91.10 167	94.48 202

thres400		87.62 179	86.64 179	90.57 177	95.99 104	78.64 227	94.58 128	91.98 271	86.94 110	88.09 129	91.77 225	69.18 241	98.10 134	70.13 291	91.10 167	94.96 176

tfpnnormal		84.72 251	83.23 254	89.20 232	92.79 219	80.05 197	94.48 134	95.81 138	82.38 203	81.08 270	91.21 242	69.01 243	96.95 225	61.69 326	80.59 294	90.58 317

thres100view900		87.63 177	86.71 176	90.38 190	96.12 96	78.55 229	95.03 101	91.58 280	87.15 103	88.06 132	92.29 206	68.91 244	98.10 134	70.13 291	91.10 167	94.48 202

thres600view7		87.65 174	86.67 178	90.59 176	96.08 100	78.72 225	94.88 110	91.58 280	87.06 106	88.08 131	92.30 205	68.91 244	98.10 134	70.05 294	91.10 167	94.96 176

PatchMatch-RL		86.77 212	85.54 216	90.47 186	95.88 107	82.71 134	90.54 269	92.31 260	79.82 249	84.32 220	91.57 235	68.77 246	96.39 257	73.16 277	93.48 140	92.32 289

XVG-OURS		89.40 131	88.70 128	91.52 147	94.06 174	81.46 161	91.27 259	96.07 118	86.14 127	88.89 122	95.77 93	68.73 247	97.26 202	87.39 113	89.96 182	95.83 151

TR-MVS		86.78 209	85.76 214	89.82 213	94.37 166	78.41 234	92.47 228	92.83 250	81.11 236	86.36 165	92.40 201	68.73 247	97.48 176	73.75 275	89.85 186	93.57 245

tpm		84.73 250	84.02 243	86.87 291	90.33 296	68.90 326	89.06 293	89.94 316	80.85 238	85.75 174	89.86 276	68.54 249	95.97 274	77.76 238	84.05 248	95.75 154

FMVSNet3		87.40 189	86.11 199	91.30 154	93.79 190	83.64 106	94.20 157	94.81 207	83.89 172	84.37 215	91.87 224	68.45 250	96.56 246	78.23 234	85.36 238	93.70 242

MVP-Stereo		85.97 227	84.86 232	89.32 229	90.92 278	82.19 145	92.11 241	94.19 225	78.76 261	78.77 295	91.63 231	68.38 251	96.56 246	75.01 266	93.95 130	89.20 323
Qingsong Yan: MVP-Stereo: A Parallel Multi-View Patchmatch Stereo Method with Dilation Matching for Photogrammetric Application.
tpm cat1		81.96 273	80.27 277	87.01 286	91.09 269	71.02 315	87.38 312	91.53 283	66.25 331	80.17 281	86.35 316	68.22 252	96.15 268	69.16 296	82.29 267	93.86 230

tpm2		84.08 256	82.94 257	87.48 275	91.39 257	71.27 311	89.23 291	90.37 307	71.95 319	84.64 206	89.33 283	67.30 253	96.55 248	75.17 263	87.09 228	94.63 188

test-LLR		85.87 229	85.41 219	87.25 280	90.95 274	71.67 309	89.55 283	89.88 319	83.41 183	84.54 209	87.95 303	67.25 254	95.11 302	81.82 185	93.37 143	94.97 173

test0.0.03 1		82.41 271	81.69 265	84.59 307	88.23 320	72.89 298	90.24 273	87.83 329	83.41 183	79.86 288	89.78 278	67.25 254	88.99 337	65.18 315	83.42 257	91.90 295

CVMVSNet		84.69 252	84.79 234	84.37 309	91.84 241	64.92 336	93.70 186	91.47 285	66.19 332	86.16 170	95.28 103	67.18 256	93.33 320	80.89 201	90.42 176	94.88 181

thisisatest0515		87.33 190	85.99 203	91.37 152	93.49 197	79.55 208	90.63 268	89.56 324	80.17 243	87.56 143	90.86 254	67.07 257	98.28 126	81.50 192	93.02 150	96.29 129

tttt0517		88.61 150	87.78 150	91.11 161	94.96 139	77.81 250	95.35 75	89.69 321	85.09 153	88.05 133	94.59 130	66.93 258	98.48 108	83.27 159	92.13 162	97.03 108

our_test_3		81.93 274	80.46 275	86.33 296	88.46 317	73.48 293	88.46 301	91.11 291	76.46 279	76.69 305	88.25 299	66.89 259	94.36 309	68.75 298	79.08 310	91.14 310

thisisatest0530		88.67 148	87.61 154	91.86 135	94.87 145	80.07 195	94.63 126	89.90 318	84.00 170	88.46 126	93.78 162	66.88 260	98.46 111	83.30 158	92.65 155	97.06 106

IterMVS-SCA-FT		85.45 235	84.53 239	88.18 260	91.71 246	76.87 267	90.19 276	92.65 256	85.40 145	81.44 265	90.54 261	66.79 261	95.00 305	81.04 196	81.05 285	92.66 278

SCA		86.32 223	85.18 224	89.73 219	92.15 229	76.60 269	91.12 262	91.69 278	83.53 180	85.50 184	88.81 289	66.79 261	96.48 251	76.65 249	90.35 177	96.12 136

D2MVS		85.90 228	85.09 226	88.35 254	90.79 283	77.42 260	91.83 246	95.70 146	80.77 239	80.08 285	90.02 272	66.74 263	96.37 258	81.88 184	87.97 217	91.26 306

IterMVS		84.88 248	83.98 245	87.60 270	91.44 252	76.03 277	90.18 277	92.41 259	83.24 188	81.06 271	90.42 265	66.60 264	94.28 311	79.46 220	80.98 290	92.48 282
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
GBi-Net		87.26 192	85.98 204	91.08 162	94.01 177	83.10 118	95.14 94	94.94 194	83.57 177	84.37 215	91.64 228	66.59 265	96.34 261	78.23 234	85.36 238	93.79 233

test1		87.26 192	85.98 204	91.08 162	94.01 177	83.10 118	95.14 94	94.94 194	83.57 177	84.37 215	91.64 228	66.59 265	96.34 261	78.23 234	85.36 238	93.79 233

FMVSNet2		87.19 199	85.82 210	91.30 154	94.01 177	83.67 105	94.79 116	94.94 194	83.57 177	83.88 230	92.05 219	66.59 265	96.51 249	77.56 241	85.01 241	93.73 240

EPMVS		83.90 259	82.70 261	87.51 272	90.23 299	72.67 301	88.62 299	81.96 341	81.37 230	85.01 202	88.34 297	66.31 268	94.45 307	75.30 262	87.12 227	95.43 162

ppachtmachnet_test		81.84 275	80.07 281	87.15 285	88.46 317	74.43 286	89.04 294	92.16 264	75.33 291	77.75 300	88.99 286	66.20 269	95.37 298	65.12 316	77.60 314	91.65 298

MDA-MVSNet_test_wron		79.21 299	77.19 300	85.29 302	88.22 321	72.77 300	85.87 319	90.06 313	74.34 301	62.62 337	87.56 309	66.14 270	91.99 329	66.90 310	73.01 322	91.10 312

YYNet1		79.22 298	77.20 299	85.28 303	88.20 322	72.66 302	85.87 319	90.05 315	74.33 302	62.70 336	87.61 308	66.09 271	92.03 328	66.94 307	72.97 323	91.15 309

JIA-IIPM		81.04 286	78.98 293	87.25 280	88.64 314	73.48 293	81.75 335	89.61 323	73.19 310	82.05 259	73.71 335	66.07 272	95.87 279	71.18 285	84.60 244	92.41 285

RRT_MVS		88.86 143	87.68 152	92.39 114	92.02 236	86.09 52	94.38 148	94.94 194	85.45 143	87.14 150	93.84 160	65.88 273	97.11 214	88.73 96	86.77 231	93.98 222

MSDG		84.86 249	83.09 255	90.14 199	93.80 188	80.05 197	89.18 292	93.09 246	78.89 257	78.19 296	91.91 222	65.86 274	97.27 200	68.47 300	88.45 207	93.11 264

jajsoiax		88.24 159	87.50 155	90.48 185	90.89 280	80.14 192	95.31 77	95.65 152	84.97 155	84.24 224	94.02 147	65.31 275	97.42 183	88.56 98	88.52 205	93.89 225

cascas		86.43 222	84.98 228	90.80 174	92.10 233	80.92 175	90.24 273	95.91 130	73.10 311	83.57 240	88.39 296	65.15 276	97.46 178	84.90 140	91.43 165	94.03 220

ADS-MVSNet2		81.66 278	79.71 285	87.50 273	91.35 259	74.19 288	83.33 330	88.48 327	72.90 313	82.24 257	85.77 318	64.98 277	93.20 322	64.57 318	83.74 250	95.12 169

ADS-MVSNet		81.56 280	79.78 283	86.90 289	91.35 259	71.82 308	83.33 330	89.16 325	72.90 313	82.24 257	85.77 318	64.98 277	93.76 315	64.57 318	83.74 250	95.12 169

pmmvs5		84.21 255	82.84 260	88.34 255	88.95 312	76.94 266	92.41 229	91.91 275	75.63 289	80.28 280	91.18 245	64.59 279	95.57 289	77.09 247	83.47 255	92.53 281

PVSNet	78.82 18	85.55 234	84.65 236	88.23 259	94.72 150	71.93 307	87.12 313	92.75 253	78.80 260	84.95 203	90.53 262	64.43 280	96.71 235	74.74 267	93.86 132	96.06 142

UGNet		89.95 113	88.95 123	92.95 87	94.51 159	83.31 114	95.70 63	95.23 181	89.37 48	87.58 142	93.94 152	64.00 281	98.78 94	83.92 151	96.31 98	96.74 119
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
RPSCF		85.07 244	84.27 240	87.48 275	92.91 217	70.62 319	91.69 252	92.46 258	76.20 285	82.67 253	95.22 106	63.94 282	97.29 199	77.51 242	85.80 235	94.53 196

mvs_tets		88.06 165	87.28 162	90.38 190	90.94 276	79.88 203	95.22 87	95.66 150	85.10 152	84.21 225	93.94 152	63.53 283	97.40 190	88.50 99	88.40 209	93.87 228

Anonymous20231211		86.59 216	85.13 225	90.98 171	96.52 86	81.50 157	96.14 42	96.16 112	73.78 305	83.65 237	92.15 210	63.26 284	97.37 194	82.82 167	81.74 277	94.06 218

dp		81.47 282	80.23 278	85.17 304	89.92 305	65.49 335	86.74 314	90.10 312	76.30 283	81.10 269	87.12 314	62.81 285	95.92 276	68.13 304	79.88 303	94.09 216

LFMVS		90.08 109	89.13 119	92.95 87	96.71 78	82.32 144	96.08 46	89.91 317	86.79 113	92.15 76	96.81 49	62.60 286	98.34 121	87.18 116	93.90 131	98.19 52

DWT-MVSNet_test		84.95 247	83.68 248	88.77 241	91.43 255	73.75 291	91.74 249	90.98 296	80.66 240	83.84 231	87.36 310	62.44 287	97.11 214	78.84 229	85.81 234	95.46 160

Anonymous20231206		81.03 287	79.77 284	84.82 306	87.85 325	70.26 321	91.42 256	92.08 266	73.67 306	77.75 300	89.25 284	62.43 288	93.08 323	61.50 327	82.00 273	91.12 311

VDD-MVS		90.74 94	89.92 105	93.20 75	96.27 92	83.02 122	95.73 61	93.86 234	88.42 75	92.53 67	96.84 46	62.09 289	98.64 99	90.95 75	92.62 156	97.93 73

MS-PatchMatch		85.05 245	84.16 241	87.73 268	91.42 256	78.51 231	91.25 260	93.53 240	77.50 272	80.15 282	91.58 233	61.99 290	95.51 293	75.69 258	94.35 128	89.16 324

OurMVSNet-221017-0		85.35 238	84.64 237	87.49 274	90.77 284	72.59 304	94.01 173	94.40 217	84.72 160	79.62 291	93.17 177	61.91 291	96.72 233	81.99 181	81.16 281	93.16 262

test20.03		79.95 293	79.08 291	82.55 315	85.79 329	67.74 330	91.09 263	91.08 292	81.23 234	74.48 317	89.96 275	61.63 292	90.15 335	60.08 330	76.38 318	89.76 319

DSMNet-mixed		76.94 303	76.29 303	78.89 319	83.10 337	56.11 344	87.78 307	79.77 343	60.65 336	75.64 311	88.71 292	61.56 293	88.34 338	60.07 331	89.29 195	92.21 292

Anonymous20240529		88.09 163	86.59 183	92.58 104	96.53 85	81.92 151	95.99 50	95.84 136	74.11 303	89.06 120	95.21 107	61.44 294	98.81 92	83.67 156	87.47 221	97.01 109

IB-MVS	80.51 15	85.24 242	83.26 253	91.19 156	92.13 231	79.86 204	91.75 248	91.29 289	83.28 187	80.66 275	88.49 295	61.28 295	98.46 111	80.99 199	79.46 307	95.25 167
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
GA-MVS		86.61 214	85.27 223	90.66 175	91.33 261	78.71 226	90.40 271	93.81 237	85.34 146	85.12 200	89.57 281	61.25 296	97.11 214	80.99 199	89.59 190	96.15 133

N_pmnet		68.89 309	68.44 311	70.23 325	89.07 311	28.79 353	88.06 304	19.50 354	69.47 327	71.86 327	84.93 321	61.24 297	91.75 331	54.70 336	77.15 317	90.15 318

EU-MVSNet		81.32 284	80.95 271	82.42 316	88.50 316	63.67 337	93.32 196	91.33 287	64.02 334	80.57 277	92.83 189	61.21 298	92.27 327	76.34 252	80.38 299	91.32 304

VDDNet		89.56 122	88.49 134	92.76 94	95.07 134	82.09 146	96.30 35	93.19 245	81.05 237	91.88 80	96.86 45	61.16 299	98.33 123	88.43 100	92.49 159	97.84 77

PVSNet_0	73.20 20	77.22 302	74.83 306	84.37 309	90.70 288	71.10 314	83.09 332	89.67 322	72.81 315	73.93 319	83.13 327	60.79 300	93.70 316	68.54 299	50.84 341	88.30 331

RRT_test8_iter05		86.90 205	86.36 189	88.52 250	93.00 214	73.27 295	94.32 151	95.96 125	85.50 142	84.26 223	92.86 186	60.76 301	97.70 164	88.32 101	82.29 267	94.60 191

SixPastTwentyTwo		83.91 258	82.90 258	86.92 288	90.99 272	70.67 318	93.48 192	91.99 270	85.54 140	77.62 302	92.11 214	60.59 302	96.87 230	76.05 256	77.75 313	93.20 260

gg-mvs-nofinetune		81.77 276	79.37 287	88.99 239	90.85 282	77.73 254	86.29 317	79.63 344	74.88 298	83.19 248	69.05 338	60.34 303	96.11 269	75.46 260	94.64 121	93.11 264

MDA-MVSNet-bldmvs		78.85 300	76.31 302	86.46 293	89.76 307	73.88 290	88.79 296	90.42 306	79.16 254	59.18 338	88.33 298	60.20 304	94.04 313	62.00 325	68.96 330	91.48 302

pmmvs6		83.42 262	81.60 266	88.87 240	88.01 323	77.87 248	94.96 103	94.24 224	74.67 299	78.80 294	91.09 250	60.17 305	96.49 250	77.06 248	75.40 320	92.23 291

ACMH	80.38 17	85.36 237	83.68 248	90.39 188	94.45 163	80.63 182	94.73 120	94.85 203	82.09 208	77.24 303	92.65 195	60.01 306	97.58 169	72.25 280	84.87 242	92.96 269
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
GG-mvs-BLEND				87.94 266	89.73 308	77.91 245	87.80 306	78.23 346		80.58 276	83.86 323	59.88 307	95.33 299	71.20 283	92.22 161	90.60 316

UniMVSNet_ETH3D		87.53 183	86.37 188	91.00 168	92.44 224	78.96 224	94.74 119	95.61 154	84.07 169	85.36 197	94.52 132	59.78 308	97.34 195	82.93 163	87.88 218	96.71 120

pmmvs-eth3d		80.97 288	78.72 294	87.74 267	84.99 333	79.97 202	90.11 278	91.65 279	75.36 290	73.51 320	86.03 317	59.45 309	93.96 314	75.17 263	72.21 325	89.29 322

test_0402		81.30 285	79.17 290	87.67 269	93.19 205	78.17 240	92.98 215	91.71 276	75.25 292	76.02 310	90.31 266	59.23 310	96.37 258	50.22 338	83.63 253	88.47 330

FMVSNet1		85.85 230	84.11 242	91.08 162	92.81 218	83.10 118	95.14 94	94.94 194	81.64 224	82.68 252	91.64 228	59.01 311	96.34 261	75.37 261	83.78 249	93.79 233

COLMAP_ROB	80.39 16	83.96 257	82.04 263	89.74 217	95.28 127	79.75 206	94.25 154	92.28 261	75.17 293	78.02 299	93.77 163	58.60 312	97.84 156	65.06 317	85.92 233	91.63 299
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
ACMH+	81.04 14	85.05 245	83.46 252	89.82 213	94.66 154	79.37 213	94.44 139	94.12 230	82.19 207	78.04 298	92.82 190	58.23 313	97.54 172	73.77 274	82.90 263	92.54 280

CMPMVS	59.16 21	80.52 290	79.20 289	84.48 308	83.98 334	67.63 331	89.95 281	93.84 236	64.79 333	66.81 334	91.14 248	57.93 314	95.17 300	76.25 253	88.10 213	90.65 313
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
ITE_SJBPF				88.24 258	91.88 240	77.05 265		92.92 248	85.54 140	80.13 284	93.30 172	57.29 315	96.20 265	72.46 279	84.71 243	91.49 301

TESTMET0.1,1		83.74 260	82.85 259	86.42 295	89.96 304	71.21 313	89.55 283	87.88 328	77.41 273	83.37 244	87.31 311	56.71 316	93.65 317	80.62 206	92.85 154	94.40 205

UnsupCasMVSNet_eth		80.07 292	78.27 295	85.46 301	85.24 331	72.63 303	88.45 302	94.87 202	82.99 192	71.64 328	88.07 302	56.34 317	91.75 331	73.48 276	63.36 337	92.01 294

K. test v3		81.59 279	80.15 280	85.91 299	89.89 306	69.42 325	92.57 226	87.71 330	85.56 139	73.44 321	89.71 279	55.58 318	95.52 292	77.17 245	69.76 328	92.78 276

test-mter		84.54 253	83.64 250	87.25 280	90.95 274	71.67 309	89.55 283	89.88 319	79.17 253	84.54 209	87.95 303	55.56 319	95.11 302	81.82 185	93.37 143	94.97 173

lessismore_v0				86.04 297	88.46 317	68.78 327		80.59 342		73.01 323	90.11 270	55.39 320	96.43 256	75.06 265	65.06 334	92.90 271

MVS-HIRNet		73.70 306	72.20 308	78.18 321	91.81 243	56.42 343	82.94 333	82.58 339	55.24 338	68.88 330	66.48 339	55.32 321	95.13 301	58.12 333	88.42 208	83.01 335

new-patchmatchnet		76.41 304	75.17 305	80.13 318	82.65 339	59.61 339	87.66 310	91.08 292	78.23 269	69.85 329	83.22 326	54.76 322	91.63 333	64.14 320	64.89 335	89.16 324

Anonymous202405211		87.68 172	86.13 197	92.31 118	96.66 79	80.74 180	94.87 111	91.49 284	80.47 241	89.46 114	95.44 99	54.72 323	98.23 127	82.19 177	89.89 184	97.97 69

XVG-ACMP-BASELINE		86.00 226	84.84 233	89.45 228	91.20 263	78.00 243	91.70 251	95.55 158	85.05 154	82.97 249	92.25 208	54.49 324	97.48 176	82.93 163	87.45 223	92.89 272

USDC		82.76 267	81.26 269	87.26 279	91.17 265	74.55 285	89.27 289	93.39 243	78.26 268	75.30 312	92.08 216	54.43 325	96.63 237	71.64 281	85.79 236	90.61 314

AllTest		83.42 262	81.39 267	89.52 225	95.01 135	77.79 251	93.12 208	90.89 300	77.41 273	76.12 308	93.34 168	54.08 326	97.51 174	68.31 302	84.27 246	93.26 255

TestCases				89.52 225	95.01 135	77.79 251		90.89 300	77.41 273	76.12 308	93.34 168	54.08 326	97.51 174	68.31 302	84.27 246	93.26 255

MIMVSNet		82.59 270	80.53 274	88.76 242	91.51 251	78.32 236	86.57 316	90.13 311	79.32 252	80.70 274	88.69 294	52.98 328	93.07 324	66.03 312	88.86 201	94.90 180

FMVSNet5		81.52 281	79.60 286	87.27 278	91.17 265	77.95 244	91.49 255	92.26 262	76.87 278	76.16 307	87.91 305	51.67 329	92.34 326	67.74 306	81.16 281	91.52 300

testgi		80.94 289	80.20 279	83.18 313	87.96 324	66.29 332	91.28 258	90.70 305	83.70 175	78.12 297	92.84 188	51.37 330	90.82 334	63.34 321	82.46 266	92.43 284

UnsupCasMVSNet_bld		76.23 305	73.27 307	85.09 305	83.79 335	72.92 297	85.65 322	93.47 242	71.52 320	68.84 331	79.08 333	49.77 331	93.21 321	66.81 311	60.52 339	89.13 326

OpenMVS_ROB	74.94 19	79.51 296	77.03 301	86.93 287	87.00 326	76.23 276	92.33 233	90.74 304	68.93 328	74.52 316	88.23 300	49.58 332	96.62 238	57.64 334	84.29 245	87.94 332

testing_2		83.40 264	81.02 270	90.56 179	85.06 332	80.51 186	91.37 257	95.57 156	82.92 194	67.06 333	85.54 320	49.47 333	97.24 204	86.74 122	85.44 237	93.93 223

TDRefinement		79.81 294	77.34 297	87.22 283	79.24 341	75.48 282	93.12 208	92.03 268	76.45 280	75.01 313	91.58 233	49.19 334	96.44 255	70.22 290	69.18 329	89.75 320

MIMVSNet1		79.38 297	77.28 298	85.69 300	86.35 328	73.67 292	91.61 254	92.75 253	78.11 271	72.64 324	88.12 301	48.16 335	91.97 330	60.32 329	77.49 315	91.43 303

MVS_0304		83.46 261	81.92 264	88.10 262	90.63 290	77.49 259	93.26 203	93.75 238	80.04 246	80.44 279	87.24 312	47.94 336	95.55 290	75.79 257	88.16 212	91.26 306

LF4IMVS		80.37 291	79.07 292	84.27 311	86.64 327	69.87 324	89.39 288	91.05 294	76.38 281	74.97 314	90.00 273	47.85 337	94.25 312	74.55 270	80.82 292	88.69 328

EG-PatchMatch MVS		82.37 272	80.34 276	88.46 251	90.27 297	79.35 214	92.80 220	94.33 220	77.14 277	73.26 322	90.18 268	47.47 338	96.72 233	70.25 288	87.32 226	89.30 321

TinyColmap		79.76 295	77.69 296	85.97 298	91.71 246	73.12 296	89.55 283	90.36 308	75.03 294	72.03 326	90.19 267	46.22 339	96.19 267	63.11 322	81.03 286	88.59 329

tmp_tt		35.64 319	39.24 320	24.84 333	14.87 353	23.90 354	62.71 344	51.51 353	6.58 349	36.66 345	62.08 342	44.37 340	30.34 351	52.40 337	22.00 347	20.27 346

new_pmnet		72.15 307	70.13 309	78.20 320	82.95 338	65.68 333	83.91 328	82.40 340	62.94 335	64.47 335	79.82 332	42.85 341	86.26 340	57.41 335	74.44 321	82.65 336

pmmvs3		71.81 308	68.71 310	81.11 317	75.86 342	70.42 320	86.74 314	83.66 338	58.95 337	68.64 332	80.89 331	36.93 342	89.52 336	63.10 323	63.59 336	83.39 334

PM-MVS		78.11 301	76.12 304	84.09 312	83.54 336	70.08 322	88.97 295	85.27 336	79.93 247	74.73 315	86.43 315	34.70 343	93.48 318	79.43 223	72.06 326	88.72 327

ambc				83.06 314	79.99 340	63.51 338	77.47 339	92.86 249		74.34 318	84.45 322	28.74 344	95.06 304	73.06 278	68.89 331	90.61 314

DeepMVS_CX				56.31 330	74.23 343	51.81 346		56.67 352	44.85 342	48.54 342	75.16 334	27.87 345	58.74 349	40.92 341	52.22 340	58.39 342

FPMVS		64.63 311	62.55 312	70.88 324	70.80 344	56.71 341	84.42 326	84.42 337	51.78 340	49.57 340	81.61 330	23.49 346	81.48 343	40.61 342	76.25 319	74.46 339

ANet_high		58.88 313	54.22 316	72.86 323	56.50 351	56.67 342	80.75 337	86.00 333	73.09 312	37.39 344	64.63 341	22.17 347	79.49 345	43.51 340	23.96 345	82.43 337

EMVS		42.07 318	41.12 319	44.92 332	63.45 349	35.56 352	73.65 340	63.48 349	33.05 346	26.88 349	45.45 346	21.27 348	67.14 347	19.80 347	23.02 346	32.06 345

Gipuma		57.99 314	54.91 315	67.24 327	88.51 315	65.59 334	52.21 346	90.33 309	43.58 343	42.84 343	51.18 344	20.29 349	85.07 341	34.77 343	70.45 327	51.05 343
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
E-PMN		43.23 317	42.29 318	46.03 331	65.58 347	37.41 350	73.51 341	64.62 348	33.99 345	28.47 348	47.87 345	19.90 350	67.91 346	22.23 346	24.45 344	32.77 344

PMMVS2		59.60 312	56.40 314	69.21 326	68.83 345	46.58 348	73.02 343	77.48 347	55.07 339	49.21 341	72.95 337	17.43 351	80.04 344	49.32 339	44.33 342	80.99 338

LCM-MVSNet		66.00 310	62.16 313	77.51 322	64.51 348	58.29 340	83.87 329	90.90 299	48.17 341	54.69 339	73.31 336	16.83 352	86.75 339	65.47 313	61.67 338	87.48 333

PMVS	47.18 22	52.22 315	48.46 317	63.48 328	45.72 352	46.20 349	73.41 342	78.31 345	41.03 344	30.06 346	65.68 340	6.05 353	83.43 342	30.04 344	65.86 333	60.80 340
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
MVE	39.65 23	43.39 316	38.59 321	57.77 329	56.52 350	48.77 347	55.38 345	58.64 351	29.33 347	28.96 347	52.65 343	4.68 354	64.62 348	28.11 345	33.07 343	59.93 341
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
wuyk23d		21.27 321	20.48 323	23.63 334	68.59 346	36.41 351	49.57 347	6.85 355	9.37 348	7.89 350	4.46 352	4.03 355	31.37 350	17.47 348	16.07 348	3.12 347

test123		8.76 323	11.22 325	1.39 335	0.85 355	0.97 355	85.76 321	0.35 357	0.54 351	2.45 352	8.14 351	0.60 356	0.48 352	2.16 350	0.17 350	2.71 348

testmvs		8.92 322	11.52 324	1.12 336	1.06 354	0.46 356	86.02 318	0.65 356	0.62 350	2.74 351	9.52 350	0.31 357	0.45 353	2.38 349	0.39 349	2.46 349

uanet_test		0.00 326	0.00 328	0.00 337	0.00 356	0.00 357	0.00 348	0.00 358	0.00 352	0.00 353	0.00 353	0.00 358	0.00 354	0.00 351	0.00 351	0.00 350

test_part1				0.00 337		0.00 357	0.00 348	97.45 11				0.00 358	0.00 354	0.00 351	0.00 351	0.00 350

sosnet-low-res		0.00 326	0.00 328	0.00 337	0.00 356	0.00 357	0.00 348	0.00 358	0.00 352	0.00 353	0.00 353	0.00 358	0.00 354	0.00 351	0.00 351	0.00 350

sosnet		0.00 326	0.00 328	0.00 337	0.00 356	0.00 357	0.00 348	0.00 358	0.00 352	0.00 353	0.00 353	0.00 358	0.00 354	0.00 351	0.00 351	0.00 350

uncertanet		0.00 326	0.00 328	0.00 337	0.00 356	0.00 357	0.00 348	0.00 358	0.00 352	0.00 353	0.00 353	0.00 358	0.00 354	0.00 351	0.00 351	0.00 350

Regformer		0.00 326	0.00 328	0.00 337	0.00 356	0.00 357	0.00 348	0.00 358	0.00 352	0.00 353	0.00 353	0.00 358	0.00 354	0.00 351	0.00 351	0.00 350

ab-mvs-re		7.82 324	10.43 326	0.00 337	0.00 356	0.00 357	0.00 348	0.00 358	0.00 352	0.00 353	93.88 157	0.00 358	0.00 354	0.00 351	0.00 351	0.00 350

uanet		0.00 326	0.00 328	0.00 337	0.00 356	0.00 357	0.00 348	0.00 358	0.00 352	0.00 353	0.00 353	0.00 358	0.00 354	0.00 351	0.00 351	0.00 350

IU-MVS					98.77 4	86.00 53		96.84 63	81.26 233	97.26 6				95.50 7	99.13 3	99.03 4

save fliter					97.85 44	85.63 67	95.21 88	96.82 66	89.44 44

test_0728_SECOND				95.01 15	98.79 1	86.43 40	97.09 11	97.49 5					99.61 3	95.62 5	99.08 7	98.99 5

GSMVS																96.12 136

test_part2					98.55 11	87.22 16				96.40 11

MTGPA								96.97 49

MTMP							96.16 40	60.64 350

gm-plane-assit					89.60 309	68.00 328			77.28 276		88.99 286		97.57 170	79.44 222

test9_res														91.91 54	98.71 30	98.07 62

agg_prior2														90.54 80	98.68 35	98.27 46

agg_prior					97.38 60	85.92 59		96.72 75		92.16 74			98.97 76

test_prior4						85.96 56	94.11 162

test_prior				93.82 64	97.29 65	84.49 82		96.88 59					98.87 83			98.11 60

旧先验2							93.36 195		71.25 322	94.37 26			97.13 213	86.74 122

新几何2							93.11 210

无先验							93.28 202	96.26 103	73.95 304				99.05 59	80.56 207		96.59 122

原ACMM2							92.94 217

testdata2													98.75 95	78.30 233

testdata1							92.15 239		87.94 86

plane_prior7					94.70 152	82.74 131

plane_prior5								96.22 108					98.12 132	88.15 102	89.99 180	94.63 188

plane_prior4											94.86 118

plane_prior3						82.75 129			90.26 30	86.91 154

plane_prior2							95.85 57		90.81 17

plane_prior1					94.59 156

plane_prior						82.73 132	95.21 88		89.66 41						89.88 185

n2								0.00 358

nn								0.00 358

door-mid								85.49 334

test11								96.57 88

door								85.33 335

HQP5-MVS						81.56 155

HQP-NCC					94.17 171		94.39 144		88.81 61	85.43 190

ACMP_Plane					94.17 171		94.39 144		88.81 61	85.43 190

BP-MVS														87.11 119

HQP4-MVS										85.43 190			97.96 150			94.51 198

HQP3-MVS								96.04 121							89.77 187

NP-MVS					94.37 166	82.42 139					93.98 150

ACMMP++_ref															87.47 221

ACMMP++															88.01 216

High-resolution multi-view benchmark