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.

test_part1								97.45 6				91.93 1			99.02 3	98.67 5

ESAPD		95.32 3	95.38 3	95.17 7	98.55 5	87.22 11	95.99 36	97.45 6	88.25 66	96.40 3	97.60 5	91.93 1	99.62 1	93.18 19	99.02 3	98.67 5

HSP-MVS		95.30 4	95.48 2	94.76 25	98.49 10	86.52 29	96.91 15	96.73 55	91.73 9	96.10 6	96.69 39	89.90 3	99.30 30	94.70 3	98.04 50	98.45 19

DeepPCF-MVS	89.96 1	94.20 26	94.77 10	92.49 89	96.52 67	80.00 173	94.00 168	97.08 30	90.05 26	95.65 8	97.29 13	89.66 4	98.97 62	93.95 9	98.71 20	98.50 12

SD-MVS		94.96 8	95.33 5	93.88 50	97.25 53	86.69 22	96.19 30	97.11 29	90.42 24	96.95 2	97.27 14	89.53 5	96.91 217	94.38 6	98.85 9	98.03 50

CNVR-MVS		95.40 2	95.37 4	95.50 4	98.11 26	88.51 3	95.29 64	96.96 38	92.09 3	95.32 10	97.08 26	89.49 6	99.33 27	95.10 2	98.85 9	98.66 7

APDe-MVS		95.46 1	95.64 1	94.91 13	98.26 20	86.29 39	97.46 2	97.40 9	89.03 47	96.20 5	98.10 1	89.39 7	99.34 24	95.88 1	99.03 2	99.10 1

MCST-MVS		94.45 14	94.20 21	95.19 6	98.46 12	87.50 9	95.00 87	97.12 27	87.13 90	92.51 51	96.30 55	89.24 8	99.34 24	93.46 13	98.62 33	98.73 3

TSAR-MVS + MP.		94.85 9	94.94 8	94.58 32	98.25 21	86.33 35	96.11 32	96.62 66	88.14 70	96.10 6	96.96 29	89.09 9	98.94 66	94.48 5	98.68 25	98.48 14

SteuartSystems-ACMMP		95.20 5	95.32 6	94.85 17	96.99 56	86.33 35	97.33 3	97.30 18	91.38 12	95.39 9	97.46 10	88.98 10	99.40 22	94.12 8	98.89 8	98.82 2
Skip Steuart: Steuart Systems R&D Blog.
SMA-MVS		95.20 5	95.10 7	95.51 3	98.14 25	88.26 4	96.26 28	97.31 17	86.04 116	97.82 1	98.10 1	88.43 11	99.56 3	94.66 4	99.13 1	98.71 4

HPM-MVS++		95.14 7	94.91 9	95.83 1	98.25 21	89.65 1	95.92 41	96.96 38	91.75 8	94.02 20	96.83 33	88.12 12	99.55 8	93.41 16	98.94 6	98.28 29

agg_prior1		93.29 44	92.97 46	94.26 43	97.38 44	85.92 45	93.92 171	96.72 57	81.96 209	92.16 57	96.23 58	87.85 13	98.97 62	91.95 42	98.55 38	97.90 59

CSCG		93.23 49	93.05 43	93.76 56	98.04 30	84.07 78	96.22 29	97.37 10	84.15 153	90.05 85	95.66 80	87.77 14	99.15 39	89.91 66	98.27 43	98.07 46

NCCC		94.81 10	94.69 11	95.17 7	97.83 33	87.46 10	95.66 52	96.93 41	92.34 2	93.94 21	96.58 46	87.74 15	99.44 21	92.83 23	98.40 40	98.62 8

TEST9					97.53 37	86.49 30	94.07 160	96.78 51	81.61 227	92.77 41	96.20 60	87.71 16	99.12 42

train_agg		93.44 40	93.08 42	94.52 34	97.53 37	86.49 30	94.07 160	96.78 51	81.86 222	92.77 41	96.20 60	87.63 17	99.12 42	92.14 36	98.69 22	97.94 54

test_8					97.49 40	86.30 38	94.02 166	96.76 54	81.86 222	92.70 45	96.20 60	87.63 17	99.02 54

TSAR-MVS + GP.		93.66 36	93.41 37	94.41 39	96.59 64	86.78 19	94.40 128	93.93 218	89.77 32	94.21 16	95.59 82	87.35 19	98.61 86	92.72 24	96.15 81	97.83 62

APD-MVS		94.24 23	94.07 25	94.75 26	98.06 29	86.90 16	95.88 42	96.94 40	85.68 122	95.05 12	97.18 22	87.31 20	99.07 45	91.90 46	98.61 34	98.28 29
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
Regformer-2		94.33 20	94.22 18	94.68 28	95.54 102	86.75 21	94.57 117	96.70 59	91.84 6	94.41 13	96.56 48	87.19 21	99.13 41	93.50 12	97.65 58	98.16 39

segment_acmp												87.16 22

Regformer-1		94.22 24	94.13 23	94.51 35	95.54 102	86.36 34	94.57 117	96.44 73	91.69 10	94.32 15	96.56 48	87.05 23	99.03 51	93.35 17	97.65 58	98.15 40

agg_prior3		93.27 45	92.89 48	94.40 40	97.49 40	86.12 42	94.07 160	96.73 55	81.46 230	92.46 53	96.05 68	86.90 24	99.15 39	92.14 36	98.69 22	97.94 54

旧先验1					96.79 60	81.81 128		95.67 124			96.81 34	86.69 25			97.66 57	96.97 89

test_prior3		93.60 37	93.53 36	93.82 52	97.29 49	84.49 65	94.12 152	96.88 44	87.67 81	92.63 46	96.39 53	86.62 26	98.87 68	91.50 50	98.67 27	98.11 44

test_prior2							94.12 152		87.67 81	92.63 46	96.39 53	86.62 26		91.50 50	98.67 27

CDPH-MVS		92.83 53	92.30 55	94.44 36	97.79 34	86.11 43	94.06 163	96.66 63	80.09 241	92.77 41	96.63 43	86.62 26	99.04 50	87.40 91	98.66 29	98.17 38

DELS-MVS		93.43 41	93.25 39	93.97 47	95.42 107	85.04 56	93.06 215	97.13 26	90.74 20	91.84 63	95.09 93	86.32 29	99.21 33	91.22 53	98.45 39	97.65 66
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
HFP-MVS		94.52 12	94.40 13	94.86 15	98.61 3	86.81 17	96.94 10	97.34 11	88.63 56	93.65 24	97.21 19	86.10 30	99.49 17	92.35 30	98.77 15	98.30 27

#test#		94.32 21	94.14 22	94.86 15	98.61 3	86.81 17	96.43 23	97.34 11	87.51 84	93.65 24	97.21 19	86.10 30	99.49 17	91.68 48	98.77 15	98.30 27

MVS_111021_HR		93.45 39	93.31 38	93.84 51	96.99 56	84.84 57	93.24 208	97.24 20	88.76 53	91.60 69	95.85 74	86.07 32	98.66 81	91.91 43	98.16 46	98.03 50

Regformer-4		93.91 31	93.81 29	94.19 45	95.36 108	85.47 52	94.68 109	96.41 76	91.60 11	93.75 23	96.71 37	85.95 33	99.10 44	93.21 18	96.65 73	98.01 52

ACMMP_Plus		94.74 11	94.56 12	95.28 5	98.02 31	87.70 5	95.68 50	97.34 11	88.28 65	95.30 11	97.67 4	85.90 34	99.54 11	93.91 10	98.95 5	98.60 9

Regformer-3		93.68 35	93.64 35	93.81 54	95.36 108	84.61 61	94.68 109	95.83 114	91.27 13	93.60 27	96.71 37	85.75 35	98.86 71	92.87 22	96.65 73	97.96 53

PHI-MVS		93.89 32	93.65 34	94.62 31	96.84 59	86.43 32	96.69 21	97.49 4	85.15 133	93.56 30	96.28 56	85.60 36	99.31 29	92.45 26	98.79 12	98.12 43

MP-MVS-pluss		94.21 25	94.00 27	94.85 17	98.17 24	86.65 25	94.82 98	97.17 25	86.26 111	92.83 39	97.87 3	85.57 37	99.56 3	94.37 7	98.92 7	98.34 24
MP-MVS-pluss: MP-MVS-pluss. MP-MVS-pluss
MP-MVS		94.25 22	94.07 25	94.77 24	98.47 11	86.31 37	96.71 20	96.98 34	89.04 46	91.98 61	97.19 21	85.43 38	99.56 3	92.06 38	98.79 12	98.44 20

DeepC-MVS_fast	89.43 2	94.04 27	93.79 30	94.80 23	97.48 42	86.78 19	95.65 54	96.89 43	89.40 38	92.81 40	96.97 28	85.37 39	99.24 32	90.87 59	98.69 22	98.38 23
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 16	94.27 17	94.92 12	98.65 1	86.67 24	96.92 14	97.23 22	88.60 58	93.58 28	97.27 14	85.22 40	99.54 11	92.21 32	98.74 19	98.56 11

CP-MVS		94.34 19	94.21 20	94.74 27	98.39 16	86.64 26	97.60 1	97.24 20	88.53 60	92.73 44	97.23 17	85.20 41	99.32 28	92.15 35	98.83 11	98.25 35

test12				94.34 41	97.13 54	86.15 41		96.29 82		91.04 76		85.08 42	99.01 56		98.13 47	97.86 60

ACMMPR		94.43 16	94.28 16	94.91 13	98.63 2	86.69 22	96.94 10	97.32 16	88.63 56	93.53 31	97.26 16	85.04 43	99.54 11	92.35 30	98.78 14	98.50 12

XVS		94.45 14	94.32 14	94.85 17	98.54 7	86.60 27	96.93 12	97.19 23	90.66 22	92.85 37	97.16 24	85.02 44	99.49 17	91.99 39	98.56 36	98.47 15

X-MVStestdata		88.31 137	86.13 184	94.85 17	98.54 7	86.60 27	96.93 12	97.19 23	90.66 22	92.85 37	23.41 352	85.02 44	99.49 17	91.99 39	98.56 36	98.47 15

MSLP-MVS++		93.72 34	94.08 24	92.65 83	97.31 47	83.43 92	95.79 45	97.33 14	90.03 27	93.58 28	96.96 29	84.87 46	97.76 140	92.19 34	98.66 29	96.76 95

HPM-MVS		94.02 28	93.88 28	94.43 38	98.39 16	85.78 50	97.25 5	97.07 31	86.90 101	92.62 48	96.80 36	84.85 47	99.17 36	92.43 27	98.65 31	98.33 25
Chunlin Ren, Qingshan Xu, Shikun Zhang, Jiaqi Yang: Hierarchical Prior Mining for Non-local Multi-View Stereo. ICCV 2023
PGM-MVS		93.96 30	93.72 33	94.68 28	98.43 13	86.22 40	95.30 62	97.78 1	87.45 85	93.26 32	97.33 12	84.62 48	99.51 15	90.75 61	98.57 35	98.32 26

EI-MVSNet-Vis-set		93.01 52	92.92 47	93.29 60	95.01 124	83.51 91	94.48 120	95.77 118	90.87 16	92.52 50	96.67 41	84.50 49	99.00 59	91.99 39	94.44 108	97.36 75

zzz-MVS		94.47 13	94.30 15	95.00 10	98.42 14	86.95 13	95.06 83	96.97 35	91.07 14	93.14 35	97.56 7	84.30 50	99.56 3	93.43 14	98.75 17	98.47 15

MTAPA		94.42 18	94.22 18	95.00 10	98.42 14	86.95 13	94.36 138	96.97 35	91.07 14	93.14 35	97.56 7	84.30 50	99.56 3	93.43 14	98.75 17	98.47 15

EI-MVSNet-UG-set		92.74 55	92.62 51	93.12 66	94.86 132	83.20 97	94.40 128	95.74 121	90.71 21	92.05 60	96.60 45	84.00 52	98.99 60	91.55 49	93.63 117	97.17 82

mPP-MVS		93.99 29	93.78 31	94.63 30	98.50 9	85.90 48	96.87 16	96.91 42	88.70 54	91.83 65	97.17 23	83.96 53	99.55 8	91.44 52	98.64 32	98.43 21

APD-MVS_3200maxsize		93.78 33	93.77 32	93.80 55	97.92 32	84.19 76	96.30 26	96.87 46	86.96 97	93.92 22	97.47 9	83.88 54	98.96 65	92.71 25	97.87 53	98.26 34

EPP-MVSNet		91.70 65	91.56 60	92.13 104	95.88 92	80.50 163	97.33 3	95.25 161	86.15 113	89.76 87	95.60 81	83.42 55	98.32 101	87.37 93	93.25 127	97.56 71

UA-Net		92.83 53	92.54 53	93.68 57	96.10 84	84.71 60	95.66 52	96.39 78	91.92 4	93.22 33	96.49 50	83.16 56	98.87 68	84.47 122	95.47 89	97.45 74

UniMVSNet_NR-MVSNet		89.92 97	89.29 96	91.81 119	93.39 182	83.72 84	94.43 126	97.12 27	89.80 31	86.46 137	93.32 143	83.16 56	97.23 194	84.92 115	81.02 275	94.49 184

1121		90.42 86	89.49 90	93.20 63	97.27 51	84.46 68	92.63 227	95.51 139	71.01 318	91.20 74	96.21 59	82.92 58	99.05 47	80.56 178	98.07 49	96.10 113

新几何1				93.10 67	97.30 48	84.35 74		95.56 132	71.09 317	91.26 73	96.24 57	82.87 59	98.86 71	79.19 206	98.10 48	96.07 115

原ACMM1				92.01 105	97.34 46	81.05 147		96.81 49	78.89 251	90.45 80	95.92 71	82.65 60	98.84 76	80.68 176	98.26 44	96.14 109

DeepC-MVS	88.79 3	93.31 43	92.99 45	94.26 43	96.07 86	85.83 49	94.89 93	96.99 33	89.02 48	89.56 88	97.37 11	82.51 61	99.38 23	92.20 33	98.30 42	97.57 70
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 42	93.22 40	93.94 49	98.36 18	84.83 58	97.15 7	96.80 50	85.77 119	92.47 52	97.13 25	82.38 62	99.07 45	90.51 63	98.40 40	97.92 58

canonicalmvs		93.27 45	92.75 50	94.85 17	95.70 98	87.66 6	96.33 25	96.41 76	90.00 28	94.09 18	94.60 107	82.33 63	98.62 85	92.40 29	92.86 135	98.27 32

DP-MVS Recon		91.95 60	91.28 63	93.96 48	98.33 19	85.92 45	94.66 112	96.66 63	82.69 198	90.03 86	95.82 75	82.30 64	99.03 51	84.57 121	96.48 78	96.91 91

PAPR		90.02 92	89.27 98	92.29 98	95.78 95	80.95 151	92.68 226	96.22 87	81.91 212	86.66 135	93.75 138	82.23 65	98.44 95	79.40 205	94.79 97	97.48 73

MVS_Test		91.31 70	91.11 65	91.93 111	94.37 150	80.14 167	93.46 197	95.80 116	86.46 107	91.35 72	93.77 136	82.21 66	98.09 123	87.57 89	94.95 96	97.55 72

nrg030		91.08 75	90.39 75	93.17 65	93.07 191	86.91 15	96.41 24	96.26 83	88.30 64	88.37 101	94.85 100	82.19 67	97.64 147	91.09 54	82.95 247	94.96 151

UniMVSNet (Re)		89.80 99	89.07 101	92.01 105	93.60 178	84.52 64	94.78 101	97.47 5	89.26 41	86.44 140	92.32 180	82.10 68	97.39 180	84.81 118	80.84 279	94.12 196

testdata				90.49 162	96.40 68	77.89 243		95.37 154	72.51 307	93.63 26	96.69 39	82.08 69	97.65 145	83.08 138	97.39 61	95.94 119

PAPM_NR		91.22 72	90.78 73	92.52 88	97.60 36	81.46 135	94.37 134	96.24 86	86.39 109	87.41 121	94.80 102	82.06 70	98.48 92	82.80 144	95.37 91	97.61 68

MG-MVS		91.77 62	91.70 59	92.00 107	97.08 55	80.03 172	93.60 192	95.18 168	87.85 76	90.89 77	96.47 51	82.06 70	98.36 96	85.07 113	97.04 66	97.62 67

CANet		93.54 38	93.20 41	94.55 33	95.65 99	85.73 51	94.94 90	96.69 61	91.89 5	90.69 78	95.88 73	81.99 72	99.54 11	93.14 21	97.95 52	98.39 22

FC-MVSNet-test		90.27 88	90.18 80	90.53 154	93.71 175	79.85 177	95.77 46	97.59 2	89.31 40	86.27 143	94.67 104	81.93 73	97.01 209	84.26 127	88.09 204	94.71 166

FIs		90.51 85	90.35 76	90.99 145	93.99 165	80.98 149	95.73 47	97.54 3	89.15 44	86.72 134	94.68 103	81.83 74	97.24 192	85.18 112	88.31 201	94.76 165

ACMMP		93.24 48	92.88 49	94.30 42	98.09 28	85.33 54	96.86 17	97.45 6	88.33 63	90.15 84	97.03 27	81.44 75	99.51 15	90.85 60	95.74 84	98.04 49
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 67	91.11 65	93.01 72	94.35 153	83.39 94	94.60 114	95.10 170	87.10 91	90.57 79	93.10 154	81.43 76	98.07 125	89.29 70	94.48 105	97.59 69

MVS_111021_LR		92.47 56	92.29 56	92.98 73	95.99 89	84.43 72	93.08 213	96.09 95	88.20 69	91.12 75	95.72 79	81.33 77	97.76 140	91.74 47	97.37 62	96.75 96

mvs_anonymous		89.37 114	89.32 95	89.51 214	93.47 180	74.22 278	91.65 255	94.83 186	82.91 192	85.45 170	93.79 135	81.23 78	96.36 248	86.47 107	94.09 111	97.94 54

PVSNet_BlendedMVS		89.98 93	89.70 87	90.82 148	96.12 79	81.25 140	93.92 171	96.83 47	83.49 169	89.10 93	92.26 185	81.04 79	98.85 74	86.72 104	87.86 206	92.35 275

PVSNet_Blended		90.73 79	90.32 77	91.98 108	96.12 79	81.25 140	92.55 231	96.83 47	82.04 208	89.10 93	92.56 173	81.04 79	98.85 74	86.72 104	95.91 82	95.84 124

alignmvs		93.08 51	92.50 54	94.81 22	95.62 101	87.61 7	95.99 36	96.07 97	89.77 32	94.12 17	94.87 97	80.56 81	98.66 81	92.42 28	93.10 130	98.15 40

abl_6		93.18 50	93.05 43	93.57 59	97.52 39	84.27 75	95.53 57	96.67 62	87.85 76	93.20 34	97.22 18	80.35 82	99.18 35	91.91 43	97.21 63	97.26 76

API-MVS		90.66 80	90.07 82	92.45 91	96.36 70	84.57 63	96.06 34	95.22 167	82.39 200	89.13 92	94.27 117	80.32 83	98.46 93	80.16 187	96.71 71	94.33 189

PVSNet_Blended_VisFu		91.38 69	90.91 70	92.80 79	96.39 69	83.17 98	94.87 96	96.66 63	83.29 175	89.27 91	94.46 109	80.29 84	99.17 36	87.57 89	95.37 91	96.05 117

test222					96.55 66	81.70 129	92.22 241	95.01 173	68.36 324	90.20 83	96.14 65	80.26 85			97.80 55	96.05 117

Test By Simon												80.02 86

IterMVS-LS		88.36 136	87.91 131	89.70 207	93.80 172	78.29 233	93.73 183	95.08 172	85.73 120	84.75 195	91.90 200	79.88 87	96.92 216	83.83 133	82.51 251	93.89 207
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
EI-MVSNet		89.10 118	88.86 108	89.80 202	91.84 215	78.30 232	93.70 187	95.01 173	85.73 120	87.15 125	95.28 86	79.87 88	97.21 196	83.81 134	87.36 210	93.88 209

TAPA-MVS	84.62 6	88.16 141	87.01 152	91.62 123	96.64 62	80.65 157	94.39 130	96.21 90	76.38 274	86.19 145	95.44 83	79.75 89	98.08 124	62.75 317	95.29 93	96.13 110
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
Fast-Effi-MVS+		89.41 111	88.64 110	91.71 121	94.74 134	80.81 155	93.54 193	95.10 170	83.11 178	86.82 133	90.67 244	79.74 90	97.75 143	80.51 180	93.55 118	96.57 100

pcd_1.5k_mvsjas		6.64 335	8.86 336	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	0.00 358	79.70 91	0.00 359	0.00 356	0.00 357	0.00 357

PS-MVSNAJss		89.97 94	89.62 88	91.02 143	91.90 213	80.85 154	95.26 71	95.98 102	86.26 111	86.21 144	94.29 114	79.70 91	97.65 145	88.87 74	88.10 202	94.57 177

PS-MVSNAJ		91.18 73	90.92 69	91.96 109	95.26 114	82.60 119	92.09 246	95.70 123	86.27 110	91.84 63	92.46 174	79.70 91	98.99 60	89.08 71	95.86 83	94.29 190

xiu_mvs_v2_base		91.13 74	90.89 71	91.86 114	94.97 127	82.42 120	92.24 240	95.64 129	86.11 115	91.74 68	93.14 152	79.67 94	98.89 67	89.06 72	95.46 90	94.28 191

WR-MVS_H		87.80 156	87.37 138	89.10 231	93.23 187	78.12 237	95.61 55	97.30 18	87.90 74	83.72 219	92.01 196	79.65 95	96.01 260	76.36 230	80.54 283	93.16 250

EPNet		91.79 61	91.02 68	94.10 46	90.10 290	85.25 55	96.03 35	92.05 251	92.83 1	87.39 123	95.78 76	79.39 96	99.01 56	88.13 82	97.48 60	98.05 48
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
NR-MVSNet		88.58 132	87.47 136	91.93 111	93.04 193	84.16 77	94.77 102	96.25 85	89.05 45	80.04 269	93.29 146	79.02 97	97.05 207	81.71 163	80.05 289	94.59 175

TAMVS		89.21 116	88.29 122	91.96 109	93.71 175	82.62 118	93.30 203	94.19 203	82.22 203	87.78 117	93.94 127	78.83 98	96.95 214	77.70 219	92.98 132	96.32 104

1112_ss		88.42 133	87.33 139	91.72 120	94.92 129	80.98 149	92.97 219	94.54 193	78.16 264	83.82 217	93.88 132	78.78 99	97.91 135	79.45 201	89.41 175	96.26 106

CDS-MVSNet		89.45 108	88.51 112	92.29 98	93.62 177	83.61 89	93.01 216	94.68 190	81.95 210	87.82 116	93.24 148	78.69 100	96.99 210	80.34 183	93.23 128	96.28 105
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
WTY-MVS		89.60 102	88.92 105	91.67 122	95.47 106	81.15 145	92.38 236	94.78 188	83.11 178	89.06 95	94.32 112	78.67 101	96.61 235	81.57 164	90.89 157	97.24 77

CPTT-MVS		91.99 59	91.80 58	92.55 86	98.24 23	81.98 127	96.76 19	96.49 72	81.89 214	90.24 82	96.44 52	78.59 102	98.61 86	89.68 67	97.85 54	97.06 87

IS-MVSNet		91.43 68	91.09 67	92.46 90	95.87 94	81.38 138	96.95 9	93.69 223	89.72 34	89.50 90	95.98 69	78.57 103	97.77 139	83.02 140	96.50 77	98.22 36

OMC-MVS		91.23 71	90.62 74	93.08 68	96.27 72	84.07 78	93.52 194	95.93 105	86.95 98	89.51 89	96.13 66	78.50 104	98.35 98	85.84 108	92.90 134	96.83 94

PCF-MVS	84.11 10	87.74 158	86.08 187	92.70 82	94.02 160	84.43 72	89.27 284	95.87 112	73.62 297	84.43 203	94.33 111	78.48 105	98.86 71	70.27 268	94.45 107	94.81 163
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
LCM-MVSNet-Re		88.30 138	88.32 120	88.27 255	94.71 137	72.41 298	93.15 209	90.98 284	87.77 78	79.25 275	91.96 197	78.35 106	95.75 271	83.04 139	95.62 85	96.65 98

HY-MVS	83.01 12	89.03 122	87.94 130	92.29 98	94.86 132	82.77 109	92.08 247	94.49 194	81.52 229	86.93 129	92.79 169	78.32 107	98.23 103	79.93 191	90.55 158	95.88 122

diffmvs		89.07 119	88.32 120	91.34 130	93.24 186	79.79 178	92.29 239	94.98 176	80.24 238	87.38 124	92.45 175	78.02 108	97.33 182	83.29 137	92.93 133	96.91 91

MVS		87.44 177	86.10 186	91.44 128	92.61 203	83.62 88	92.63 227	95.66 126	67.26 328	81.47 249	92.15 187	77.95 109	98.22 104	79.71 197	95.48 88	92.47 270

MVSFormer		91.68 66	91.30 62	92.80 79	93.86 169	83.88 81	95.96 39	95.90 109	84.66 142	91.76 66	94.91 95	77.92 110	97.30 184	89.64 68	97.11 64	97.24 77

lupinMVS		90.92 76	90.21 78	93.03 71	93.86 169	83.88 81	92.81 222	93.86 219	79.84 243	91.76 66	94.29 114	77.92 110	98.04 127	90.48 64	97.11 64	97.17 82

Test_1112_low_res		87.65 160	86.51 176	91.08 139	94.94 128	79.28 203	91.77 249	94.30 201	76.04 279	83.51 225	92.37 178	77.86 112	97.73 144	78.69 210	89.13 188	96.22 107

VNet		92.24 58	91.91 57	93.24 62	96.59 64	83.43 92	94.84 97	96.44 73	89.19 43	94.08 19	95.90 72	77.85 113	98.17 106	88.90 73	93.38 124	98.13 42

DU-MVS		89.34 115	88.50 113	91.85 115	93.04 193	83.72 84	94.47 123	96.59 68	89.50 36	86.46 137	93.29 146	77.25 114	97.23 194	84.92 115	81.02 275	94.59 175

Baseline_NR-MVSNet		87.07 189	86.63 174	88.40 252	91.44 227	77.87 244	94.23 143	92.57 241	84.12 154	85.74 155	92.08 192	77.25 114	96.04 257	82.29 153	79.94 292	91.30 293

jason		90.80 77	90.10 81	92.90 76	93.04 193	83.53 90	93.08 213	94.15 205	80.22 239	91.41 71	94.91 95	76.87 116	97.93 134	90.28 65	96.90 67	97.24 77
jason: jason.
PAPM		86.68 197	85.39 202	90.53 154	93.05 192	79.33 202	89.79 277	94.77 189	78.82 253	81.95 245	93.24 148	76.81 117	97.30 184	66.94 295	93.16 129	94.95 158

Vis-MVSNet (Re-imp)		89.59 103	89.44 92	90.03 192	95.74 96	75.85 272	95.61 55	90.80 289	87.66 83	87.83 115	95.40 85	76.79 118	96.46 243	78.37 211	96.73 70	97.80 63

114514_t		89.51 105	88.50 113	92.54 87	98.11 26	81.99 126	95.16 77	96.36 80	70.19 320	85.81 149	95.25 88	76.70 119	98.63 84	82.07 155	96.86 69	97.00 88

PLC	84.53 7	89.06 121	88.03 127	92.15 102	97.27 51	82.69 116	94.29 139	95.44 148	79.71 245	84.01 214	94.18 119	76.68 120	98.75 79	77.28 223	93.41 123	95.02 146
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
TranMVSNet+NR-MVSNet		88.84 126	87.95 129	91.49 126	92.68 202	83.01 104	94.92 92	96.31 81	89.88 30	85.53 164	93.85 134	76.63 121	96.96 213	81.91 159	79.87 294	94.50 182

MAR-MVS		90.30 87	89.37 94	93.07 70	96.61 63	84.48 67	95.68 50	95.67 124	82.36 202	87.85 110	92.85 163	76.63 121	98.80 77	80.01 188	96.68 72	95.91 120
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
v18		84.97 236	83.76 234	88.60 242	91.36 237	79.41 191	93.82 176	94.04 208	83.00 188	76.61 289	86.60 298	76.19 123	95.43 283	80.39 181	71.79 316	90.96 298

v6		87.98 146	87.25 143	90.16 177	91.36 237	79.39 196	94.37 134	95.27 160	84.48 145	85.78 150	91.51 216	76.15 124	97.46 157	84.46 123	81.88 262	93.62 233

v1neww		87.98 146	87.25 143	90.16 177	91.38 234	79.41 191	94.37 134	95.28 157	84.48 145	85.77 151	91.53 214	76.12 125	97.45 159	84.45 124	81.89 260	93.61 234

v7new		87.98 146	87.25 143	90.16 177	91.38 234	79.41 191	94.37 134	95.28 157	84.48 145	85.77 151	91.53 214	76.12 125	97.45 159	84.45 124	81.89 260	93.61 234

WR-MVS		88.38 134	87.67 133	90.52 160	93.30 185	80.18 165	93.26 206	95.96 104	88.57 59	85.47 169	92.81 167	76.12 125	96.91 217	81.24 166	82.29 253	94.47 187

v16		84.96 237	83.74 236	88.62 240	91.40 232	79.48 185	93.83 174	94.04 208	83.03 185	76.54 290	86.59 299	76.11 128	95.42 284	80.33 184	71.80 315	90.95 300

v17		84.93 239	83.70 238	88.62 240	91.36 237	79.48 185	93.83 174	94.03 210	83.04 184	76.51 291	86.57 300	76.05 129	95.42 284	80.31 186	71.65 317	90.96 298

v8		87.50 176	86.71 163	89.89 197	91.37 236	79.40 195	94.50 119	95.38 152	84.81 139	83.60 223	91.33 224	76.05 129	97.42 170	82.84 143	80.51 286	92.84 260

v148		87.04 190	86.32 180	89.21 227	90.94 265	77.26 260	93.71 186	94.43 196	84.84 138	84.36 207	90.80 242	76.04 131	97.05 207	82.12 154	79.60 295	93.31 245

v15		84.79 242	83.53 243	88.57 246	91.30 248	79.41 191	93.70 187	94.01 211	83.06 181	76.27 292	86.42 304	76.03 132	95.38 286	80.01 188	71.00 320	90.92 301

V14		84.79 242	83.52 244	88.57 246	91.32 244	79.43 190	93.72 185	94.01 211	83.06 181	76.22 293	86.43 301	76.01 133	95.37 287	79.96 190	70.99 321	90.91 302

V9		84.77 244	83.50 245	88.58 243	91.33 242	79.46 187	93.75 181	94.00 214	83.07 180	76.07 298	86.43 301	75.97 134	95.37 287	79.91 193	70.93 323	90.91 302

v1		87.85 151	87.10 146	90.11 188	91.21 251	79.24 207	94.09 156	95.24 162	84.44 149	85.70 156	91.31 227	75.96 135	97.45 159	84.18 128	81.73 267	93.64 230

3Dnovator+	87.14 4	92.42 57	91.37 61	95.55 2	95.63 100	88.73 2	97.07 8	96.77 53	90.84 17	84.02 213	96.62 44	75.95 136	99.34 24	87.77 86	97.68 56	98.59 10

v1141		87.84 152	87.09 147	90.11 188	91.23 249	79.25 205	94.08 158	95.24 162	84.44 149	85.69 158	91.31 227	75.91 137	97.44 166	84.17 129	81.74 266	93.63 232

divwei89l23v2f112		87.84 152	87.09 147	90.10 190	91.23 249	79.24 207	94.09 156	95.24 162	84.44 149	85.70 156	91.31 227	75.91 137	97.44 166	84.17 129	81.73 267	93.64 230

v13		84.72 247	83.44 248	88.58 243	91.31 247	79.52 181	93.77 179	94.00 214	83.03 185	75.85 301	86.38 306	75.84 139	95.35 290	79.83 195	70.95 322	90.87 305

v12		84.74 245	83.46 246	88.58 243	91.32 244	79.50 182	93.75 181	94.01 211	83.06 181	75.98 300	86.41 305	75.82 140	95.36 289	79.87 194	70.89 324	90.89 304

BH-untuned		88.60 131	88.13 126	90.01 194	95.24 121	78.50 227	93.29 204	94.15 205	84.75 140	84.46 201	93.40 140	75.76 141	97.40 177	77.59 220	94.52 104	94.12 196

BH-w/o		87.57 174	87.05 151	89.12 229	94.90 131	77.90 242	92.41 234	93.51 225	82.89 193	83.70 220	91.34 223	75.75 142	97.07 205	75.49 237	93.49 120	92.39 273

cdsmvs_eth3d_5k		22.14 330	29.52 331	0.00 346	0.00 360	0.00 361	0.00 352	95.76 119	0.00 356	0.00 357	94.29 114	75.66 143	0.00 359	0.00 356	0.00 357	0.00 357

CNLPA		89.07 119	87.98 128	92.34 96	96.87 58	84.78 59	94.08 158	93.24 228	81.41 231	84.46 201	95.13 92	75.57 144	96.62 233	77.21 224	93.84 115	95.61 133

v11		84.67 250	83.41 249	88.44 251	91.32 244	79.13 210	93.69 190	93.99 216	82.81 194	76.20 294	86.24 308	75.48 145	95.35 290	79.53 199	71.48 319	90.85 306

CHOSEN 1792x2688		88.84 126	87.69 132	92.30 97	96.14 78	81.42 137	90.01 273	95.86 113	74.52 292	87.41 121	93.94 127	75.46 146	98.36 96	80.36 182	95.53 86	97.12 85

MVS_0304		93.25 47	92.62 51	95.14 9	95.72 97	87.58 8	94.71 108	96.59 68	91.78 7	91.46 70	96.18 64	75.45 147	99.55 8	93.53 11	98.19 45	98.28 29

CP-MVSNet		87.63 164	87.26 142	88.74 236	93.12 190	76.59 266	95.29 64	96.58 70	88.43 61	83.49 226	92.98 161	75.28 148	95.83 267	78.97 207	81.15 272	93.79 215

v7		87.75 157	86.96 153	90.12 183	91.20 252	79.50 182	94.28 140	95.46 142	83.45 170	85.75 153	91.56 213	75.13 149	97.43 168	83.60 135	82.18 255	93.42 243

v10		87.25 183	86.38 177	89.85 198	91.19 254	79.50 182	94.48 120	95.45 146	83.79 160	83.62 222	91.19 232	75.13 149	97.42 170	81.94 158	80.60 281	92.63 266

Vis-MVSNet		91.75 63	91.23 64	93.29 60	95.32 111	83.78 83	96.14 31	95.98 102	89.89 29	90.45 80	96.58 46	75.09 151	98.31 102	84.75 119	96.90 67	97.78 65
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
sss		88.93 125	88.26 124	90.94 147	94.05 159	80.78 156	91.71 252	95.38 152	81.55 228	88.63 97	93.91 131	75.04 152	95.47 282	82.47 149	91.61 142	96.57 100

v1144		87.61 172	86.79 160	90.06 191	91.01 260	79.34 199	93.95 170	95.42 151	83.36 174	85.66 160	91.31 227	74.98 153	97.42 170	83.37 136	82.06 256	93.42 243

V42		87.68 159	86.86 155	90.15 181	90.58 279	80.14 167	94.24 142	95.28 157	83.66 162	85.67 159	91.33 224	74.73 154	97.41 175	84.43 126	81.83 263	92.89 258

XVG-OURS-SEG-HR		89.95 95	89.45 91	91.47 127	94.00 164	81.21 143	91.87 248	96.06 99	85.78 118	88.55 98	95.73 78	74.67 155	97.27 188	88.71 75	89.64 173	95.91 120

v2v482		87.84 152	87.06 150	90.17 176	90.99 261	79.23 209	94.00 168	95.13 169	84.87 137	85.53 164	92.07 194	74.45 156	97.45 159	84.71 120	81.75 265	93.85 213

CLD-MVS		89.47 107	88.90 106	91.18 135	94.22 154	82.07 125	92.13 244	96.09 95	87.90 74	85.37 180	92.45 175	74.38 157	97.56 150	87.15 96	90.43 159	93.93 205
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
XXY-MVS		87.65 160	86.85 156	90.03 192	92.14 209	80.60 160	93.76 180	95.23 165	82.94 190	84.60 197	94.02 123	74.27 158	95.49 281	81.04 168	83.68 240	94.01 204

HQP_MVS		90.60 84	90.19 79	91.82 117	94.70 138	82.73 113	95.85 43	96.22 87	90.81 18	86.91 130	94.86 98	74.23 159	98.12 110	88.15 80	89.99 166	94.63 171

plane_prior6					94.52 144	82.75 110						74.23 159

v144192		87.19 187	86.35 178	89.74 203	90.64 278	78.24 235	93.92 171	95.43 149	81.93 211	85.51 166	91.05 239	74.21 161	97.45 159	82.86 142	81.56 269	93.53 238

VPA-MVSNet		89.62 101	88.96 103	91.60 124	93.86 169	82.89 108	95.46 58	97.33 14	87.91 73	88.43 100	93.31 144	74.17 162	97.40 177	87.32 94	82.86 249	94.52 180

ab-mvs		89.41 111	88.35 117	92.60 84	95.15 122	82.65 117	92.20 242	95.60 130	83.97 155	88.55 98	93.70 139	74.16 163	98.21 105	82.46 150	89.37 176	96.94 90

1314		87.51 175	86.57 175	90.34 173	92.42 205	79.74 180	92.63 227	95.35 156	78.35 260	80.14 267	91.62 209	74.05 164	97.15 198	81.05 167	93.53 119	94.12 196

test_djsdf		89.03 122	88.64 110	90.21 175	90.74 274	79.28 203	95.96 39	95.90 109	84.66 142	85.33 182	92.94 162	74.02 165	97.30 184	89.64 68	88.53 194	94.05 201

AdaColmap		89.89 98	89.07 101	92.37 95	97.41 43	83.03 102	94.42 127	95.92 106	82.81 194	86.34 142	94.65 105	73.89 166	99.02 54	80.69 175	95.51 87	95.05 145

HyFIR lowres test		88.09 144	86.81 158	91.93 111	96.00 88	80.63 158	90.01 273	95.79 117	73.42 298	87.68 119	92.10 191	73.86 167	97.96 131	80.75 174	91.70 141	97.19 81

HQP2-MVS												73.83 168

HQP-MVS		89.80 99	89.28 97	91.34 130	94.17 155	81.56 130	94.39 130	96.04 100	88.81 50	85.43 173	93.97 126	73.83 168	97.96 131	87.11 98	89.77 171	94.50 182

3Dnovator	86.66 5	91.73 64	90.82 72	94.44 36	94.59 142	86.37 33	97.18 6	97.02 32	89.20 42	84.31 209	96.66 42	73.74 170	99.17 36	86.74 101	97.96 51	97.79 64

EPNet_dtu		86.49 202	85.94 191	88.14 260	90.24 288	72.82 290	94.11 154	92.20 247	86.66 105	79.42 274	92.36 179	73.52 171	95.81 269	71.26 263	93.66 116	95.80 127
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
TransMVSNet (Re)		84.43 252	83.06 253	88.54 248	91.72 219	78.44 228	95.18 75	92.82 235	82.73 196	79.67 271	92.12 188	73.49 172	95.96 262	71.10 267	68.73 331	91.21 294

Effi-MVS+-dtu		88.65 130	88.35 117	89.54 211	93.33 183	76.39 267	94.47 123	94.36 198	87.70 79	85.43 173	89.56 266	73.45 173	97.26 190	85.57 110	91.28 144	94.97 148

mvs-test1		89.45 108	89.14 99	90.38 170	93.33 183	77.63 252	94.95 89	94.36 198	87.70 79	87.10 127	92.81 167	73.45 173	98.03 128	85.57 110	93.04 131	95.48 135

PEN-MVS		86.80 193	86.27 182	88.40 252	92.32 207	75.71 273	95.18 75	96.38 79	87.97 71	82.82 233	93.15 151	73.39 175	95.92 263	76.15 234	79.03 297	93.59 236

v1192		87.25 183	86.33 179	90.00 195	90.76 273	79.04 211	93.80 177	95.48 141	82.57 199	85.48 168	91.18 233	73.38 176	97.42 170	82.30 152	82.06 256	93.53 238

QAPM		89.51 105	88.15 125	93.59 58	94.92 129	84.58 62	96.82 18	96.70 59	78.43 259	83.41 227	96.19 63	73.18 177	99.30 30	77.11 226	96.54 76	96.89 93

tpmrst		85.35 228	84.99 207	86.43 290	90.88 270	67.88 321	88.71 291	91.43 272	80.13 240	86.08 147	88.80 273	73.05 178	96.02 259	82.48 148	83.40 246	95.40 138

PS-CasMVS		87.32 180	86.88 154	88.63 239	92.99 196	76.33 269	95.33 59	96.61 67	88.22 68	83.30 229	93.07 155	73.03 179	95.79 270	78.36 212	81.00 277	93.75 221

DTE-MVSNet		86.11 206	85.48 200	87.98 262	91.65 223	74.92 276	94.93 91	95.75 120	87.36 86	82.26 238	93.04 156	72.85 180	95.82 268	74.04 250	77.46 302	93.20 248

MVSTER		88.84 126	88.29 122	90.51 161	92.95 197	80.44 164	93.73 183	95.01 173	84.66 142	87.15 125	93.12 153	72.79 181	97.21 196	87.86 85	87.36 210	93.87 210

pcd1.5k->3k		37.02 328	38.84 329	31.53 341	92.33 206	0.00 361	0.00 352	96.13 93	0.00 356	0.00 357	0.00 358	72.70 182	0.00 359	0.00 356	88.43 198	94.60 174

v1921920		86.97 191	86.06 188	89.69 208	90.53 283	78.11 238	93.80 177	95.43 149	81.90 213	85.33 182	91.05 239	72.66 183	97.41 175	82.05 156	81.80 264	93.53 238

DP-MVS		87.25 183	85.36 203	92.90 76	97.65 35	83.24 96	94.81 99	92.00 253	74.99 287	81.92 246	95.00 94	72.66 183	99.05 47	66.92 297	92.33 139	96.40 102

v7n		86.81 192	85.76 194	89.95 196	90.72 275	79.25 205	95.07 81	95.92 106	84.45 148	82.29 237	90.86 241	72.60 185	97.53 152	79.42 204	80.52 285	93.08 255

v748		86.27 204	85.28 204	89.25 226	90.26 287	77.58 259	94.89 93	95.50 140	84.28 152	81.41 251	90.46 252	72.57 186	97.32 183	79.81 196	78.36 298	92.84 260

OPM-MVS		90.12 90	89.56 89	91.82 117	93.14 189	83.90 80	94.16 151	95.74 121	88.96 49	87.86 109	95.43 84	72.48 187	97.91 135	88.10 83	90.18 165	93.65 229

LS3D		87.89 150	86.32 180	92.59 85	96.07 86	82.92 107	95.23 72	94.92 181	75.66 281	82.89 232	95.98 69	72.48 187	99.21 33	68.43 288	95.23 95	95.64 132

pm-mvs1		86.61 198	85.54 196	89.82 199	91.44 227	80.18 165	95.28 70	94.85 184	83.84 157	81.66 248	92.62 172	72.45 189	96.48 241	79.67 198	78.06 299	92.82 262

PMMVS		85.71 224	84.96 210	87.95 263	88.90 305	77.09 261	88.68 292	90.06 301	72.32 308	86.47 136	90.76 243	72.15 190	94.40 302	81.78 162	93.49 120	92.36 274

PatchmatchNet		85.85 213	84.70 218	89.29 225	91.76 218	75.54 274	88.49 294	91.30 274	81.63 226	85.05 185	88.70 275	71.71 191	96.24 252	74.61 247	89.05 189	96.08 114
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
sam_mvs1												71.70 192				96.12 111

V4		86.50 200	85.54 196	89.39 218	89.13 301	78.99 212	94.73 103	95.54 135	83.59 163	82.10 241	90.61 246	71.60 193	97.45 159	82.52 146	80.01 290	91.74 284

v52		86.50 200	85.53 199	89.39 218	89.17 300	78.99 212	94.72 106	95.54 135	83.59 163	82.10 241	90.60 247	71.59 194	97.45 159	82.52 146	79.99 291	91.73 285

patchmatchnet-post											83.76 317	71.53 195	96.48 241

v1240		86.78 194	85.85 192	89.56 210	90.45 284	77.79 246	93.61 191	95.37 154	81.65 224	85.43 173	91.15 235	71.50 196	97.43 168	81.47 165	82.05 258	93.47 242

anonymousdsp		87.84 152	87.09 147	90.12 183	89.13 301	80.54 161	94.67 111	95.55 133	82.05 206	83.82 217	92.12 188	71.47 197	97.15 198	87.15 96	87.80 207	92.67 264

Patchmatch-test		81.37 279	79.30 283	87.58 269	90.92 267	74.16 280	80.99 334	87.68 330	70.52 319	76.63 288	88.81 272	71.21 198	92.76 319	60.01 325	86.93 216	95.83 125

F-COLMAP		87.95 149	86.80 159	91.40 129	96.35 71	80.88 153	94.73 103	95.45 146	79.65 246	82.04 244	94.61 106	71.13 199	98.50 91	76.24 233	91.05 151	94.80 164

pmmvs4		85.43 226	83.86 233	90.16 177	90.02 293	82.97 106	90.27 268	92.67 239	75.93 280	80.73 258	91.74 204	71.05 200	95.73 272	78.85 208	83.46 244	91.78 283

CR-MVSNet		85.35 228	83.76 234	90.12 183	90.58 279	79.34 199	85.24 317	91.96 257	78.27 261	85.55 162	87.87 289	71.03 201	95.61 273	73.96 252	89.36 177	95.40 138

Patchmtry		82.71 266	80.93 269	88.06 261	90.05 292	76.37 268	84.74 319	91.96 257	72.28 309	81.32 253	87.87 289	71.03 201	95.50 280	68.97 284	80.15 288	92.32 276

PatchFormer-LS_test		86.02 209	85.13 206	88.70 237	91.52 224	74.12 281	91.19 263	92.09 249	82.71 197	84.30 210	87.24 295	70.87 203	96.98 211	81.04 168	85.17 227	95.00 147

RPMNet		83.18 264	80.87 270	90.12 183	90.58 279	79.34 199	85.24 317	90.78 290	71.44 313	85.55 162	82.97 322	70.87 203	95.61 273	61.01 321	89.36 177	95.40 138

Patchmatch-RL test		81.67 273	79.96 277	86.81 288	85.42 322	71.23 304	82.17 332	87.50 332	78.47 258	77.19 287	82.50 323	70.81 205	93.48 311	82.66 145	72.89 312	95.71 130

CostFormer		85.77 218	84.94 211	88.26 256	91.16 257	72.58 297	89.47 282	91.04 283	76.26 277	86.45 139	89.97 259	70.74 206	96.86 220	82.35 151	87.07 215	95.34 141

sam_mvs												70.60 207

xiu_mvs_v1_base_debu		90.64 81	90.05 83	92.40 92	93.97 166	84.46 68	93.32 199	95.46 142	85.17 130	92.25 54	94.03 120	70.59 208	98.57 88	90.97 55	94.67 98	94.18 192

xiu_mvs_v1_base		90.64 81	90.05 83	92.40 92	93.97 166	84.46 68	93.32 199	95.46 142	85.17 130	92.25 54	94.03 120	70.59 208	98.57 88	90.97 55	94.67 98	94.18 192

xiu_mvs_v1_base_debi		90.64 81	90.05 83	92.40 92	93.97 166	84.46 68	93.32 199	95.46 142	85.17 130	92.25 54	94.03 120	70.59 208	98.57 88	90.97 55	94.67 98	94.18 192

test_post											10.29 353	70.57 211	95.91 265

CANet_DTU		90.26 89	89.41 93	92.81 78	93.46 181	83.01 104	93.48 195	94.47 195	89.43 37	87.76 118	94.23 118	70.54 212	99.03 51	84.97 114	96.39 79	96.38 103

BH-RMVSNet		88.37 135	87.48 135	91.02 143	95.28 112	79.45 189	92.89 221	93.07 231	85.45 126	86.91 130	94.84 101	70.35 213	97.76 140	73.97 251	94.59 102	95.85 123

Fast-Effi-MVS+-dtu		87.44 177	86.72 162	89.63 209	92.04 212	77.68 251	94.03 165	93.94 217	85.81 117	82.42 236	91.32 226	70.33 214	97.06 206	80.33 184	90.23 164	94.14 195

MDTV_nov1_ep13_2view						55.91 343	87.62 303		73.32 299	84.59 198		70.33 214		74.65 246		95.50 134

ACMM	84.12 9	89.14 117	88.48 116	91.12 136	94.65 141	81.22 142	95.31 60	96.12 94	85.31 129	85.92 148	94.34 110	70.19 216	98.06 126	85.65 109	88.86 191	94.08 200
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
LPG-MVS_test		89.45 108	88.90 106	91.12 136	94.47 146	81.49 133	95.30 62	96.14 91	86.73 103	85.45 170	95.16 90	69.89 217	98.10 116	87.70 87	89.23 180	93.77 219

LGP-MVS_train				91.12 136	94.47 146	81.49 133		96.14 91	86.73 103	85.45 170	95.16 90	69.89 217	98.10 116	87.70 87	89.23 180	93.77 219

CHOSEN 280x420		85.15 232	83.99 231	88.65 238	92.47 204	78.40 230	79.68 337	92.76 236	74.90 289	81.41 251	89.59 264	69.85 219	95.51 278	79.92 192	95.29 93	92.03 280

LTVRE_ROB	82.13 13	86.26 205	84.90 213	90.34 173	94.44 149	81.50 132	92.31 238	94.89 182	83.03 185	79.63 272	92.67 170	69.69 220	97.79 138	71.20 264	86.26 218	91.72 286
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	88.74 129	87.29 140	93.08 68	92.70 201	85.39 53	96.57 22	96.43 75	78.74 256	80.85 257	96.07 67	69.64 221	99.01 56	78.01 217	96.65 73	94.83 162

MDTV_nov1_ep13			83.56 242		91.69 222	69.93 315	87.75 301	91.54 269	78.60 257	84.86 194	88.90 271	69.54 222	96.03 258	70.25 269	88.93 190

PatchT		82.68 267	81.27 265	86.89 286	90.09 291	70.94 309	84.06 324	90.15 298	74.91 288	85.63 161	83.57 318	69.37 223	94.87 300	65.19 308	88.50 196	94.84 161

VPNet		88.20 140	87.47 136	90.39 168	93.56 179	79.46 187	94.04 164	95.54 135	88.67 55	86.96 128	94.58 108	69.33 224	97.15 198	84.05 131	80.53 284	94.56 178

ACMP	84.23 8	89.01 124	88.35 117	90.99 145	94.73 135	81.27 139	95.07 81	95.89 111	86.48 106	83.67 221	94.30 113	69.33 224	97.99 130	87.10 100	88.55 193	93.72 223
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
test_post1							88.00 298				9.81 354	69.31 226	95.53 276	76.65 229

tpmvs		83.35 263	82.07 260	87.20 281	91.07 259	71.00 308	88.31 296	91.70 261	78.91 250	80.49 263	87.18 296	69.30 227	97.08 204	68.12 292	83.56 242	93.51 241

thres200		87.21 186	86.24 183	90.12 183	95.36 108	78.53 221	93.26 206	92.10 248	86.42 108	88.00 108	91.11 237	69.24 228	98.00 129	69.58 278	91.04 152	93.83 214

tfpn200view9		87.58 173	86.64 172	90.41 167	95.99 89	78.64 216	94.58 115	91.98 255	86.94 99	88.09 102	91.77 202	69.18 229	98.10 116	70.13 272	91.10 145	94.48 185

thres400		87.62 167	86.64 172	90.57 152	95.99 89	78.64 216	94.58 115	91.98 255	86.94 99	88.09 102	91.77 202	69.18 229	98.10 116	70.13 272	91.10 145	94.96 151

tfpnnormal		84.72 247	83.23 251	89.20 228	92.79 200	80.05 170	94.48 120	95.81 115	82.38 201	81.08 255	91.21 231	69.01 231	96.95 214	61.69 319	80.59 282	90.58 311

tfpn111		87.63 164	86.68 166	90.47 164	96.12 79	78.55 218	95.03 84	91.58 264	87.15 87	88.06 105	92.29 182	68.91 232	98.15 109	69.88 277	91.10 145	94.71 166

conf200view11		87.65 160	86.71 163	90.46 166	96.12 79	78.55 218	95.03 84	91.58 264	87.15 87	88.06 105	92.29 182	68.91 232	98.10 116	70.13 272	91.10 145	94.71 166

thres100view900		87.63 164	86.71 163	90.38 170	96.12 79	78.55 218	95.03 84	91.58 264	87.15 87	88.06 105	92.29 182	68.91 232	98.10 116	70.13 272	91.10 145	94.48 185

thres600view7		87.65 160	86.67 167	90.59 151	96.08 85	78.72 214	94.88 95	91.58 264	87.06 96	88.08 104	92.30 181	68.91 232	98.10 116	70.05 276	91.10 145	94.96 151

view600		87.62 167	86.65 168	90.53 154	96.19 74	78.52 222	95.29 64	91.09 276	87.08 92	87.84 111	93.03 157	68.86 236	98.11 112	69.44 279	91.02 153	94.96 151

view800		87.62 167	86.65 168	90.53 154	96.19 74	78.52 222	95.29 64	91.09 276	87.08 92	87.84 111	93.03 157	68.86 236	98.11 112	69.44 279	91.02 153	94.96 151

conf0.05thres1000		87.62 167	86.65 168	90.53 154	96.19 74	78.52 222	95.29 64	91.09 276	87.08 92	87.84 111	93.03 157	68.86 236	98.11 112	69.44 279	91.02 153	94.96 151

tfpn		87.62 167	86.65 168	90.53 154	96.19 74	78.52 222	95.29 64	91.09 276	87.08 92	87.84 111	93.03 157	68.86 236	98.11 112	69.44 279	91.02 153	94.96 151

PatchMatch-RL		86.77 196	85.54 196	90.47 164	95.88 92	82.71 115	90.54 266	92.31 244	79.82 244	84.32 208	91.57 212	68.77 240	96.39 246	73.16 256	93.48 122	92.32 276

XVG-OURS		89.40 113	88.70 109	91.52 125	94.06 158	81.46 135	91.27 261	96.07 97	86.14 114	88.89 96	95.77 77	68.73 241	97.26 190	87.39 92	89.96 168	95.83 125

TR-MVS		86.78 194	85.76 194	89.82 199	94.37 150	78.41 229	92.47 233	92.83 234	81.11 234	86.36 141	92.40 177	68.73 241	97.48 155	73.75 254	89.85 170	93.57 237

tpm		84.73 246	84.02 230	86.87 287	90.33 285	68.90 318	89.06 288	89.94 304	80.85 236	85.75 153	89.86 261	68.54 243	95.97 261	77.76 218	84.05 236	95.75 129

DI_MVS_plusplus_test		88.15 142	86.82 157	92.14 103	90.67 277	81.07 146	93.01 216	94.59 192	83.83 159	77.78 282	90.63 245	68.51 244	98.16 107	88.02 84	94.37 109	97.17 82

test_normal		88.13 143	86.78 161	92.18 101	90.55 282	81.19 144	92.74 224	94.64 191	83.84 157	77.49 285	90.51 251	68.49 245	98.16 107	88.22 79	94.55 103	97.21 80

tfpn_ndepth		86.10 207	84.98 208	89.43 217	95.52 105	78.29 233	94.62 113	89.60 311	81.88 221	85.43 173	90.54 248	68.47 246	96.85 221	68.46 287	90.34 162	93.15 252

FMVSNet3		87.40 179	86.11 185	91.30 132	93.79 174	83.64 87	94.20 150	94.81 187	83.89 156	84.37 204	91.87 201	68.45 247	96.56 236	78.23 214	85.36 224	93.70 224

conf0.01		85.83 215	84.54 221	89.71 205	95.26 114	77.63 252	94.21 144	89.33 313	81.89 214	84.94 188	91.51 216	68.43 248	96.80 222	66.05 300	89.23 180	94.71 166

conf0.002		85.83 215	84.54 221	89.71 205	95.26 114	77.63 252	94.21 144	89.33 313	81.89 214	84.94 188	91.51 216	68.43 248	96.80 222	66.05 300	89.23 180	94.71 166

thresconf0.02		85.75 219	84.54 221	89.38 220	95.26 114	77.63 252	94.21 144	89.33 313	81.89 214	84.94 188	91.51 216	68.43 248	96.80 222	66.05 300	89.23 180	93.70 224

tfpn_n400		85.75 219	84.54 221	89.38 220	95.26 114	77.63 252	94.21 144	89.33 313	81.89 214	84.94 188	91.51 216	68.43 248	96.80 222	66.05 300	89.23 180	93.70 224

tfpnconf		85.75 219	84.54 221	89.38 220	95.26 114	77.63 252	94.21 144	89.33 313	81.89 214	84.94 188	91.51 216	68.43 248	96.80 222	66.05 300	89.23 180	93.70 224

tfpnview11		85.75 219	84.54 221	89.38 220	95.26 114	77.63 252	94.21 144	89.33 313	81.89 214	84.94 188	91.51 216	68.43 248	96.80 222	66.05 300	89.23 180	93.70 224

MVP-Stereo		85.97 211	84.86 214	89.32 224	90.92 267	82.19 123	92.11 245	94.19 203	78.76 255	78.77 277	91.63 208	68.38 254	96.56 236	75.01 244	93.95 112	89.20 318
Qingsong Yan: MVP-Stereo: A Parallel Multi-View Patchmatch Stereo Method with Dilation Matching for Photogrammetric Application.
tfpn1000		86.06 208	84.92 212	89.49 215	95.54 102	77.79 246	94.72 106	89.07 320	82.05 206	85.36 181	91.94 198	68.32 255	96.65 231	67.04 294	90.24 163	94.02 203

tpm cat1		81.96 271	80.27 273	87.01 282	91.09 258	71.02 307	87.38 304	91.53 270	66.25 329	80.17 265	86.35 307	68.22 256	96.15 255	69.16 283	82.29 253	93.86 212

tpm2		84.08 254	82.94 254	87.48 273	91.39 233	71.27 303	89.23 286	90.37 294	71.95 311	84.64 196	89.33 267	67.30 257	96.55 238	75.17 241	87.09 214	94.63 171

test-LLR		85.87 212	85.41 201	87.25 277	90.95 263	71.67 301	89.55 278	89.88 306	83.41 171	84.54 199	87.95 286	67.25 258	95.11 295	81.82 160	93.37 125	94.97 148

test0.0.03 1		82.41 269	81.69 262	84.59 303	88.23 310	72.89 289	90.24 269	87.83 328	83.41 171	79.86 270	89.78 262	67.25 258	88.99 331	65.18 309	83.42 245	91.90 282

CVMVSNet		84.69 249	84.79 216	84.37 305	91.84 215	64.92 329	93.70 187	91.47 271	66.19 330	86.16 146	95.28 86	67.18 260	93.33 313	80.89 173	90.42 160	94.88 160

Patchmatch-test1		85.81 217	84.71 217	89.12 229	92.15 208	76.60 265	91.12 264	91.69 262	83.53 168	85.50 167	88.56 278	66.79 261	95.00 298	72.69 258	90.35 161	95.76 128

IterMVS		84.88 240	83.98 232	87.60 268	91.44 227	76.03 271	90.18 271	92.41 243	83.24 177	81.06 256	90.42 253	66.60 262	94.28 303	79.46 200	80.98 278	92.48 269
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
GBi-Net		87.26 181	85.98 189	91.08 139	94.01 161	83.10 99	95.14 78	94.94 177	83.57 165	84.37 204	91.64 205	66.59 263	96.34 249	78.23 214	85.36 224	93.79 215

test1		87.26 181	85.98 189	91.08 139	94.01 161	83.10 99	95.14 78	94.94 177	83.57 165	84.37 204	91.64 205	66.59 263	96.34 249	78.23 214	85.36 224	93.79 215

FMVSNet2		87.19 187	85.82 193	91.30 132	94.01 161	83.67 86	94.79 100	94.94 177	83.57 165	83.88 215	92.05 195	66.59 263	96.51 239	77.56 221	85.01 228	93.73 222

EPMVS		83.90 257	82.70 258	87.51 270	90.23 289	72.67 293	88.62 293	81.96 343	81.37 232	85.01 186	88.34 281	66.31 266	94.45 301	75.30 240	87.12 213	95.43 137

semantic-postprocess				88.18 259	91.71 220	76.87 264		92.65 240	85.40 127	81.44 250	90.54 248	66.21 267	95.00 298	81.04 168	81.05 273	92.66 265

tpmp4_e23		83.87 258	82.33 259	88.48 249	91.46 226	72.82 290	89.82 276	91.57 268	73.02 303	81.86 247	89.05 269	66.20 268	96.97 212	71.57 262	86.39 217	95.66 131

MDA-MVSNet_test_wron		79.21 295	77.19 295	85.29 298	88.22 311	72.77 292	85.87 312	90.06 301	74.34 293	62.62 336	87.56 292	66.14 269	91.99 323	66.90 298	73.01 310	91.10 297

YYNet1		79.22 294	77.20 294	85.28 299	88.20 312	72.66 294	85.87 312	90.05 303	74.33 294	62.70 335	87.61 291	66.09 270	92.03 322	66.94 295	72.97 311	91.15 295

JIA-IIPM		81.04 282	78.98 288	87.25 277	88.64 306	73.48 286	81.75 333	89.61 310	73.19 300	82.05 243	73.71 337	66.07 271	95.87 266	71.18 266	84.60 231	92.41 272

MSDG		84.86 241	83.09 252	90.14 182	93.80 172	80.05 170	89.18 287	93.09 230	78.89 251	78.19 278	91.91 199	65.86 272	97.27 188	68.47 286	88.45 197	93.11 253

jajsoiax		88.24 139	87.50 134	90.48 163	90.89 269	80.14 167	95.31 60	95.65 128	84.97 136	84.24 211	94.02 123	65.31 273	97.42 170	88.56 76	88.52 195	93.89 207

cascas		86.43 203	84.98 208	90.80 149	92.10 211	80.92 152	90.24 269	95.91 108	73.10 301	83.57 224	88.39 280	65.15 274	97.46 157	84.90 117	91.43 143	94.03 202

ADS-MVSNet2		81.66 274	79.71 280	87.50 271	91.35 240	74.19 279	83.33 328	88.48 324	72.90 304	82.24 239	85.77 310	64.98 275	93.20 315	64.57 311	83.74 238	95.12 143

ADS-MVSNet		81.56 276	79.78 278	86.90 285	91.35 240	71.82 300	83.33 328	89.16 319	72.90 304	82.24 239	85.77 310	64.98 275	93.76 307	64.57 311	83.74 238	95.12 143

pmmvs5		84.21 253	82.84 257	88.34 254	88.95 304	76.94 263	92.41 234	91.91 259	75.63 282	80.28 264	91.18 233	64.59 277	95.57 275	77.09 227	83.47 243	92.53 268

PVSNet	78.82 18	85.55 225	84.65 219	88.23 258	94.72 136	71.93 299	87.12 305	92.75 237	78.80 254	84.95 187	90.53 250	64.43 278	96.71 230	74.74 245	93.86 114	96.06 116

UGNet		89.95 95	88.95 104	92.95 74	94.51 145	83.31 95	95.70 49	95.23 165	89.37 39	87.58 120	93.94 127	64.00 279	98.78 78	83.92 132	96.31 80	96.74 97
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		85.07 233	84.27 227	87.48 273	92.91 198	70.62 311	91.69 254	92.46 242	76.20 278	82.67 235	95.22 89	63.94 280	97.29 187	77.51 222	85.80 221	94.53 179

mvs_tets		88.06 145	87.28 141	90.38 170	90.94 265	79.88 175	95.22 73	95.66 126	85.10 134	84.21 212	93.94 127	63.53 281	97.40 177	88.50 77	88.40 200	93.87 210

dp		81.47 278	80.23 274	85.17 300	89.92 295	65.49 328	86.74 306	90.10 300	76.30 276	81.10 254	87.12 297	62.81 282	95.92 263	68.13 291	79.88 293	94.09 199

LFMVS		90.08 91	89.13 100	92.95 74	96.71 61	82.32 122	96.08 33	89.91 305	86.79 102	92.15 59	96.81 34	62.60 283	98.34 99	87.18 95	93.90 113	98.19 37

DWT-MVSNet_test		84.95 238	83.68 239	88.77 234	91.43 230	73.75 284	91.74 251	90.98 284	80.66 237	83.84 216	87.36 293	62.44 284	97.11 202	78.84 209	85.81 220	95.46 136

Anonymous20231206		81.03 283	79.77 279	84.82 302	87.85 317	70.26 313	91.42 258	92.08 250	73.67 296	77.75 283	89.25 268	62.43 285	93.08 317	61.50 320	82.00 259	91.12 296

VDD-MVS		90.74 78	89.92 86	93.20 63	96.27 72	83.02 103	95.73 47	93.86 219	88.42 62	92.53 49	96.84 32	62.09 286	98.64 83	90.95 58	92.62 137	97.93 57

MS-PatchMatch		85.05 234	84.16 228	87.73 266	91.42 231	78.51 226	91.25 262	93.53 224	77.50 266	80.15 266	91.58 210	61.99 287	95.51 278	75.69 236	94.35 110	89.16 319

OurMVSNet-221017-0		85.35 228	84.64 220	87.49 272	90.77 272	72.59 296	94.01 167	94.40 197	84.72 141	79.62 273	93.17 150	61.91 288	96.72 228	81.99 157	81.16 270	93.16 250

test20.03		79.95 289	79.08 286	82.55 312	85.79 321	67.74 322	91.09 265	91.08 280	81.23 233	74.48 308	89.96 260	61.63 289	90.15 329	60.08 323	76.38 304	89.76 313

DSMNet-mixed		76.94 299	76.29 298	78.89 316	83.10 330	56.11 342	87.78 300	79.77 346	60.65 338	75.64 302	88.71 274	61.56 290	88.34 333	60.07 324	89.29 179	92.21 279

IB-MVS	80.51 15	85.24 231	83.26 250	91.19 134	92.13 210	79.86 176	91.75 250	91.29 275	83.28 176	80.66 260	88.49 279	61.28 291	98.46 93	80.99 171	79.46 296	95.25 142
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		86.61 198	85.27 205	90.66 150	91.33 242	78.71 215	90.40 267	93.81 222	85.34 128	85.12 184	89.57 265	61.25 292	97.11 202	80.99 171	89.59 174	96.15 108

N_pmnet		68.89 312	68.44 313	70.23 328	89.07 303	28.79 357	88.06 297	19.50 358	69.47 322	71.86 320	84.93 313	61.24 293	91.75 325	54.70 329	77.15 303	90.15 312

EU-MVSNet		81.32 280	80.95 268	82.42 313	88.50 308	63.67 330	93.32 199	91.33 273	64.02 334	80.57 262	92.83 165	61.21 294	92.27 321	76.34 231	80.38 287	91.32 292

VDDNet		89.56 104	88.49 115	92.76 81	95.07 123	82.09 124	96.30 26	93.19 229	81.05 235	91.88 62	96.86 31	61.16 295	98.33 100	88.43 78	92.49 138	97.84 61

PVSNet_0	73.20 20	77.22 298	74.83 301	84.37 305	90.70 276	71.10 306	83.09 330	89.67 309	72.81 306	73.93 310	83.13 321	60.79 296	93.70 308	68.54 285	50.84 343	88.30 330

SixPastTwentyTwo		83.91 256	82.90 255	86.92 284	90.99 261	70.67 310	93.48 195	91.99 254	85.54 124	77.62 284	92.11 190	60.59 297	96.87 219	76.05 235	77.75 300	93.20 248

gg-mvs-nofinetune		81.77 272	79.37 282	88.99 232	90.85 271	77.73 250	86.29 309	79.63 347	74.88 290	83.19 230	69.05 340	60.34 298	96.11 256	75.46 238	94.64 101	93.11 253

MDA-MVSNet-bldmvs		78.85 296	76.31 297	86.46 289	89.76 297	73.88 283	88.79 290	90.42 293	79.16 249	59.18 337	88.33 282	60.20 299	94.04 305	62.00 318	68.96 329	91.48 290

pmmvs6		83.42 260	81.60 263	88.87 233	88.01 314	77.87 244	94.96 88	94.24 202	74.67 291	78.80 276	91.09 238	60.17 300	96.49 240	77.06 228	75.40 307	92.23 278

ACMH	80.38 17	85.36 227	83.68 239	90.39 168	94.45 148	80.63 158	94.73 103	94.85 184	82.09 205	77.24 286	92.65 171	60.01 301	97.58 148	72.25 260	84.87 229	92.96 256
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
GG-mvs-BLEND				87.94 264	89.73 298	77.91 241	87.80 299	78.23 349		80.58 261	83.86 316	59.88 302	95.33 292	71.20 264	92.22 140	90.60 310

pmmvs-eth3d		80.97 284	78.72 289	87.74 265	84.99 325	79.97 174	90.11 272	91.65 263	75.36 283	73.51 311	86.03 309	59.45 303	93.96 306	75.17 241	72.21 313	89.29 317

test_0402		81.30 281	79.17 285	87.67 267	93.19 188	78.17 236	92.98 218	91.71 260	75.25 284	76.02 299	90.31 254	59.23 304	96.37 247	50.22 335	83.63 241	88.47 329

FMVSNet1		85.85 213	84.11 229	91.08 139	92.81 199	83.10 99	95.14 78	94.94 177	81.64 225	82.68 234	91.64 205	59.01 305	96.34 249	75.37 239	83.78 237	93.79 215

COLMAP_ROB	80.39 16	83.96 255	82.04 261	89.74 203	95.28 112	79.75 179	94.25 141	92.28 245	75.17 285	78.02 281	93.77 136	58.60 306	97.84 137	65.06 310	85.92 219	91.63 287
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	85.05 234	83.46 246	89.82 199	94.66 140	79.37 197	94.44 125	94.12 207	82.19 204	78.04 280	92.82 166	58.23 307	97.54 151	73.77 253	82.90 248	92.54 267

LP		75.51 302	72.15 306	85.61 296	87.86 316	73.93 282	80.20 336	88.43 325	67.39 325	70.05 322	80.56 330	58.18 308	93.18 316	46.28 341	70.36 326	89.71 315

CMPMVS	59.16 21	80.52 286	79.20 284	84.48 304	83.98 327	67.63 323	89.95 275	93.84 221	64.79 333	66.81 330	91.14 236	57.93 309	95.17 293	76.25 232	88.10 202	90.65 307
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
ITE_SJBPF				88.24 257	91.88 214	77.05 262		92.92 232	85.54 124	80.13 268	93.30 145	57.29 310	96.20 253	72.46 259	84.71 230	91.49 289

TESTMET0.1,1		83.74 259	82.85 256	86.42 291	89.96 294	71.21 305	89.55 278	87.88 327	77.41 267	83.37 228	87.31 294	56.71 311	93.65 309	80.62 177	92.85 136	94.40 188

UnsupCasMVSNet_eth		80.07 288	78.27 290	85.46 297	85.24 323	72.63 295	88.45 295	94.87 183	82.99 189	71.64 321	88.07 285	56.34 312	91.75 325	73.48 255	63.36 338	92.01 281

K. test v3		81.59 275	80.15 276	85.91 294	89.89 296	69.42 317	92.57 230	87.71 329	85.56 123	73.44 312	89.71 263	55.58 313	95.52 277	77.17 225	69.76 327	92.78 263

test-mter		84.54 251	83.64 241	87.25 277	90.95 263	71.67 301	89.55 278	89.88 306	79.17 248	84.54 199	87.95 286	55.56 314	95.11 295	81.82 160	93.37 125	94.97 148

lessismore_v0				86.04 292	88.46 309	68.78 319		80.59 345		73.01 315	90.11 257	55.39 315	96.43 245	75.06 243	65.06 333	92.90 257

MVS-HIRNet		73.70 305	72.20 305	78.18 319	91.81 217	56.42 341	82.94 331	82.58 341	55.24 340	68.88 324	66.48 341	55.32 316	95.13 294	58.12 326	88.42 199	83.01 336

new-patchmatchnet		76.41 300	75.17 300	80.13 315	82.65 333	59.61 335	87.66 302	91.08 280	78.23 263	69.85 323	83.22 320	54.76 317	91.63 327	64.14 313	64.89 334	89.16 319

XVG-ACMP-BASELINE		86.00 210	84.84 215	89.45 216	91.20 252	78.00 239	91.70 253	95.55 133	85.05 135	82.97 231	92.25 186	54.49 318	97.48 155	82.93 141	87.45 209	92.89 258

USDC		82.76 265	81.26 266	87.26 276	91.17 255	74.55 277	89.27 284	93.39 227	78.26 262	75.30 303	92.08 192	54.43 319	96.63 232	71.64 261	85.79 222	90.61 308

AllTest		83.42 260	81.39 264	89.52 212	95.01 124	77.79 246	93.12 210	90.89 287	77.41 267	76.12 296	93.34 141	54.08 320	97.51 153	68.31 289	84.27 234	93.26 246

TestCases				89.52 212	95.01 124	77.79 246		90.89 287	77.41 267	76.12 296	93.34 141	54.08 320	97.51 153	68.31 289	84.27 234	93.26 246

MIMVSNet		82.59 268	80.53 271	88.76 235	91.51 225	78.32 231	86.57 308	90.13 299	79.32 247	80.70 259	88.69 276	52.98 322	93.07 318	66.03 306	88.86 191	94.90 159

FMVSNet5		81.52 277	79.60 281	87.27 275	91.17 255	77.95 240	91.49 257	92.26 246	76.87 272	76.16 295	87.91 288	51.67 323	92.34 320	67.74 293	81.16 270	91.52 288

testgi		80.94 285	80.20 275	83.18 309	87.96 315	66.29 325	91.28 260	90.70 292	83.70 161	78.12 279	92.84 164	51.37 324	90.82 328	63.34 314	82.46 252	92.43 271

Test4		85.75 219	83.72 237	91.83 116	88.08 313	81.03 148	92.48 232	95.54 135	83.38 173	73.40 313	88.57 277	50.99 325	97.37 181	86.61 106	94.47 106	97.09 86

UnsupCasMVSNet_bld		76.23 301	73.27 303	85.09 301	83.79 328	72.92 288	85.65 316	93.47 226	71.52 312	68.84 325	79.08 333	49.77 326	93.21 314	66.81 299	60.52 340	89.13 321

OpenMVS_ROB	74.94 19	79.51 292	77.03 296	86.93 283	87.00 318	76.23 270	92.33 237	90.74 291	68.93 323	74.52 307	88.23 283	49.58 327	96.62 233	57.64 327	84.29 233	87.94 331

testing_2		83.40 262	81.02 267	90.56 153	85.06 324	80.51 162	91.37 259	95.57 131	82.92 191	67.06 329	85.54 312	49.47 328	97.24 192	86.74 101	85.44 223	93.93 205

TDRefinement		79.81 290	77.34 292	87.22 280	79.24 339	75.48 275	93.12 210	92.03 252	76.45 273	75.01 304	91.58 210	49.19 329	96.44 244	70.22 271	69.18 328	89.75 314

MIMVSNet1		79.38 293	77.28 293	85.69 295	86.35 320	73.67 285	91.61 256	92.75 237	78.11 265	72.64 317	88.12 284	48.16 330	91.97 324	60.32 322	77.49 301	91.43 291

LF4IMVS		80.37 287	79.07 287	84.27 307	86.64 319	69.87 316	89.39 283	91.05 282	76.38 274	74.97 305	90.00 258	47.85 331	94.25 304	74.55 248	80.82 280	88.69 324

EG-PatchMatch MVS		82.37 270	80.34 272	88.46 250	90.27 286	79.35 198	92.80 223	94.33 200	77.14 271	73.26 314	90.18 256	47.47 332	96.72 228	70.25 269	87.32 212	89.30 316

testpf		71.41 310	72.11 307	69.30 330	84.53 326	59.79 334	62.74 347	83.14 340	71.11 316	68.83 326	81.57 328	46.70 333	84.83 344	74.51 249	75.86 306	63.30 344

TinyColmap		79.76 291	77.69 291	85.97 293	91.71 220	73.12 287	89.55 278	90.36 295	75.03 286	72.03 319	90.19 255	46.22 334	96.19 254	63.11 315	81.03 274	88.59 325

test2356		74.50 303	73.27 303	78.20 317	80.81 335	59.84 333	83.76 327	88.33 326	71.43 314	72.37 318	81.84 326	45.60 335	86.26 339	50.97 333	84.32 232	88.50 326

tmp_tt		35.64 329	39.24 328	24.84 342	14.87 357	23.90 358	62.71 348	51.51 357	6.58 353	36.66 347	62.08 345	44.37 336	30.34 356	52.40 330	22.00 352	20.27 352

new_pmnet		72.15 308	70.13 309	78.20 317	82.95 332	65.68 326	83.91 325	82.40 342	62.94 336	64.47 333	79.82 332	42.85 337	86.26 339	57.41 328	74.44 309	82.65 337

test1235678		72.22 307	70.31 308	77.93 320	78.04 340	58.04 337	85.76 314	89.80 308	70.15 321	63.43 334	80.20 331	42.24 338	87.24 336	48.68 337	74.50 308	88.50 326

1111		70.54 311	69.71 310	73.04 325	79.30 337	44.83 350	84.23 322	88.96 321	67.33 326	65.42 331	82.28 324	41.11 339	88.11 334	47.12 339	71.60 318	86.19 333

.test1245		57.63 321	61.79 318	45.14 339	79.30 337	44.83 350	84.23 322	88.96 321	67.33 326	65.42 331	82.28 324	41.11 339	88.11 334	47.12 339	0.39 354	2.46 355

testus		74.41 304	73.35 302	77.59 321	82.49 334	57.08 338	86.02 310	90.21 297	72.28 309	72.89 316	84.32 315	37.08 341	86.96 337	52.24 331	82.65 250	88.73 322

pmmvs3		71.81 309	68.71 312	81.11 314	75.86 341	70.42 312	86.74 306	83.66 339	58.95 339	68.64 327	80.89 329	36.93 342	89.52 330	63.10 316	63.59 337	83.39 335

PM-MVS		78.11 297	76.12 299	84.09 308	83.54 329	70.08 314	88.97 289	85.27 337	79.93 242	74.73 306	86.43 301	34.70 343	93.48 311	79.43 203	72.06 314	88.72 323

Anonymous20231211		72.97 306	69.63 311	83.00 311	83.05 331	66.91 324	92.69 225	89.45 312	61.06 337	67.50 328	83.46 319	34.34 344	93.61 310	51.11 332	63.97 336	88.48 328

ambc				83.06 310	79.99 336	63.51 331	77.47 340	92.86 233		74.34 309	84.45 314	28.74 345	95.06 297	73.06 257	68.89 330	90.61 308

test12356		64.99 315	63.78 314	68.61 332	72.69 343	39.14 353	78.46 338	87.61 331	64.91 332	55.77 338	77.48 334	28.10 346	85.59 341	44.69 342	64.35 335	81.12 339

DeepMVS_CX				56.31 337	74.23 342	51.81 346		56.67 356	44.85 344	48.54 343	75.16 335	27.87 347	58.74 354	40.92 345	52.22 342	58.39 348

no-one		61.56 317	56.58 319	76.49 323	67.80 349	62.76 332	78.13 339	86.11 333	63.16 335	43.24 344	64.70 343	26.12 348	88.95 332	50.84 334	29.15 346	77.77 341

testmv		65.49 314	62.66 315	73.96 324	68.78 346	53.14 345	84.70 320	88.56 323	65.94 331	52.35 340	74.65 336	25.02 349	85.14 342	43.54 343	60.40 341	83.60 334

FPMVS		64.63 316	62.55 316	70.88 327	70.80 344	56.71 339	84.42 321	84.42 338	51.78 342	49.57 341	81.61 327	23.49 350	81.48 346	40.61 346	76.25 305	74.46 343

ANet_high		58.88 319	54.22 322	72.86 326	56.50 355	56.67 340	80.75 335	86.00 334	73.09 302	37.39 346	64.63 344	22.17 351	79.49 349	43.51 344	23.96 350	82.43 338

EMVS		42.07 327	41.12 327	44.92 340	63.45 352	35.56 356	73.65 341	63.48 353	33.05 350	26.88 352	45.45 351	21.27 352	67.14 352	19.80 352	23.02 351	32.06 351

Gipuma		57.99 320	54.91 321	67.24 333	88.51 307	65.59 327	52.21 350	90.33 296	43.58 346	42.84 345	51.18 348	20.29 353	85.07 343	34.77 348	70.45 325	51.05 349
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
E-PMN		43.23 326	42.29 326	46.03 338	65.58 350	37.41 354	73.51 342	64.62 352	33.99 349	28.47 351	47.87 349	19.90 354	67.91 351	22.23 351	24.45 349	32.77 350

PMMVS2		59.60 318	56.40 320	69.21 331	68.83 345	46.58 348	73.02 345	77.48 350	55.07 341	49.21 342	72.95 339	17.43 355	80.04 347	49.32 336	44.33 344	80.99 340

LCM-MVSNet		66.00 313	62.16 317	77.51 322	64.51 351	58.29 336	83.87 326	90.90 286	48.17 343	54.69 339	73.31 338	16.83 356	86.75 338	65.47 307	61.67 339	87.48 332

PNet_i23d		50.48 324	47.18 324	60.36 335	68.59 347	44.56 352	72.75 346	72.61 351	43.92 345	33.91 348	60.19 346	6.16 357	73.52 350	38.50 347	28.04 347	63.01 345

PMVS	47.18 22	52.22 322	48.46 323	63.48 334	45.72 356	46.20 349	73.41 343	78.31 348	41.03 347	30.06 349	65.68 342	6.05 358	83.43 345	30.04 349	65.86 332	60.80 346
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
MVE	39.65 23	43.39 325	38.59 330	57.77 336	56.52 354	48.77 347	55.38 349	58.64 355	29.33 351	28.96 350	52.65 347	4.68 359	64.62 353	28.11 350	33.07 345	59.93 347
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
wuyk23d		21.27 331	20.48 332	23.63 343	68.59 347	36.41 355	49.57 351	6.85 359	9.37 352	7.89 354	4.46 357	4.03 360	31.37 355	17.47 353	16.07 353	3.12 353

wuykxyi23d		50.55 323	44.13 325	69.81 329	56.77 353	54.58 344	73.22 344	80.78 344	39.79 348	22.08 353	46.69 350	4.03 360	79.71 348	47.65 338	26.13 348	75.14 342

test123		8.76 333	11.22 334	1.39 344	0.85 359	0.97 359	85.76 314	0.35 361	0.54 355	2.45 356	8.14 356	0.60 362	0.48 357	2.16 355	0.17 356	2.71 354

testmvs		8.92 332	11.52 333	1.12 345	1.06 358	0.46 360	86.02 310	0.65 360	0.62 354	2.74 355	9.52 355	0.31 363	0.45 358	2.38 354	0.39 354	2.46 355

sosnet-low-res		0.00 336	0.00 337	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	0.00 358	0.00 364	0.00 359	0.00 356	0.00 357	0.00 357

sosnet		0.00 336	0.00 337	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	0.00 358	0.00 364	0.00 359	0.00 356	0.00 357	0.00 357

uncertanet		0.00 336	0.00 337	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	0.00 358	0.00 364	0.00 359	0.00 356	0.00 357	0.00 357

Regformer		0.00 336	0.00 337	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	0.00 358	0.00 364	0.00 359	0.00 356	0.00 357	0.00 357

ab-mvs-re		7.82 334	10.43 335	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	93.88 132	0.00 364	0.00 359	0.00 356	0.00 357	0.00 357

uanet		0.00 336	0.00 337	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	0.00 358	0.00 364	0.00 359	0.00 356	0.00 357	0.00 357

GSMVS																96.12 111

test_part3							95.99 36		88.25 66		97.60 5		99.62 1	93.18 19

test_part2					98.55 5	87.22 11				96.40 3

MTGPA								96.97 35

MTMP								60.64 354

gm-plane-assit					89.60 299	68.00 320			77.28 270		88.99 270		97.57 149	79.44 202

test9_res														91.91 43	98.71 20	98.07 46

agg_prior2														90.54 62	98.68 25	98.27 32

agg_prior					97.38 44	85.92 45		96.72 57		92.16 57			98.97 62

test_prior4						85.96 44	94.11 154

test_prior				93.82 52	97.29 49	84.49 65		96.88 44					98.87 68			98.11 44

旧先验2							93.36 198		71.25 315	94.37 14			97.13 201	86.74 101

新几何2							93.11 212

无先验							93.28 205	96.26 83	73.95 295				99.05 47	80.56 178		96.59 99

原ACMM2							92.94 220

testdata2													98.75 79	78.30 213

testdata1							92.15 243		87.94 72

plane_prior7					94.70 138	82.74 112

plane_prior5								96.22 87					98.12 110	88.15 80	89.99 166	94.63 171

plane_prior4											94.86 98

plane_prior3						82.75 110			90.26 25	86.91 130

plane_prior2							95.85 43		90.81 18

plane_prior1					94.59 142

plane_prior						82.73 113	95.21 74		89.66 35						89.88 169

n2								0.00 362

nn								0.00 362

door-mid								85.49 335

test11								96.57 71

door								85.33 336

HQP5-MVS						81.56 130

HQP-NCC					94.17 155		94.39 130		88.81 50	85.43 173

ACMP_Plane					94.17 155		94.39 130		88.81 50	85.43 173

BP-MVS														87.11 98

HQP4-MVS										85.43 173			97.96 131			94.51 181

HQP3-MVS								96.04 100							89.77 171

NP-MVS					94.37 150	82.42 120					93.98 125

ACMMP++_ref															87.47 208

ACMMP++															88.01 205

High-resolution multi-view benchmark