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.

LCM-MVSNet		99.43 1	99.49 1	99.24 1	99.95 1	98.13 1	99.37 1	99.57 1	99.82 1	99.86 1	99.85 1	99.52 1	99.73 1	97.58 1	99.94 1	99.85 1

Anonymous20231211		97.78 3	98.31 2	96.16 47	99.55 2	89.37 81	98.40 5	98.89 4	98.75 2	99.48 3	99.62 2	98.70 2	99.40 36	91.60 106	99.84 5	99.71 3

PEN-MVS		96.69 20	97.39 8	94.61 99	99.16 3	84.50 156	96.54 29	98.05 37	98.06 4	98.64 13	98.25 38	95.01 39	99.65 3	92.95 73	99.83 8	99.68 5

MIMVSNet1		95.52 59	95.45 70	95.72 66	99.14 4	89.02 85	96.23 46	96.87 145	93.73 51	97.87 32	98.49 26	90.73 123	99.05 82	86.43 193	99.60 32	99.10 55

PS-CasMVS		96.69 20	97.43 5	94.49 109	99.13 5	84.09 163	96.61 25	97.97 48	97.91 5	98.64 13	98.13 40	95.24 31	99.65 3	93.39 60	99.84 5	99.72 2

DTE-MVSNet		96.74 18	97.43 5	94.67 97	99.13 5	84.68 155	96.51 30	97.94 54	98.14 3	98.67 12	98.32 35	95.04 36	99.69 2	93.27 64	99.82 10	99.62 11

pmmvs6		96.80 14	97.36 9	95.15 86	99.12 7	87.82 112	96.68 23	97.86 58	96.10 24	98.14 25	99.28 3	97.94 4	98.21 200	91.38 113	99.69 15	99.42 27

HPM-MVS_fast		97.01 7	96.89 17	97.39 18	99.12 7	93.92 24	97.16 12	98.17 26	93.11 62	96.48 76	97.36 79	96.92 7	99.34 49	94.31 33	99.38 64	98.92 83

MP-MVS-pluss		96.08 47	95.92 53	96.57 41	99.06 9	91.21 60	93.25 142	98.32 13	87.89 190	96.86 62	97.38 76	95.55 21	99.39 41	95.47 13	99.47 48	99.11 52
MP-MVS-pluss: MP-MVS-pluss. MP-MVS-pluss
OurMVSNet-221017-0		96.80 14	96.75 20	96.96 32	99.03 10	91.85 52	97.98 6	98.01 43	94.15 44	98.93 4	99.07 5	88.07 169	99.57 13	95.86 11	99.69 15	99.46 25

WR-MVS_H		96.60 25	97.05 15	95.24 82	99.02 11	86.44 131	96.78 22	98.08 32	97.42 7	98.48 18	97.86 55	91.76 97	99.63 6	94.23 37	99.84 5	99.66 7

TDRefinement		97.68 4	97.60 4	97.93 2	99.02 11	95.95 6	98.61 3	98.81 5	97.41 8	97.28 48	98.46 28	94.62 47	98.84 122	94.64 26	99.53 43	98.99 70

CP-MVSNet		96.19 44	96.80 19	94.38 115	98.99 13	83.82 165	96.31 41	97.53 87	97.60 6	98.34 22	97.52 68	91.98 93	99.63 6	93.08 71	99.81 11	99.70 4

PMVS	87.21 14	94.97 82	95.33 77	93.91 129	98.97 14	97.16 2	95.54 65	95.85 193	96.47 18	93.40 183	97.46 72	95.31 28	95.47 302	86.18 196	98.78 125	89.11 336
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
zzz-MVS		96.47 31	96.14 40	97.47 11	98.95 15	94.05 18	93.69 129	97.62 75	94.46 40	96.29 84	96.94 95	93.56 58	99.37 45	94.29 35	99.42 56	98.99 70

MTAPA		96.65 22	96.38 31	97.47 11	98.95 15	94.05 18	95.88 55	97.62 75	94.46 40	96.29 84	96.94 95	93.56 58	99.37 45	94.29 35	99.42 56	98.99 70

ACMMP_Plus		96.21 43	96.12 42	96.49 46	98.90 17	91.42 57	94.57 99	98.03 40	90.42 135	96.37 79	97.35 80	95.68 19	99.25 60	94.44 31	99.34 66	98.80 93

HPM-MVS		96.81 13	96.62 24	97.36 20	98.89 18	93.53 34	97.51 9	98.44 8	92.35 81	95.95 103	96.41 126	96.71 9	99.42 28	93.99 42	99.36 65	99.13 50
Chunlin Ren, Qingshan Xu, Shikun Zhang, Jiaqi Yang: Hierarchical Prior Mining for Non-local Multi-View Stereo. ICCV 2023
VDDNet		94.03 118	94.27 111	93.31 146	98.87 19	82.36 179	95.51 66	91.78 274	97.19 10	96.32 81	98.60 20	84.24 221	98.75 139	87.09 182	98.83 117	98.81 92

TSAR-MVS + MP.		94.96 83	94.75 93	95.57 72	98.86 20	88.69 91	96.37 38	96.81 147	85.23 224	94.75 149	97.12 90	91.85 95	99.40 36	93.45 57	98.33 159	98.62 107

mvs_tets		96.83 10	96.71 21	97.17 25	98.83 21	92.51 43	96.58 27	97.61 78	87.57 196	98.80 8	98.90 9	96.50 11	99.59 12	96.15 9	99.47 48	99.40 31

PS-MVSNAJss		96.01 49	96.04 48	95.89 58	98.82 22	88.51 99	95.57 63	97.88 56	88.72 169	98.81 7	98.86 10	90.77 119	99.60 8	95.43 14	99.53 43	99.57 15

MP-MVS		96.14 45	95.68 63	97.51 10	98.81 23	94.06 16	96.10 47	97.78 67	92.73 69	93.48 180	96.72 111	94.23 52	99.42 28	91.99 96	99.29 73	99.05 63

LTVRE_ROB	93.87 1	97.93 2	98.16 3	97.26 23	98.81 23	93.86 27	99.07 2	98.98 3	97.01 11	98.92 5	98.78 14	95.22 32	98.61 158	96.85 4	99.77 12	99.31 38
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
jajsoiax		96.59 26	96.42 29	97.12 27	98.76 25	92.49 44	96.44 35	97.42 96	86.96 206	98.71 10	98.72 17	95.36 26	99.56 16	95.92 10	99.45 52	99.32 37

pcd1.5k->3k		41.03 329	43.65 331	33.18 342	98.74 26	0.00 361	0.00 352	97.57 82	0.00 356	0.00 357	0.00 358	97.01 6	0.00 359	0.00 356	99.52 45	99.53 17

HSP-MVS		95.18 75	94.49 102	97.23 24	98.67 27	94.05 18	96.41 37	97.00 129	91.26 116	95.12 135	95.15 187	86.60 203	99.50 18	93.43 59	96.81 235	98.13 133

SteuartSystems-ACMMP		96.40 37	96.30 33	96.71 38	98.63 28	91.96 50	95.70 59	98.01 43	93.34 60	96.64 71	96.57 117	94.99 40	99.36 47	93.48 55	99.34 66	98.82 91
Skip Steuart: Steuart Systems R&D Blog.
region2R		96.41 36	96.09 44	97.38 19	98.62 29	93.81 31	96.32 40	97.96 49	92.26 84	95.28 129	96.57 117	95.02 38	99.41 32	93.63 49	99.11 90	98.94 78

mPP-MVS		96.46 32	96.05 47	97.69 5	98.62 29	94.65 9	96.45 33	97.74 68	92.59 75	95.47 122	96.68 113	94.50 50	99.42 28	93.10 69	99.26 75	98.99 70

ACMMP		96.61 24	96.34 32	97.43 15	98.61 31	93.88 25	96.95 17	98.18 25	92.26 84	96.33 80	96.84 104	95.10 35	99.40 36	93.47 56	99.33 69	99.02 67
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
VPNet		93.08 148	93.76 126	91.03 224	98.60 32	75.83 279	91.51 210	95.62 198	91.84 99	95.74 115	97.10 91	89.31 146	98.32 191	85.07 208	99.06 94	98.93 79

ACMMPR		96.46 32	96.14 40	97.41 17	98.60 32	93.82 29	96.30 43	97.96 49	92.35 81	95.57 120	96.61 115	94.93 42	99.41 32	93.78 45	99.15 86	99.00 68

PGM-MVS		96.32 40	95.94 51	97.43 15	98.59 34	93.84 28	95.33 70	98.30 16	91.40 114	95.76 114	96.87 101	95.26 30	99.45 23	92.77 75	99.21 81	99.00 68

XVS		96.49 29	96.18 38	97.44 13	98.56 35	93.99 22	96.50 31	97.95 51	94.58 36	94.38 158	96.49 119	94.56 48	99.39 41	93.57 50	99.05 96	98.93 79

X-MVStestdata		90.70 197	88.45 224	97.44 13	98.56 35	93.99 22	96.50 31	97.95 51	94.58 36	94.38 158	26.89 353	94.56 48	99.39 41	93.57 50	99.05 96	98.93 79

ACMH	88.36 12	96.59 26	97.43 5	94.07 122	98.56 35	85.33 150	96.33 39	98.30 16	94.66 35	98.72 9	98.30 36	97.51 5	98.00 211	94.87 21	99.59 34	98.86 86
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
test_djsdf		96.62 23	96.49 28	97.01 30	98.55 38	91.77 54	97.15 13	97.37 100	88.98 157	98.26 23	98.86 10	93.35 67	99.60 8	96.41 6	99.45 52	99.66 7

v7n		96.82 11	97.31 10	95.33 79	98.54 39	86.81 125	96.83 19	98.07 35	96.59 17	98.46 19	98.43 32	92.91 75	99.52 17	96.25 8	99.76 13	99.65 9

abl_6		97.31 6	97.12 14	97.86 3	98.54 39	95.32 8	96.61 25	98.35 12	95.81 29	97.55 39	97.44 73	96.51 10	99.40 36	94.06 41	99.23 79	98.85 89

ACMH+	88.43 11	96.48 30	96.82 18	95.47 75	98.54 39	89.06 84	95.65 61	98.61 7	96.10 24	98.16 24	97.52 68	96.90 8	98.62 157	90.30 126	99.60 32	98.72 101

SixPastTwentyTwo		94.91 85	95.21 83	93.98 124	98.52 42	83.19 171	95.93 52	94.84 216	94.86 34	98.49 17	98.74 16	81.45 241	99.60 8	94.69 25	99.39 63	99.15 48

HFP-MVS		96.39 38	96.17 39	97.04 28	98.51 43	93.37 35	96.30 43	97.98 45	92.35 81	95.63 118	96.47 121	95.37 24	99.27 58	93.78 45	99.14 87	98.48 112

#test#		95.89 50	95.51 67	97.04 28	98.51 43	93.37 35	95.14 75	97.98 45	89.34 151	95.63 118	96.47 121	95.37 24	99.27 58	91.99 96	99.14 87	98.48 112

Baseline_NR-MVSNet		94.47 106	95.09 88	92.60 178	98.50 45	80.82 196	92.08 182	96.68 154	93.82 50	96.29 84	98.56 22	90.10 137	97.75 237	90.10 134	99.66 23	99.24 42

OPM-MVS		95.61 57	95.45 70	96.08 50	98.49 46	91.00 62	92.65 158	97.33 109	90.05 140	96.77 66	96.85 102	95.04 36	98.56 166	92.77 75	99.06 94	98.70 102

FC-MVSNet-test		95.32 67	95.88 55	93.62 135	98.49 46	81.77 184	95.90 54	98.32 13	93.93 48	97.53 40	97.56 65	88.48 155	99.40 36	92.91 74	99.83 8	99.68 5

XVG-ACMP-BASELINE		95.68 55	95.34 75	96.69 39	98.40 48	93.04 38	94.54 103	98.05 37	90.45 134	96.31 82	96.76 107	92.91 75	98.72 144	91.19 114	99.42 56	98.32 118

ACMM	88.83 9	96.30 42	96.07 46	96.97 31	98.39 49	92.95 41	94.74 90	98.03 40	90.82 125	97.15 52	96.85 102	96.25 14	99.00 92	93.10 69	99.33 69	98.95 77
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
pm-mvs1		95.43 61	95.94 51	93.93 128	98.38 50	85.08 152	95.46 67	97.12 125	91.84 99	97.28 48	98.46 28	95.30 29	97.71 239	90.17 130	99.42 56	98.99 70

COLMAP_ROB	91.06 5	96.75 17	96.62 24	97.13 26	98.38 50	94.31 12	96.79 21	98.32 13	96.69 15	96.86 62	97.56 65	95.48 22	98.77 138	90.11 132	99.44 54	98.31 120
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
TransMVSNet (Re)		95.27 73	96.04 48	92.97 157	98.37 52	81.92 183	95.07 79	96.76 151	93.97 47	97.77 34	98.57 21	95.72 18	97.90 214	88.89 157	99.23 79	99.08 59

LPG-MVS_test		96.38 39	96.23 36	96.84 36	98.36 53	92.13 47	95.33 70	98.25 19	91.78 104	97.07 54	97.22 84	96.38 12	99.28 56	92.07 94	99.59 34	99.11 52

LGP-MVS_train				96.84 36	98.36 53	92.13 47		98.25 19	91.78 104	97.07 54	97.22 84	96.38 12	99.28 56	92.07 94	99.59 34	99.11 52

v748		96.51 28	97.05 15	94.89 91	98.35 55	85.82 144	96.58 27	97.47 93	96.25 21	98.46 19	98.35 33	93.27 68	99.33 52	95.13 19	99.59 34	99.52 20

CP-MVS		96.44 35	96.08 45	97.54 9	98.29 56	94.62 10	96.80 20	98.08 32	92.67 72	95.08 140	96.39 131	94.77 44	99.42 28	93.17 67	99.44 54	98.58 111

FIs		94.90 86	95.35 74	93.55 138	98.28 57	81.76 185	95.33 70	98.14 28	93.05 63	97.07 54	97.18 86	87.65 175	99.29 54	91.72 102	99.69 15	99.61 12

SMA-MVS		95.85 52	95.63 65	96.51 43	98.27 58	91.30 58	95.09 77	97.88 56	86.59 211	97.63 38	97.51 70	94.82 43	99.29 54	93.55 52	99.34 66	98.93 79

TranMVSNet+NR-MVSNet		96.07 48	96.26 35	95.50 74	98.26 59	87.69 113	93.75 127	97.86 58	95.96 28	97.48 42	97.14 88	95.33 27	99.44 24	90.79 116	99.76 13	99.38 32

IS-MVSNet		94.49 105	94.35 106	94.92 90	98.25 60	86.46 130	97.13 15	94.31 230	96.24 22	96.28 87	96.36 136	82.88 228	99.35 48	88.19 168	99.52 45	98.96 76

UA-Net		97.35 5	97.24 13	97.69 5	98.22 61	93.87 26	98.42 4	98.19 24	96.95 12	95.46 124	99.23 4	93.45 60	99.57 13	95.34 17	99.89 4	99.63 10

test_part2					98.21 62	89.41 77				96.72 67

ESAPD		95.42 63	95.34 75	95.68 69	98.21 62	89.41 77	93.92 122	98.14 28	91.83 101	96.72 67	96.39 131	94.69 45	99.44 24	89.00 154	99.10 91	98.17 128

test_0402		95.73 53	96.22 37	94.26 118	98.19 64	85.77 145	93.24 143	97.24 117	96.88 14	97.69 36	97.77 58	94.12 54	99.13 72	91.54 110	99.29 73	97.88 150

ACMP	88.15 13	95.71 54	95.43 73	96.54 42	98.17 65	91.73 55	94.24 110	98.08 32	89.46 149	96.61 73	96.47 121	95.85 17	99.12 74	90.45 118	99.56 41	98.77 96
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
CPTT-MVS		94.74 95	94.12 114	96.60 40	98.15 66	93.01 39	95.84 56	97.66 73	89.21 156	93.28 187	95.46 177	88.89 151	98.98 93	89.80 138	98.82 120	97.80 157

v52		96.93 8	97.29 11	95.86 59	98.12 67	88.48 100	97.69 7	97.74 68	94.90 33	98.55 15	98.72 17	93.39 64	99.49 21	96.92 2	99.62 29	99.61 12

V4		96.93 8	97.29 11	95.86 59	98.11 68	88.47 101	97.69 7	97.74 68	94.91 31	98.55 15	98.72 17	93.37 65	99.49 21	96.92 2	99.62 29	99.61 12

Vis-MVSNet		95.50 60	95.48 68	95.56 73	98.11 68	89.40 79	95.35 69	98.22 23	92.36 79	94.11 166	98.07 41	92.02 90	99.44 24	93.38 61	97.67 207	97.85 153
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
XVG-OURS-SEG-HR		95.38 65	95.00 89	96.51 43	98.10 70	94.07 15	92.46 168	98.13 31	90.69 127	93.75 174	96.25 141	98.03 3	97.02 264	92.08 93	95.55 265	98.45 115

EPP-MVSNet		93.91 120	93.68 132	94.59 104	98.08 71	85.55 148	97.44 10	94.03 235	94.22 43	94.94 144	96.19 149	82.07 236	99.57 13	87.28 181	98.89 107	98.65 103

K. test v3		93.37 137	93.27 144	93.66 134	98.05 72	82.62 177	94.35 107	86.62 305	96.05 26	97.51 41	98.85 12	76.59 274	99.65 3	93.21 66	98.20 176	98.73 100

lessismore_v0				93.87 131	98.05 72	83.77 166		80.32 348		97.13 53	97.91 52	77.49 264	99.11 75	92.62 81	98.08 187	98.74 98

AllTest		94.88 88	94.51 101	96.00 51	98.02 74	92.17 45	95.26 73	98.43 9	90.48 132	95.04 141	96.74 109	92.54 83	97.86 225	85.11 206	98.98 102	97.98 140

TestCases				96.00 51	98.02 74	92.17 45		98.43 9	90.48 132	95.04 141	96.74 109	92.54 83	97.86 225	85.11 206	98.98 102	97.98 140

anonymousdsp		96.74 18	96.42 29	97.68 7	98.00 76	94.03 21	96.97 16	97.61 78	87.68 195	98.45 21	98.77 15	94.20 53	99.50 18	96.70 5	99.40 61	99.53 17

XVG-OURS		94.72 96	94.12 114	96.50 45	98.00 76	94.23 13	91.48 211	98.17 26	90.72 126	95.30 128	96.47 121	87.94 172	96.98 265	91.41 112	97.61 210	98.30 121

114514_t		90.51 199	89.80 209	92.63 176	98.00 76	82.24 180	93.40 134	97.29 113	65.84 340	89.40 270	94.80 204	86.99 192	98.75 139	83.88 218	98.61 135	96.89 200

Gipuma		95.31 69	95.80 59	93.81 133	97.99 79	90.91 64	96.42 36	97.95 51	96.69 15	91.78 221	98.85 12	91.77 96	95.49 301	91.72 102	99.08 93	95.02 262
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
APD-MVS_3200maxsize		96.82 11	96.65 22	97.32 22	97.95 80	93.82 29	96.31 41	98.25 19	95.51 30	96.99 60	97.05 94	95.63 20	99.39 41	93.31 63	98.88 109	98.75 97

HPM-MVS++		95.02 80	94.39 103	96.91 34	97.88 81	93.58 33	94.09 113	96.99 131	91.05 121	92.40 208	95.22 186	91.03 117	99.25 60	92.11 91	98.69 132	97.90 148

EG-PatchMatch MVS		94.54 104	94.67 97	94.14 120	97.87 82	86.50 127	92.00 185	96.74 152	88.16 186	96.93 61	97.61 63	93.04 73	97.90 214	91.60 106	98.12 183	98.03 137

nrg030		96.32 40	96.55 27	95.62 70	97.83 83	88.55 97	95.77 58	98.29 18	92.68 70	98.03 27	97.91 52	95.13 33	98.95 100	93.85 43	99.49 47	99.36 35

UniMVSNet (Re)		95.32 67	95.15 85	95.80 62	97.79 84	88.91 87	92.91 151	98.07 35	93.46 57	96.31 82	95.97 157	90.14 133	99.34 49	92.11 91	99.64 26	99.16 47

VPA-MVSNet		95.14 77	95.67 64	93.58 137	97.76 85	83.15 172	94.58 98	97.58 81	93.39 59	97.05 58	98.04 42	93.25 69	98.51 175	89.75 139	99.59 34	99.08 59

DU-MVS		95.28 71	95.12 87	95.75 65	97.75 86	88.59 95	92.58 159	97.81 63	93.99 45	96.80 64	95.90 158	90.10 137	99.41 32	91.60 106	99.58 39	99.26 40

NR-MVSNet		95.28 71	95.28 80	95.26 81	97.75 86	87.21 119	95.08 78	97.37 100	93.92 49	97.65 37	95.90 158	90.10 137	99.33 52	90.11 132	99.66 23	99.26 40

XXY-MVS		92.58 165	93.16 146	90.84 229	97.75 86	79.84 219	91.87 194	96.22 182	85.94 217	95.53 121	97.68 60	92.69 80	94.48 314	83.21 223	97.51 212	98.21 126

wuykxyi23d		96.76 16	96.57 26	97.34 21	97.75 86	96.73 3	94.37 106	96.48 164	91.00 122	99.72 2	98.99 6	96.06 15	98.21 200	94.86 22	99.90 2	97.09 191

PVSNet_Blended_VisFu		91.63 180	91.20 188	92.94 160	97.73 90	83.95 164	92.14 181	97.46 94	78.85 282	92.35 210	94.98 197	84.16 222	99.08 77	86.36 194	96.77 237	95.79 240

tfpnnormal		94.27 112	94.87 92	92.48 184	97.71 91	80.88 195	94.55 102	95.41 208	93.70 52	96.67 70	97.72 59	91.40 103	98.18 205	87.45 177	99.18 84	98.36 116

HQP_MVS		94.26 113	93.93 117	95.23 83	97.71 91	88.12 106	94.56 100	97.81 63	91.74 108	93.31 184	95.59 169	86.93 194	98.95 100	89.26 149	98.51 143	98.60 109

plane_prior7					97.71 91	88.68 92

UniMVSNet_NR-MVSNet		95.35 66	95.21 83	95.76 64	97.69 94	88.59 95	92.26 177	97.84 61	94.91 31	96.80 64	95.78 166	90.42 129	99.41 32	91.60 106	99.58 39	99.29 39

APDe-MVS		96.46 32	96.64 23	95.93 56	97.68 95	89.38 80	96.90 18	98.41 11	92.52 76	97.43 45	97.92 50	95.11 34	99.50 18	94.45 30	99.30 71	98.92 83

DeepC-MVS	91.39 4	95.43 61	95.33 77	95.71 67	97.67 96	90.17 68	93.86 125	98.02 42	87.35 198	96.22 90	97.99 47	94.48 51	99.05 82	92.73 78	99.68 18	97.93 144
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
Vis-MVSNet (Re-imp)		90.42 202	90.16 205	91.20 222	97.66 97	77.32 262	94.33 108	87.66 298	91.20 118	92.99 196	95.13 189	75.40 276	98.28 193	77.86 275	99.19 82	97.99 139

FMVSNet1		94.84 90	95.13 86	93.97 125	97.60 98	84.29 157	95.99 48	96.56 158	92.38 78	97.03 59	98.53 23	90.12 134	98.98 93	88.78 159	99.16 85	98.65 103

RPSCF		95.58 58	94.89 91	97.62 8	97.58 99	96.30 5	95.97 51	97.53 87	92.42 77	93.41 181	97.78 56	91.21 111	97.77 234	91.06 115	97.06 228	98.80 93

WR-MVS		93.49 131	93.72 129	92.80 168	97.57 100	80.03 212	90.14 249	95.68 197	93.70 52	96.62 72	95.39 183	87.21 186	99.04 85	87.50 176	99.64 26	99.33 36

CSCG		94.69 97	94.75 93	94.52 107	97.55 101	87.87 110	95.01 82	97.57 82	92.68 70	96.20 92	93.44 245	91.92 94	98.78 134	89.11 153	99.24 77	96.92 198

MCST-MVS		92.91 154	92.51 160	94.10 121	97.52 102	85.72 146	91.36 215	97.13 124	80.33 268	92.91 199	94.24 221	91.23 110	98.72 144	89.99 136	97.93 196	97.86 152

F-COLMAP		92.28 172	91.06 191	95.95 53	97.52 102	91.90 51	93.53 131	97.18 120	83.98 238	88.70 283	94.04 229	88.41 158	98.55 172	80.17 252	95.99 257	97.39 180

VDD-MVS		94.37 107	94.37 105	94.40 114	97.49 104	86.07 139	93.97 117	93.28 248	94.49 39	96.24 88	97.78 56	87.99 171	98.79 131	88.92 156	99.14 87	98.34 117

testgi		90.38 204	91.34 185	87.50 293	97.49 104	71.54 316	89.43 270	95.16 211	88.38 179	94.54 155	94.68 209	92.88 77	93.09 328	71.60 317	97.85 200	97.88 150

plane_prior1					97.38 106

APD-MVS		95.00 81	94.69 95	95.93 56	97.38 106	90.88 65	94.59 96	97.81 63	89.22 155	95.46 124	96.17 151	93.42 63	99.34 49	89.30 145	98.87 112	97.56 172
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
ITE_SJBPF				95.95 53	97.34 108	93.36 37		96.55 161	91.93 94	94.82 147	95.39 183	91.99 92	97.08 262	85.53 200	97.96 194	97.41 177

OMC-MVS		94.22 114	93.69 131	95.81 61	97.25 109	91.27 59	92.27 176	97.40 98	87.10 204	94.56 154	95.42 180	93.74 56	98.11 208	86.62 189	98.85 113	98.06 135

v13		95.39 64	96.12 42	93.18 149	97.22 110	80.81 197	95.55 64	97.57 82	93.42 58	98.02 29	98.49 26	89.62 142	99.18 65	95.54 12	99.68 18	99.54 16

plane_prior6					97.21 111	88.23 105						86.93 194

DP-MVS Recon		92.31 171	91.88 169	93.60 136	97.18 112	86.87 124	91.10 221	97.37 100	84.92 232	92.08 217	94.08 228	88.59 154	98.20 202	83.50 220	98.14 180	95.73 242

新几何1				93.17 150	97.16 113	87.29 116		94.43 227	67.95 333	91.29 227	94.94 198	86.97 193	98.23 199	81.06 244	97.75 201	93.98 287

DP-MVS		95.62 56	95.84 57	94.97 89	97.16 113	88.62 94	94.54 103	97.64 74	96.94 13	96.58 74	97.32 81	93.07 72	98.72 144	90.45 118	98.84 114	97.57 170

1121		90.26 209	89.23 211	93.34 144	97.15 115	87.40 115	91.94 188	94.39 228	67.88 334	91.02 239	94.91 199	86.91 196	98.59 162	81.17 242	97.71 204	94.02 286

v12		95.29 70	96.02 50	93.10 151	97.14 116	80.63 198	95.39 68	97.55 86	93.19 61	97.98 30	98.44 30	89.40 145	99.16 66	95.38 16	99.67 21	99.52 20

CHOSEN 1792x2688		87.19 264	85.92 277	91.00 227	97.13 117	79.41 231	84.51 324	95.60 199	64.14 343	90.07 255	94.81 201	78.26 260	97.14 260	73.34 304	95.38 272	96.46 219

HyFIR lowres test		87.19 264	85.51 279	92.24 191	97.12 118	80.51 199	85.03 319	96.06 185	66.11 339	91.66 222	92.98 252	70.12 287	99.14 70	75.29 298	95.23 275	97.07 192

V9		95.17 76	95.89 54	93.02 154	97.04 119	80.42 200	95.22 74	97.53 87	92.92 68	97.90 31	98.35 33	89.15 149	99.14 70	95.21 18	99.65 25	99.50 22

ab-mvs		92.40 169	92.62 157	91.74 204	97.02 120	81.65 186	95.84 56	95.50 206	86.95 207	92.95 198	97.56 65	90.70 125	97.50 246	79.63 258	97.43 219	96.06 233

v11		95.10 78	95.88 55	92.76 169	96.98 121	79.64 226	95.12 76	97.60 80	92.64 73	98.03 27	98.44 30	89.06 150	99.15 68	95.42 15	99.67 21	99.50 22

test222					96.95 122	85.27 151	88.83 285	93.61 242	65.09 342	90.74 243	94.85 200	84.62 220			97.36 222	93.91 288

V14		95.05 79	95.75 61	92.94 160	96.94 123	80.21 203	95.03 81	97.50 91	92.62 74	97.84 33	98.28 37	88.87 152	99.13 72	95.03 20	99.64 26	99.48 24

CDPH-MVS		92.67 162	91.83 170	95.18 85	96.94 123	88.46 102	90.70 231	97.07 126	77.38 290	92.34 212	95.08 191	92.67 81	98.88 110	85.74 198	98.57 137	98.20 127

CNVR-MVS		94.58 102	94.29 108	95.46 76	96.94 123	89.35 82	91.81 203	96.80 148	89.66 147	93.90 172	95.44 179	92.80 79	98.72 144	92.74 77	98.52 142	98.32 118

原ACMM1				92.87 164	96.91 126	84.22 160		97.01 128	76.84 294	89.64 267	94.46 213	88.00 170	98.70 150	81.53 237	98.01 192	95.70 244

ambc				92.98 156	96.88 127	83.01 175	95.92 53	96.38 171		96.41 77	97.48 71	88.26 160	97.80 231	89.96 137	98.93 106	98.12 134

testdata				91.03 224	96.87 128	82.01 181		94.28 231	71.55 317	92.46 206	95.42 180	85.65 214	97.38 254	82.64 228	97.27 224	93.70 295

v15		94.93 84	95.62 66	92.86 165	96.83 129	80.01 216	94.84 88	97.48 92	92.36 79	97.76 35	98.20 39	88.61 153	99.11 75	94.86 22	99.62 29	99.46 25

NP-MVS					96.82 130	87.10 120					93.40 246

3Dnovator+	92.74 2	95.86 51	95.77 60	96.13 49	96.81 131	90.79 67	96.30 43	97.82 62	96.13 23	94.74 150	97.23 83	91.33 105	99.16 66	93.25 65	98.30 164	98.46 114

Test_1112_low_res		87.50 255	86.58 260	90.25 240	96.80 132	77.75 257	87.53 299	96.25 178	69.73 328	86.47 304	93.61 240	75.67 275	97.88 222	79.95 254	93.20 305	95.11 260

testing_2		94.03 118	94.38 104	93.00 155	96.79 133	81.41 190	92.87 153	96.96 133	85.88 219	97.06 57	97.92 50	91.18 115	98.71 149	91.72 102	99.04 99	98.87 85

v17		94.80 92	95.46 69	92.83 166	96.76 134	80.02 214	94.85 86	97.40 98	92.23 86	97.45 44	98.04 42	88.46 157	99.06 80	94.56 27	99.40 61	99.41 28

v16		94.79 94	95.44 72	92.83 166	96.73 135	80.03 212	94.85 86	97.41 97	92.23 86	97.41 47	98.04 42	88.40 159	99.06 80	94.56 27	99.30 71	99.41 28

PAPM_NR		91.03 193	90.81 196	91.68 207	96.73 135	81.10 193	93.72 128	96.35 175	88.19 185	88.77 281	92.12 273	85.09 217	97.25 256	82.40 231	93.90 297	96.68 207

1112_ss		88.42 235	87.41 243	91.45 214	96.69 137	80.99 194	89.72 264	96.72 153	73.37 309	87.00 302	90.69 296	77.38 266	98.20 202	81.38 238	93.72 300	95.15 258

v8		94.65 99	95.29 79	92.74 170	96.65 138	79.77 222	94.59 96	97.17 121	91.86 98	97.47 43	97.93 49	88.16 163	99.08 77	94.32 32	99.47 48	99.38 32

v6		93.59 127	93.93 117	92.56 180	96.65 138	79.77 222	92.50 165	96.40 168	88.55 174	95.94 105	96.23 144	88.13 164	98.87 116	92.46 87	98.50 145	99.06 62

MVS_111021_HR		93.63 126	93.42 140	94.26 118	96.65 138	86.96 123	89.30 275	96.23 180	88.36 180	93.57 178	94.60 210	93.45 60	97.77 234	90.23 128	98.38 152	98.03 137

ANet_high		94.83 91	96.28 34	90.47 233	96.65 138	73.16 306	94.33 108	98.74 6	96.39 20	98.09 26	98.93 8	93.37 65	98.70 150	90.38 121	99.68 18	99.53 17

v1neww		93.58 128	93.92 119	92.56 180	96.64 142	79.77 222	92.50 165	96.41 166	88.55 174	95.93 106	96.24 142	88.08 166	98.87 116	92.45 88	98.50 145	99.05 63

v7new		93.58 128	93.92 119	92.56 180	96.64 142	79.77 222	92.50 165	96.41 166	88.55 174	95.93 106	96.24 142	88.08 166	98.87 116	92.45 88	98.50 145	99.05 63

SD-MVS		95.19 74	95.73 62	93.55 138	96.62 144	88.88 90	94.67 92	98.05 37	91.26 116	97.25 51	96.40 127	95.42 23	94.36 318	92.72 79	99.19 82	97.40 179

v18		94.63 100	95.26 82	92.74 170	96.60 145	79.81 220	94.64 95	97.37 100	91.87 97	97.26 50	97.91 52	88.13 164	99.04 85	94.30 34	99.24 77	99.38 32

PM-MVS		93.33 138	92.67 156	95.33 79	96.58 146	94.06 16	92.26 177	92.18 266	85.92 218	96.22 90	96.61 115	85.64 215	95.99 295	90.35 124	98.23 171	95.93 237

v10		94.68 98	95.27 81	92.90 163	96.57 147	80.15 205	94.65 94	97.57 82	90.68 128	97.43 45	98.00 46	88.18 161	99.15 68	94.84 24	99.55 42	99.41 28

v7		93.66 124	93.97 116	92.73 172	96.55 148	80.15 205	92.54 160	96.99 131	87.36 197	95.99 100	96.48 120	88.18 161	98.94 103	93.35 62	98.31 161	99.09 56

PLC	85.34 15	90.40 203	88.92 219	94.85 92	96.53 149	90.02 69	91.58 208	96.48 164	80.16 269	86.14 306	92.18 271	85.73 212	98.25 198	76.87 285	94.61 287	96.30 225
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
TAPA-MVS	88.58 10	92.49 168	91.75 174	94.73 96	96.50 150	89.69 73	92.91 151	97.68 72	78.02 287	92.79 200	94.10 227	90.85 118	97.96 213	84.76 211	98.16 178	96.54 208
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
view600		88.32 237	87.94 236	89.46 255	96.49 151	73.31 301	93.95 118	84.46 328	93.02 64	94.18 162	92.68 259	63.33 319	98.56 166	75.87 293	97.50 213	96.51 210

view800		88.32 237	87.94 236	89.46 255	96.49 151	73.31 301	93.95 118	84.46 328	93.02 64	94.18 162	92.68 259	63.33 319	98.56 166	75.87 293	97.50 213	96.51 210

conf0.05thres1000		88.32 237	87.94 236	89.46 255	96.49 151	73.31 301	93.95 118	84.46 328	93.02 64	94.18 162	92.68 259	63.33 319	98.56 166	75.87 293	97.50 213	96.51 210

tfpn		88.32 237	87.94 236	89.46 255	96.49 151	73.31 301	93.95 118	84.46 328	93.02 64	94.18 162	92.68 259	63.33 319	98.56 166	75.87 293	97.50 213	96.51 210

NCCC		94.08 117	93.54 137	95.70 68	96.49 151	89.90 72	92.39 171	96.91 141	90.64 129	92.33 213	94.60 210	90.58 128	98.96 98	90.21 129	97.70 205	98.23 124

TAMVS		90.16 211	89.05 215	93.49 143	96.49 151	86.37 133	90.34 242	92.55 262	80.84 266	92.99 196	94.57 212	81.94 239	98.20 202	73.51 303	98.21 174	95.90 238

TEST9					96.45 157	89.46 74	90.60 234	96.92 138	79.09 280	90.49 248	94.39 217	91.31 106	98.88 110

train_agg		92.71 161	91.83 170	95.35 77	96.45 157	89.46 74	90.60 234	96.92 138	79.37 276	90.49 248	94.39 217	91.20 112	98.88 110	88.66 162	98.43 148	97.72 160

agg_prior3		92.56 167	91.62 175	95.35 77	96.39 159	89.45 76	90.61 233	96.82 146	78.82 283	90.03 256	94.14 226	90.72 124	98.88 110	88.66 162	98.43 148	97.72 160

test_8					96.37 160	89.14 83	90.51 238	96.89 142	79.37 276	90.42 250	94.36 219	91.20 112	98.82 124

v1141		93.42 135	93.76 126	92.40 188	96.37 160	79.24 234	91.84 199	96.38 171	88.33 181	95.86 111	96.23 144	87.41 181	98.89 106	92.61 82	98.82 120	99.08 59

divwei89l23v2f112		93.42 135	93.76 126	92.41 186	96.37 160	79.24 234	91.84 199	96.38 171	88.33 181	95.86 111	96.23 144	87.41 181	98.89 106	92.61 82	98.83 117	99.09 56

v1		93.43 133	93.77 125	92.41 186	96.37 160	79.24 234	91.84 199	96.38 171	88.33 181	95.87 110	96.22 147	87.45 179	98.89 106	92.61 82	98.83 117	99.09 56

CLD-MVS		91.82 178	91.41 182	93.04 152	96.37 160	83.65 167	86.82 308	97.29 113	84.65 235	92.27 214	89.67 308	92.20 87	97.85 228	83.95 217	99.47 48	97.62 168
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
HQP-NCC					96.36 165		91.37 212		87.16 201	88.81 277

ACMP_Plane					96.36 165		91.37 212		87.16 201	88.81 277

HQP-MVS		92.09 175	91.49 180	93.88 130	96.36 165	84.89 153	91.37 212	97.31 110	87.16 201	88.81 277	93.40 246	84.76 218	98.60 160	86.55 191	97.73 202	98.14 132

v2v482		93.29 139	93.63 133	92.29 189	96.35 168	78.82 245	91.77 206	96.28 176	88.45 177	95.70 117	96.26 140	86.02 210	98.90 104	93.02 72	98.81 123	99.14 49

MSLP-MVS++		93.25 144	93.88 121	91.37 216	96.34 169	82.81 176	93.11 144	97.74 68	89.37 150	94.08 168	95.29 185	90.40 132	96.35 289	90.35 124	98.25 169	94.96 263

FPMVS		84.50 289	83.28 292	88.16 286	96.32 170	94.49 11	85.76 315	85.47 316	83.09 246	85.20 311	94.26 220	63.79 314	86.58 349	63.72 340	91.88 323	83.40 344

Anonymous20231206		88.77 231	88.29 226	90.20 244	96.31 171	78.81 246	89.56 268	93.49 246	74.26 304	92.38 209	95.58 172	82.21 234	95.43 304	72.07 312	98.75 129	96.34 223

MVP-Stereo		90.07 213	88.92 219	93.54 140	96.31 171	86.49 128	90.93 225	95.59 202	79.80 270	91.48 223	95.59 169	80.79 249	97.39 252	78.57 272	91.19 325	96.76 206
Qingsong Yan: MVP-Stereo: A Parallel Multi-View Patchmatch Stereo Method with Dilation Matching for Photogrammetric Application.
v1144		93.50 130	93.81 122	92.57 179	96.28 173	79.61 228	91.86 198	96.96 133	86.95 207	95.91 109	96.32 137	87.65 175	98.96 98	93.51 53	98.88 109	99.13 50

LFMVS		91.33 190	91.16 190	91.82 202	96.27 174	79.36 232	95.01 82	85.61 315	96.04 27	94.82 147	97.06 93	72.03 282	98.46 182	84.96 209	98.70 131	97.65 166

VNet		92.67 162	92.96 147	91.79 203	96.27 174	80.15 205	91.95 186	94.98 213	92.19 89	94.52 156	96.07 153	87.43 180	97.39 252	84.83 210	98.38 152	97.83 154

IterMVS-LS		93.78 122	94.28 109	92.27 190	96.27 174	79.21 239	91.87 194	96.78 149	91.77 106	96.57 75	97.07 92	87.15 187	98.74 142	91.99 96	99.03 100	98.86 86
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
v148		92.87 156	93.29 141	91.62 208	96.25 177	77.72 258	91.28 216	95.05 212	89.69 146	95.93 106	96.04 154	87.34 183	98.38 187	90.05 135	97.99 193	98.78 95

MVS_111021_LR		93.66 124	93.28 143	94.80 94	96.25 177	90.95 63	90.21 245	95.43 207	87.91 188	93.74 176	94.40 216	92.88 77	96.38 287	90.39 120	98.28 165	97.07 192

agg_prior1		92.60 164	91.76 173	95.10 87	96.20 179	88.89 88	90.37 240	96.88 143	79.67 273	90.21 251	94.41 214	91.30 107	98.78 134	88.46 165	98.37 157	97.64 167

agg_prior					96.20 179	88.89 88		96.88 143		90.21 251			98.78 134

旧先验1					96.20 179	84.17 161		94.82 217			95.57 173	89.57 143			97.89 198	96.32 224

CNLPA		91.72 179	91.20 188	93.26 147	96.17 182	91.02 61	91.14 219	95.55 204	90.16 139	90.87 240	93.56 242	86.31 206	94.40 317	79.92 257	97.12 226	94.37 277

v1192		93.49 131	93.78 124	92.62 177	96.16 183	79.62 227	91.83 202	97.22 119	86.07 215	96.10 98	96.38 134	87.22 185	99.02 89	94.14 40	98.88 109	99.22 43

tfpn111		87.60 253	87.12 250	89.04 268	96.14 184	73.09 307	93.00 146	85.31 318	92.13 90	93.26 189	90.96 289	63.42 315	98.48 179	72.87 308	96.98 232	95.56 248

conf200view11		87.41 256	86.89 254	88.97 269	96.14 184	73.09 307	93.00 146	85.31 318	92.13 90	93.26 189	90.96 289	63.42 315	98.28 193	71.27 320	96.54 246	95.56 248

thres100view900		87.35 258	86.89 254	88.72 274	96.14 184	73.09 307	93.00 146	85.31 318	92.13 90	93.26 189	90.96 289	63.42 315	98.28 193	71.27 320	96.54 246	94.79 266

DeepC-MVS_fast	89.96 7	93.73 123	93.44 139	94.60 103	96.14 184	87.90 109	93.36 135	97.14 122	85.53 223	93.90 172	95.45 178	91.30 107	98.59 162	89.51 142	98.62 134	97.31 185
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
PCF-MVS	84.52 17	89.12 223	87.71 241	93.34 144	96.06 188	85.84 143	86.58 312	97.31 110	68.46 332	93.61 177	93.89 234	87.51 178	98.52 174	67.85 330	98.11 184	95.66 245
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
v144192		93.20 147	93.54 137	92.16 194	96.05 189	78.26 252	91.95 186	97.14 122	84.98 231	95.96 102	96.11 152	87.08 189	99.04 85	93.79 44	98.84 114	99.17 46

thres600view7		87.66 251	87.10 252	89.36 262	96.05 189	73.17 305	92.72 155	85.31 318	91.89 96	93.29 186	90.97 288	63.42 315	98.39 185	73.23 305	96.99 231	96.51 210

MIMVSNet		87.13 266	86.54 262	88.89 271	96.05 189	76.11 274	94.39 105	88.51 289	81.37 262	88.27 289	96.75 108	72.38 280	95.52 300	65.71 337	95.47 269	95.03 261

v1921920		93.26 142	93.61 134	92.19 192	96.04 192	78.31 251	91.88 193	97.24 117	85.17 225	96.19 94	96.19 149	86.76 199	99.05 82	94.18 39	98.84 114	99.22 43

v1240		93.29 139	93.71 130	92.06 197	96.01 193	77.89 256	91.81 203	97.37 100	85.12 227	96.69 69	96.40 127	86.67 200	99.07 79	94.51 29	98.76 127	99.22 43

testmv		88.46 234	88.11 233	89.48 253	96.00 194	76.14 273	86.20 314	93.75 240	84.48 236	93.57 178	95.52 176	80.91 248	95.09 310	63.97 339	98.61 135	97.22 188

conf0.01		86.95 269	86.04 269	89.70 250	95.99 195	75.66 280	93.28 136	82.70 335	88.81 162	91.26 228	88.01 320	58.77 334	97.89 216	78.93 264	96.60 240	95.56 248

conf0.002		86.95 269	86.04 269	89.70 250	95.99 195	75.66 280	93.28 136	82.70 335	88.81 162	91.26 228	88.01 320	58.77 334	97.89 216	78.93 264	96.60 240	95.56 248

thresconf0.02		86.69 274	86.04 269	88.64 277	95.99 195	75.66 280	93.28 136	82.70 335	88.81 162	91.26 228	88.01 320	58.77 334	97.89 216	78.93 264	96.60 240	92.36 316

tfpn_n400		86.69 274	86.04 269	88.64 277	95.99 195	75.66 280	93.28 136	82.70 335	88.81 162	91.26 228	88.01 320	58.77 334	97.89 216	78.93 264	96.60 240	92.36 316

tfpnconf		86.69 274	86.04 269	88.64 277	95.99 195	75.66 280	93.28 136	82.70 335	88.81 162	91.26 228	88.01 320	58.77 334	97.89 216	78.93 264	96.60 240	92.36 316

tfpnview11		86.69 274	86.04 269	88.64 277	95.99 195	75.66 280	93.28 136	82.70 335	88.81 162	91.26 228	88.01 320	58.77 334	97.89 216	78.93 264	96.60 240	92.36 316

BH-untuned		90.68 198	90.90 192	90.05 246	95.98 201	79.57 229	90.04 253	94.94 215	87.91 188	94.07 169	93.00 251	87.76 174	97.78 233	79.19 262	95.17 276	92.80 310

DeepPCF-MVS	90.46 6	94.20 115	93.56 136	96.14 48	95.96 202	92.96 40	89.48 269	97.46 94	85.14 226	96.23 89	95.42 180	93.19 70	98.08 209	90.37 122	98.76 127	97.38 182

test_prior3		93.29 139	92.85 150	94.61 99	95.95 203	87.23 117	90.21 245	97.36 106	89.33 152	90.77 241	94.81 201	90.41 130	98.68 152	88.21 166	98.55 138	97.93 144

test_prior				94.61 99	95.95 203	87.23 117		97.36 106					98.68 152			97.93 144

test12				94.43 113	95.95 203	86.75 126		96.24 179		89.76 265		89.79 141	98.79 131		97.95 195	97.75 159

LCM-MVSNet-Re		94.20 115	94.58 99	93.04 152	95.91 206	83.13 173	93.79 126	99.19 2	92.00 93	98.84 6	98.04 42	93.64 57	99.02 89	81.28 239	98.54 140	96.96 196

PatchMatch-RL		89.18 221	88.02 235	92.64 175	95.90 207	92.87 42	88.67 288	91.06 278	80.34 267	90.03 256	91.67 279	83.34 224	94.42 316	76.35 289	94.84 282	90.64 332

TSAR-MVS + GP.		93.07 150	92.41 162	95.06 88	95.82 208	90.87 66	90.97 223	92.61 261	88.04 187	94.61 153	93.79 237	88.08 166	97.81 230	89.41 144	98.39 151	96.50 217

QAPM		92.88 155	92.77 152	93.22 148	95.82 208	83.31 169	96.45 33	97.35 108	83.91 239	93.75 174	96.77 105	89.25 147	98.88 110	84.56 213	97.02 230	97.49 174

tfpn200view9		87.05 267	86.52 263	88.67 275	95.77 210	72.94 310	91.89 191	86.00 310	90.84 123	92.61 203	89.80 303	63.93 312	98.28 193	71.27 320	96.54 246	94.79 266

thres400		87.20 263	86.52 263	89.24 266	95.77 210	72.94 310	91.89 191	86.00 310	90.84 123	92.61 203	89.80 303	63.93 312	98.28 193	71.27 320	96.54 246	96.51 210

pmmvs-eth3d		91.54 182	90.73 199	93.99 123	95.76 212	87.86 111	90.83 227	93.98 237	78.23 286	94.02 170	96.22 147	82.62 233	96.83 271	86.57 190	98.33 159	97.29 186

jason		89.17 222	88.32 225	91.70 206	95.73 213	80.07 209	88.10 292	93.22 250	71.98 316	90.09 253	92.79 254	78.53 258	98.56 166	87.43 178	97.06 228	96.46 219
jason: jason.
alignmvs		93.26 142	92.85 150	94.50 108	95.70 214	87.45 114	93.45 133	95.76 195	91.58 111	95.25 131	92.42 267	81.96 238	98.72 144	91.61 105	97.87 199	97.33 184

xiu_mvs_v1_base_debu		91.47 185	91.52 177	91.33 217	95.69 215	81.56 187	89.92 257	96.05 186	83.22 243	91.26 228	90.74 293	91.55 100	98.82 124	89.29 146	95.91 258	93.62 297

xiu_mvs_v1_base		91.47 185	91.52 177	91.33 217	95.69 215	81.56 187	89.92 257	96.05 186	83.22 243	91.26 228	90.74 293	91.55 100	98.82 124	89.29 146	95.91 258	93.62 297

xiu_mvs_v1_base_debi		91.47 185	91.52 177	91.33 217	95.69 215	81.56 187	89.92 257	96.05 186	83.22 243	91.26 228	90.74 293	91.55 100	98.82 124	89.29 146	95.91 258	93.62 297

PHI-MVS		94.34 110	93.80 123	95.95 53	95.65 218	91.67 56	94.82 89	97.86 58	87.86 191	93.04 195	94.16 225	91.58 99	98.78 134	90.27 127	98.96 105	97.41 177

LF4IMVS		92.72 160	92.02 167	94.84 93	95.65 218	91.99 49	92.92 150	96.60 157	85.08 229	92.44 207	93.62 239	86.80 198	96.35 289	86.81 184	98.25 169	96.18 229

test20.03		90.80 195	90.85 195	90.63 231	95.63 220	79.24 234	89.81 263	92.87 254	89.90 144	94.39 157	96.40 127	85.77 211	95.27 309	73.86 302	99.05 96	97.39 180

TinyColmap		92.00 177	92.76 153	89.71 249	95.62 221	77.02 266	90.72 230	96.17 184	87.70 194	95.26 130	96.29 138	92.54 83	96.45 283	81.77 234	98.77 126	95.66 245

canonicalmvs		94.59 101	94.69 95	94.30 117	95.60 222	87.03 122	95.59 62	98.24 22	91.56 112	95.21 134	92.04 274	94.95 41	98.66 154	91.45 111	97.57 211	97.20 189

AdaColmap		91.63 180	91.36 184	92.47 185	95.56 223	86.36 134	92.24 179	96.27 177	88.88 161	89.90 261	92.69 258	91.65 98	98.32 191	77.38 282	97.64 208	92.72 312

tfpn1000		86.83 272	86.23 268	88.64 277	95.53 224	75.25 287	93.57 130	82.28 342	89.27 154	91.46 224	89.24 311	57.22 342	97.86 225	80.63 247	96.88 234	92.81 309

UnsupCasMVSNet_bld		88.50 233	88.03 234	89.90 247	95.52 225	78.88 244	87.39 300	94.02 236	79.32 279	93.06 194	94.02 231	80.72 250	94.27 319	75.16 299	93.08 309	96.54 208

3Dnovator	92.54 3	94.80 92	94.90 90	94.47 110	95.47 226	87.06 121	96.63 24	97.28 115	91.82 103	94.34 161	97.41 74	90.60 127	98.65 156	92.47 86	98.11 184	97.70 162

Fast-Effi-MVS+		91.28 191	90.86 194	92.53 183	95.45 227	82.53 178	89.25 278	96.52 162	85.00 230	89.91 260	88.55 315	92.94 74	98.84 122	84.72 212	95.44 270	96.22 228

GBi-Net		93.21 145	92.96 147	93.97 125	95.40 228	84.29 157	95.99 48	96.56 158	88.63 170	95.10 137	98.53 23	81.31 244	98.98 93	86.74 185	98.38 152	98.65 103

test1		93.21 145	92.96 147	93.97 125	95.40 228	84.29 157	95.99 48	96.56 158	88.63 170	95.10 137	98.53 23	81.31 244	98.98 93	86.74 185	98.38 152	98.65 103

FMVSNet2		92.78 158	92.73 155	92.95 159	95.40 228	81.98 182	94.18 112	95.53 205	88.63 170	96.05 99	97.37 77	81.31 244	98.81 129	87.38 180	98.67 133	98.06 135

CDS-MVSNet		89.55 216	88.22 230	93.53 141	95.37 231	86.49 128	89.26 276	93.59 243	79.76 271	91.15 237	92.31 269	77.12 268	98.38 187	77.51 280	97.92 197	95.71 243
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
Test4		91.41 189	91.25 187	91.89 200	95.35 232	80.32 201	90.97 223	96.92 138	81.96 258	95.11 136	93.81 236	81.34 243	98.48 179	88.71 161	97.08 227	96.87 202

V42		93.43 133	93.58 135	92.97 157	95.34 233	81.22 191	92.67 157	96.49 163	87.25 200	96.20 92	96.37 135	87.32 184	98.85 121	92.39 90	98.21 174	98.85 89

DI_MVS_plusplus_test		91.42 188	91.41 182	91.46 213	95.34 233	79.06 241	90.58 236	93.74 241	82.59 252	94.69 152	94.76 205	86.54 204	98.44 184	87.93 172	96.49 251	96.87 202

Patchmatch-RL test		88.81 230	88.52 223	89.69 252	95.33 235	79.94 217	86.22 313	92.71 259	78.46 284	95.80 113	94.18 224	66.25 302	95.33 307	89.22 151	98.53 141	93.78 292

test_normal		91.49 184	91.44 181	91.62 208	95.21 236	79.44 230	90.08 252	93.84 239	82.60 251	94.37 160	94.74 206	86.66 201	98.46 182	88.58 164	96.92 233	96.95 197

BH-RMVSNet		90.47 200	90.44 202	90.56 232	95.21 236	78.65 249	89.15 279	93.94 238	88.21 184	92.74 201	94.22 222	86.38 205	97.88 222	78.67 271	95.39 271	95.14 259

tfpn_ndepth		85.85 281	85.15 282	87.98 287	95.19 238	75.36 286	92.79 154	83.18 334	86.97 205	89.92 259	86.43 332	57.44 341	97.85 228	78.18 273	96.22 254	90.72 331

Effi-MVS+		92.79 157	92.74 154	92.94 160	95.10 239	83.30 170	94.00 115	97.53 87	91.36 115	89.35 271	90.65 298	94.01 55	98.66 154	87.40 179	95.30 273	96.88 201

USDC		89.02 224	89.08 214	88.84 272	95.07 240	74.50 293	88.97 282	96.39 170	73.21 310	93.27 188	96.28 139	82.16 235	96.39 286	77.55 279	98.80 124	95.62 247

WTY-MVS		86.93 271	86.50 265	88.24 285	94.96 241	74.64 289	87.19 303	92.07 271	78.29 285	88.32 288	91.59 282	78.06 261	94.27 319	74.88 300	93.15 307	95.80 239

PS-MVSNAJ		88.86 229	88.99 218	88.48 282	94.88 242	74.71 288	86.69 309	95.60 199	80.88 264	87.83 293	87.37 327	90.77 119	98.82 124	82.52 229	94.37 290	91.93 323

MG-MVS		89.54 217	89.80 209	88.76 273	94.88 242	72.47 313	89.60 266	92.44 264	85.82 220	89.48 269	95.98 156	82.85 229	97.74 238	81.87 233	95.27 274	96.08 232

xiu_mvs_v2_base		89.00 225	89.19 212	88.46 283	94.86 244	74.63 290	86.97 305	95.60 199	80.88 264	87.83 293	88.62 314	91.04 116	98.81 129	82.51 230	94.38 289	91.93 323

MAR-MVS		90.32 208	88.87 221	94.66 98	94.82 245	91.85 52	94.22 111	94.75 220	80.91 263	87.52 298	88.07 319	86.63 202	97.87 224	76.67 286	96.21 255	94.25 279
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
PVSNet_BlendedMVS		90.35 206	89.96 207	91.54 212	94.81 246	78.80 247	90.14 249	96.93 136	79.43 274	88.68 284	95.06 193	86.27 207	98.15 206	80.27 249	98.04 190	97.68 164

PVSNet_Blended		88.74 232	88.16 232	90.46 234	94.81 246	78.80 247	86.64 310	96.93 136	74.67 299	88.68 284	89.18 312	86.27 207	98.15 206	80.27 249	96.00 256	94.44 276

BH-w/o		87.21 262	87.02 253	87.79 291	94.77 248	77.27 263	87.90 293	93.21 252	81.74 260	89.99 258	88.39 317	83.47 223	96.93 267	71.29 319	92.43 315	89.15 335

LS3D		96.11 46	95.83 58	96.95 33	94.75 249	94.20 14	97.34 11	97.98 45	97.31 9	95.32 127	96.77 105	93.08 71	99.20 64	91.79 101	98.16 178	97.44 176

Effi-MVS+-dtu		93.90 121	92.60 158	97.77 4	94.74 250	96.67 4	94.00 115	95.41 208	89.94 142	91.93 220	92.13 272	90.12 134	98.97 97	87.68 174	97.48 217	97.67 165

mvs-test1		93.07 150	91.80 172	96.89 35	94.74 250	95.83 7	92.17 180	95.41 208	89.94 142	89.85 262	90.59 299	90.12 134	98.88 110	87.68 174	95.66 263	95.97 235

MVSFormer		92.18 174	92.23 163	92.04 198	94.74 250	80.06 210	97.15 13	97.37 100	88.98 157	88.83 275	92.79 254	77.02 269	99.60 8	96.41 6	96.75 238	96.46 219

lupinMVS		88.34 236	87.31 244	91.45 214	94.74 250	80.06 210	87.23 301	92.27 265	71.10 320	88.83 275	91.15 285	77.02 269	98.53 173	86.67 188	96.75 238	95.76 241

MDA-MVSNet-bldmvs		91.04 192	90.88 193	91.55 211	94.68 254	80.16 204	85.49 317	92.14 269	90.41 136	94.93 145	95.79 164	85.10 216	96.93 267	85.15 204	94.19 295	97.57 170

MVS_0304		92.99 152	92.54 159	94.35 116	94.67 255	86.06 140	91.16 218	97.92 55	90.01 141	88.33 287	94.41 214	87.02 190	99.22 62	90.36 123	99.00 101	97.76 158

Fast-Effi-MVS+-dtu		92.77 159	92.16 164	94.58 106	94.66 256	88.25 104	92.05 183	96.65 155	89.62 148	90.08 254	91.23 284	92.56 82	98.60 160	86.30 195	96.27 253	96.90 199

UnsupCasMVSNet_eth		90.33 207	90.34 203	90.28 238	94.64 257	80.24 202	89.69 265	95.88 191	85.77 221	93.94 171	95.69 168	81.99 237	92.98 329	84.21 215	91.30 324	97.62 168

1111		80.36 316	81.32 304	77.48 333	94.61 258	44.56 354	81.59 334	90.66 282	86.78 209	90.60 246	93.52 243	30.37 359	90.67 337	66.36 334	97.42 220	97.20 189

.test1245		64.72 328	70.88 329	46.22 341	94.61 258	44.56 354	81.59 334	90.66 282	86.78 209	90.60 246	93.52 243	30.37 359	90.67 337	66.36 334	3.45 355	3.44 355

OpenMVS_ROB	85.12 16	89.52 218	89.05 215	90.92 228	94.58 260	81.21 192	91.10 221	93.41 247	77.03 293	93.41 181	93.99 233	83.23 225	97.80 231	79.93 256	94.80 283	93.74 294

OpenMVS	89.45 8	92.27 173	92.13 166	92.68 174	94.53 261	84.10 162	95.70 59	97.03 127	82.44 255	91.14 238	96.42 125	88.47 156	98.38 187	85.95 197	97.47 218	95.55 252

thres200		85.85 281	85.18 281	87.88 290	94.44 262	72.52 312	89.08 280	86.21 307	88.57 173	91.44 225	88.40 316	64.22 310	98.00 211	68.35 329	95.88 261	93.12 305

DELS-MVS		92.05 176	92.16 164	91.72 205	94.44 262	80.13 208	87.62 295	97.25 116	87.34 199	92.22 215	93.18 250	89.54 144	98.73 143	89.67 141	98.20 176	96.30 225
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
N_pmnet		88.90 228	87.25 246	93.83 132	94.40 264	93.81 31	84.73 321	87.09 302	79.36 278	93.26 189	92.43 266	79.29 254	91.68 334	77.50 281	97.22 225	96.00 234

pmmvs4		88.95 227	87.70 242	92.70 173	94.30 265	85.60 147	87.22 302	92.16 268	74.62 300	89.75 266	94.19 223	77.97 262	96.41 285	82.71 227	96.36 252	96.09 231

new-patchmatchnet		88.97 226	90.79 197	83.50 321	94.28 266	55.83 351	85.34 318	93.56 244	86.18 213	95.47 122	95.73 167	83.10 226	96.51 280	85.40 201	98.06 188	98.16 130

API-MVS		91.52 183	91.61 176	91.26 220	94.16 267	86.26 136	94.66 93	94.82 217	91.17 119	92.13 216	91.08 287	90.03 140	97.06 263	79.09 263	97.35 223	90.45 333

MSDG		90.82 194	90.67 200	91.26 220	94.16 267	83.08 174	86.63 311	96.19 183	90.60 131	91.94 219	91.89 275	89.16 148	95.75 298	80.96 246	94.51 288	94.95 264

TR-MVS		87.70 249	87.17 248	89.27 264	94.11 269	79.26 233	88.69 287	91.86 272	81.94 259	90.69 244	89.79 305	82.82 230	97.42 249	72.65 310	91.98 321	91.14 328

sss		87.23 261	86.82 256	88.46 283	93.96 270	77.94 253	86.84 307	92.78 258	77.59 288	87.61 297	91.83 276	78.75 256	91.92 333	77.84 276	94.20 294	95.52 253

PVSNet	76.22 20	82.89 297	82.37 296	84.48 316	93.96 270	64.38 341	78.60 341	88.61 288	71.50 318	84.43 318	86.36 333	74.27 277	94.60 313	69.87 327	93.69 301	94.46 275

semantic-postprocess				91.94 199	93.89 272	79.22 238		93.51 245	91.53 113	95.37 126	96.62 114	77.17 267	98.90 104	91.89 100	94.95 279	97.70 162

UGNet		93.08 148	92.50 161	94.79 95	93.87 273	87.99 108	95.07 79	94.26 232	90.64 129	87.33 299	97.67 61	86.89 197	98.49 176	88.10 170	98.71 130	97.91 147
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
PAPM		81.91 305	80.11 315	87.31 294	93.87 273	72.32 314	84.02 327	93.22 250	69.47 329	76.13 348	89.84 302	72.15 281	97.23 257	53.27 349	89.02 330	92.37 315

CANet		92.38 170	91.99 168	93.52 142	93.82 275	83.46 168	91.14 219	97.00 129	89.81 145	86.47 304	94.04 229	87.90 173	99.21 63	89.50 143	98.27 166	97.90 148

test1235678		84.54 288	83.85 290	86.59 299	93.81 276	73.41 300	82.38 331	91.79 273	79.43 274	89.50 268	91.61 281	70.59 285	92.94 330	58.14 345	97.40 221	93.44 301

HY-MVS	82.50 18	86.81 273	85.93 276	89.47 254	93.63 277	77.93 254	94.02 114	91.58 275	75.68 296	83.64 322	93.64 238	77.40 265	97.42 249	71.70 316	92.07 320	93.05 306

no-one		87.84 246	87.21 247	89.74 248	93.58 278	78.64 250	81.28 336	92.69 260	74.36 302	92.05 218	97.14 88	81.86 240	96.07 293	72.03 313	99.90 2	94.52 273

MVS_Test		92.57 166	93.29 141	90.40 235	93.53 279	75.85 277	92.52 162	96.96 133	88.73 168	92.35 210	96.70 112	90.77 119	98.37 190	92.53 85	95.49 267	96.99 195

EU-MVSNet		87.39 257	86.71 259	89.44 259	93.40 280	76.11 274	94.93 85	90.00 284	57.17 349	95.71 116	97.37 77	64.77 309	97.68 241	92.67 80	94.37 290	94.52 273

MS-PatchMatch		88.05 243	87.75 240	88.95 270	93.28 281	77.93 254	87.88 294	92.49 263	75.42 298	92.57 205	93.59 241	80.44 251	94.24 321	81.28 239	92.75 312	94.69 270

GA-MVS		87.70 249	86.82 256	90.31 237	93.27 282	77.22 264	84.72 323	92.79 257	85.11 228	89.82 263	90.07 300	66.80 297	97.76 236	84.56 213	94.27 293	95.96 236

pmmvs5		87.87 245	87.14 249	90.07 245	93.26 283	76.97 268	88.89 284	92.18 266	73.71 308	88.36 286	93.89 234	76.86 272	96.73 274	80.32 248	96.81 235	96.51 210

IterMVS		90.18 210	90.16 205	90.21 243	93.15 284	75.98 276	87.56 298	92.97 253	86.43 212	94.09 167	96.40 127	78.32 259	97.43 248	87.87 173	94.69 285	97.23 187
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
MVS-HIRNet		78.83 321	80.60 310	73.51 337	93.07 285	47.37 352	87.10 304	78.00 350	68.94 330	77.53 346	97.26 82	71.45 283	94.62 312	63.28 341	88.74 331	78.55 349

FMVSNet3		90.78 196	90.32 204	92.16 194	93.03 286	79.92 218	92.54 160	94.95 214	86.17 214	95.10 137	96.01 155	69.97 288	98.75 139	86.74 185	98.38 152	97.82 156

PAPR		87.65 252	86.77 258	90.27 239	92.85 287	77.38 261	88.56 289	96.23 180	76.82 295	84.98 313	89.75 307	86.08 209	97.16 259	72.33 311	93.35 303	96.26 227

Regformer-1		94.55 103	94.33 107	95.19 84	92.83 288	88.54 98	91.87 194	95.84 194	93.99 45	95.95 103	95.04 194	92.00 91	98.79 131	93.14 68	98.31 161	98.23 124

Regformer-2		94.86 89	94.55 100	95.77 63	92.83 288	89.98 70	91.87 194	96.40 168	94.38 42	96.19 94	95.04 194	92.47 86	99.04 85	93.49 54	98.31 161	98.28 122

Regformer-3		94.28 111	94.23 113	94.46 111	92.78 290	86.28 135	92.39 171	94.70 222	93.69 55	95.97 101	95.56 174	91.34 104	98.48 179	93.45 57	98.14 180	98.62 107

Regformer-4		94.90 86	94.67 97	95.59 71	92.78 290	89.02 85	92.39 171	95.91 190	94.50 38	96.41 77	95.56 174	92.10 89	99.01 91	94.23 37	98.14 180	98.74 98

EI-MVSNet-Vis-set		94.36 108	94.28 109	94.61 99	92.55 292	85.98 141	92.44 169	94.69 223	93.70 52	96.12 97	95.81 163	91.24 109	98.86 119	93.76 48	98.22 173	98.98 75

EI-MVSNet-UG-set		94.35 109	94.27 111	94.59 104	92.46 293	85.87 142	92.42 170	94.69 223	93.67 56	96.13 96	95.84 162	91.20 112	98.86 119	93.78 45	98.23 171	99.03 66

testus		82.09 304	81.78 299	83.03 323	92.35 294	64.37 342	79.44 339	93.27 249	73.08 311	87.06 301	85.21 337	76.80 273	89.27 344	53.30 348	95.48 268	95.46 254

FMVSNet5		87.82 248	86.56 261	91.62 208	92.31 295	79.81 220	93.49 132	94.81 219	83.26 242	91.36 226	96.93 97	52.77 348	97.49 247	76.07 290	98.03 191	97.55 173

diffmvs		90.45 201	90.49 201	90.34 236	92.25 296	77.09 265	91.80 205	95.96 189	82.68 250	85.83 308	95.07 192	87.01 191	97.09 261	89.68 140	94.10 296	96.83 204

MDA-MVSNet_test_wron		88.16 242	88.23 229	87.93 288	92.22 297	73.71 297	80.71 338	88.84 286	82.52 253	94.88 146	95.14 188	82.70 231	93.61 324	83.28 222	93.80 299	96.46 219

YYNet1		88.17 241	88.24 228	87.93 288	92.21 298	73.62 298	80.75 337	88.77 287	82.51 254	94.99 143	95.11 190	82.70 231	93.70 323	83.33 221	93.83 298	96.48 218

CANet_DTU		89.85 214	89.17 213	91.87 201	92.20 299	80.02 214	90.79 228	95.87 192	86.02 216	82.53 329	91.77 277	80.01 252	98.57 165	85.66 199	97.70 205	97.01 194

mvs_anonymous		90.37 205	91.30 186	87.58 292	92.17 300	68.00 325	89.84 262	94.73 221	83.82 241	93.22 193	97.40 75	87.54 177	97.40 251	87.94 171	95.05 278	97.34 183

EI-MVSNet		92.99 152	93.26 145	92.19 192	92.12 301	79.21 239	92.32 174	94.67 225	91.77 106	95.24 132	95.85 160	87.14 188	98.49 176	91.99 96	98.26 167	98.86 86

CVMVSNet		85.16 285	84.72 283	86.48 300	92.12 301	70.19 319	92.32 174	88.17 294	56.15 350	90.64 245	95.85 160	67.97 292	96.69 275	88.78 159	90.52 328	92.56 313

Patchmatch-test1		87.28 259	87.30 245	87.22 295	92.01 303	71.98 315	89.43 270	88.11 295	82.26 257	88.71 282	92.20 270	78.65 257	95.81 297	80.99 245	93.30 304	93.87 291

1314		86.46 278	86.33 266	86.87 298	91.65 304	74.54 291	91.94 188	94.10 234	74.28 303	84.78 315	87.33 328	83.03 227	95.00 311	78.72 270	91.16 326	91.06 329

cascas		87.02 268	86.28 267	89.25 265	91.56 305	76.45 270	84.33 325	96.78 149	71.01 321	86.89 303	85.91 334	81.35 242	96.94 266	83.09 224	95.60 264	94.35 278

IB-MVS	77.21 19	83.11 294	81.05 306	89.29 263	91.15 306	75.85 277	85.66 316	86.00 310	79.70 272	82.02 334	86.61 329	48.26 352	98.39 185	77.84 276	92.22 318	93.63 296
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
MVS		84.98 287	84.30 286	87.01 296	91.03 307	77.69 259	91.94 188	94.16 233	59.36 348	84.23 319	87.50 326	85.66 213	96.80 272	71.79 314	93.05 310	86.54 341

CR-MVSNet		87.89 244	87.12 250	90.22 241	91.01 308	78.93 242	92.52 162	92.81 255	73.08 311	89.10 272	96.93 97	67.11 294	97.64 242	88.80 158	92.70 313	94.08 281

RPMNet		89.30 220	89.00 217	90.22 241	91.01 308	78.93 242	92.52 162	87.85 297	91.91 95	89.10 272	96.89 100	68.84 289	97.64 242	90.17 130	92.70 313	94.08 281

new_pmnet		81.22 309	81.01 308	81.86 327	90.92 310	70.15 320	84.03 326	80.25 349	70.83 323	85.97 307	89.78 306	67.93 293	84.65 350	67.44 331	91.90 322	90.78 330

test12356		76.35 322	77.41 323	73.19 338	90.70 311	38.86 357	74.56 343	91.14 277	74.55 301	80.54 341	88.18 318	52.36 349	90.49 341	52.38 350	92.26 317	90.21 334

PatchT		87.51 254	88.17 231	85.55 306	90.64 312	66.91 329	92.02 184	86.09 308	92.20 88	89.05 274	97.16 87	64.15 311	96.37 288	89.21 152	92.98 311	93.37 303

Patchmatch-test		86.10 280	86.01 275	86.38 302	90.63 313	74.22 296	89.57 267	86.69 304	85.73 222	89.81 264	92.83 253	65.24 307	91.04 336	77.82 278	95.78 262	93.88 290

PVSNet_0	70.34 21	74.58 324	72.96 326	79.47 331	90.63 313	66.24 334	73.26 344	83.40 333	63.67 345	78.02 345	78.35 349	72.53 279	89.59 343	56.68 346	60.05 352	82.57 347

PMMVS2		81.31 308	83.44 291	74.92 336	90.52 315	46.49 353	69.19 349	85.23 323	84.30 237	87.95 292	94.71 208	76.95 271	84.36 351	64.07 338	98.09 186	93.89 289

tpm		84.38 290	84.08 287	85.30 311	90.47 316	63.43 344	89.34 273	85.63 314	77.24 292	87.62 296	95.03 196	61.00 330	97.30 255	79.26 261	91.09 327	95.16 257

PNet_i23d		72.03 327	70.91 328	75.38 335	90.46 317	57.84 349	71.73 348	81.53 345	83.86 240	82.21 330	83.49 342	29.97 361	87.80 348	60.78 342	54.12 353	80.51 348

wuyk23d		87.83 247	90.79 197	78.96 332	90.46 317	88.63 93	92.72 155	90.67 281	91.65 110	98.68 11	97.64 62	96.06 15	77.53 353	59.84 343	99.41 60	70.73 350

Patchmtry		90.11 212	89.92 208	90.66 230	90.35 319	77.00 267	92.96 149	92.81 255	90.25 138	94.74 150	96.93 97	67.11 294	97.52 245	85.17 202	98.98 102	97.46 175

CHOSEN 280x420		80.04 318	77.97 322	86.23 304	90.13 320	74.53 292	72.87 346	89.59 285	66.38 338	76.29 347	85.32 336	56.96 343	95.36 305	69.49 328	94.72 284	88.79 338

MVSTER		89.32 219	88.75 222	91.03 224	90.10 321	76.62 269	90.85 226	94.67 225	82.27 256	95.24 132	95.79 164	61.09 329	98.49 176	90.49 117	98.26 167	97.97 143

tpm2		81.46 307	80.35 313	84.80 313	89.90 322	65.14 337	90.44 239	85.36 317	65.82 341	82.05 333	92.44 265	57.94 340	96.69 275	70.71 324	88.49 333	92.56 313

test0.0.03 1		82.48 300	81.47 303	85.48 307	89.70 323	73.57 299	84.73 321	81.64 344	83.07 247	88.13 290	86.61 329	62.86 324	89.10 346	66.24 336	90.29 329	93.77 293

test-LLR		83.58 293	83.17 293	84.79 314	89.68 324	66.86 331	83.08 328	84.52 326	83.07 247	82.85 327	84.78 338	62.86 324	93.49 325	82.85 225	94.86 280	94.03 284

test-mter		81.21 310	80.01 316	84.79 314	89.68 324	66.86 331	83.08 328	84.52 326	73.85 307	82.85 327	84.78 338	43.66 356	93.49 325	82.85 225	94.86 280	94.03 284

DSMNet-mixed		82.21 302	81.56 301	84.16 318	89.57 326	70.00 321	90.65 232	77.66 351	54.99 351	83.30 325	97.57 64	77.89 263	90.50 340	66.86 333	95.54 266	91.97 322

PatchmatchNet		85.22 284	84.64 284	86.98 297	89.51 327	69.83 322	90.52 237	87.34 301	78.87 281	87.22 300	92.74 256	66.91 296	96.53 278	81.77 234	86.88 336	94.58 272
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
MDTV_nov1_ep13			83.88 289		89.42 328	61.52 345	88.74 286	87.41 300	73.99 306	84.96 314	94.01 232	65.25 306	95.53 299	78.02 274	93.16 306

tpmp4_e23		81.87 306	80.41 311	86.27 303	89.29 329	67.84 326	91.58 208	87.61 299	67.42 335	78.60 344	92.71 257	56.42 345	96.87 269	71.44 318	88.63 332	94.10 280

CostFormer		83.09 295	82.21 297	85.73 305	89.27 330	67.01 328	90.35 241	86.47 306	70.42 325	83.52 324	93.23 249	61.18 328	96.85 270	77.21 283	88.26 334	93.34 304

ADS-MVSNet2		84.01 292	82.20 298	89.41 260	89.04 331	76.37 271	87.57 296	90.98 280	72.71 314	84.46 316	92.45 263	68.08 290	96.48 281	70.58 325	83.97 338	95.38 255

ADS-MVSNet		82.25 301	81.55 302	84.34 317	89.04 331	65.30 335	87.57 296	85.13 324	72.71 314	84.46 316	92.45 263	68.08 290	92.33 332	70.58 325	83.97 338	95.38 255

tpm cat1		80.61 315	79.46 317	84.07 319	88.78 333	65.06 339	89.26 276	88.23 292	62.27 346	81.90 335	89.66 309	62.70 326	95.29 308	71.72 315	80.60 347	91.86 325

CMPMVS	68.83 22	87.28 259	85.67 278	92.09 196	88.77 334	85.42 149	90.31 243	94.38 229	70.02 327	88.00 291	93.30 248	73.78 278	94.03 322	75.96 292	96.54 246	96.83 204
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
PatchFormer-LS_test		82.62 299	81.71 300	85.32 310	87.92 335	67.31 327	89.03 281	88.20 293	77.58 289	83.79 321	80.50 348	60.96 331	96.42 284	83.86 219	83.59 340	92.23 320

LP		86.29 279	85.35 280	89.10 267	87.80 336	76.21 272	89.92 257	90.99 279	84.86 233	87.66 295	92.32 268	70.40 286	96.48 281	81.94 232	82.24 345	94.63 271

tpmrst		82.85 298	82.93 295	82.64 325	87.65 337	58.99 348	90.14 249	87.90 296	75.54 297	83.93 320	91.63 280	66.79 299	95.36 305	81.21 241	81.54 346	93.57 300

JIA-IIPM		85.08 286	83.04 294	91.19 223	87.56 338	86.14 138	89.40 272	84.44 332	88.98 157	82.20 331	97.95 48	56.82 344	96.15 291	76.55 288	83.45 341	91.30 327

TESTMET0.1,1		79.09 320	78.04 321	82.25 326	87.52 339	64.03 343	83.08 328	80.62 347	70.28 326	80.16 342	83.22 343	44.13 355	90.56 339	79.95 254	93.36 302	92.15 321

DWT-MVSNet_test		80.74 313	79.18 318	85.43 308	87.51 340	66.87 330	89.87 261	86.01 309	74.20 305	80.86 338	80.62 347	48.84 351	96.68 277	81.54 236	83.14 343	92.75 311

gg-mvs-nofinetune		82.10 303	81.02 307	85.34 309	87.46 341	71.04 317	94.74 90	67.56 354	96.44 19	79.43 343	98.99 6	45.24 353	96.15 291	67.18 332	92.17 319	88.85 337

pmmvs3		80.83 312	78.96 319	86.45 301	87.23 342	77.48 260	84.87 320	82.31 341	63.83 344	85.03 312	89.50 310	49.66 350	93.10 327	73.12 307	95.10 277	88.78 339

tpmvs		84.22 291	83.97 288	84.94 312	87.09 343	65.18 336	91.21 217	88.35 290	82.87 249	85.21 310	90.96 289	65.24 307	96.75 273	79.60 260	85.25 337	92.90 308

gm-plane-assit					87.08 344	59.33 347			71.22 319		83.58 341		97.20 258	73.95 301

MVE	59.87 23	73.86 326	72.65 327	77.47 334	87.00 345	74.35 294	61.37 351	60.93 356	67.27 336	69.69 352	86.49 331	81.24 247	72.33 354	56.45 347	83.45 341	85.74 342
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
EPNet_dtu		85.63 283	84.37 285	89.40 261	86.30 346	74.33 295	91.64 207	88.26 291	84.84 234	72.96 351	89.85 301	71.27 284	97.69 240	76.60 287	97.62 209	96.18 229
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
test2356		75.58 323	73.13 325	82.95 324	86.10 347	66.42 333	75.07 342	84.87 325	70.91 322	80.85 339	80.66 346	38.02 358	88.98 347	49.32 351	92.35 316	93.44 301

dp		79.28 319	78.62 320	81.24 328	85.97 348	56.45 350	86.91 306	85.26 322	72.97 313	81.45 337	89.17 313	56.01 347	95.45 303	73.19 306	76.68 349	91.82 326

EPMVS		81.17 311	80.37 312	83.58 320	85.58 349	65.08 338	90.31 243	71.34 353	77.31 291	85.80 309	91.30 283	59.38 332	92.70 331	79.99 253	82.34 344	92.96 307

E-PMN		80.72 314	80.86 309	80.29 330	85.11 350	68.77 324	72.96 345	81.97 343	87.76 193	83.25 326	83.01 344	62.22 327	89.17 345	77.15 284	94.31 292	82.93 345

GG-mvs-BLEND				83.24 322	85.06 351	71.03 318	94.99 84	65.55 355		74.09 350	75.51 350	44.57 354	94.46 315	59.57 344	87.54 335	84.24 343

EMVS		80.35 317	80.28 314	80.54 329	84.73 352	69.07 323	72.54 347	80.73 346	87.80 192	81.66 336	81.73 345	62.89 323	89.84 342	75.79 297	94.65 286	82.71 346

EPNet		89.80 215	88.25 227	94.45 112	83.91 353	86.18 137	93.87 124	87.07 303	91.16 120	80.64 340	94.72 207	78.83 255	98.89 106	85.17 202	98.89 107	98.28 122
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
PMMVS		83.00 296	81.11 305	88.66 276	83.81 354	86.44 131	82.24 333	85.65 313	61.75 347	82.07 332	85.64 335	79.75 253	91.59 335	75.99 291	93.09 308	87.94 340

testpf		74.01 325	76.37 324	66.95 339	80.56 355	60.00 346	88.43 291	75.07 352	81.54 261	75.75 349	83.73 340	38.93 357	83.09 352	84.01 216	79.32 348	57.75 351

DeepMVS_CX				53.83 340	70.38 356	64.56 340		48.52 358	33.01 352	65.50 353	74.21 351	56.19 346	46.64 355	38.45 353	70.07 350	50.30 352

tmp_tt		37.97 330	44.33 330	18.88 343	11.80 357	21.54 358	63.51 350	45.66 359	4.23 353	51.34 354	50.48 352	59.08 333	22.11 356	44.50 352	68.35 351	13.00 353

test123		9.49 332	12.01 333	1.91 344	2.87 358	1.30 359	82.38 331	1.34 361	1.36 354	2.84 355	6.56 355	2.45 362	0.97 357	2.73 354	5.56 354	3.47 354

testmvs		9.02 333	11.42 334	1.81 345	2.77 359	1.13 360	79.44 339	1.90 360	1.18 355	2.65 356	6.80 354	1.95 363	0.87 358	2.62 355	3.45 355	3.44 355

cdsmvs_eth3d_5k		23.35 331	31.13 332	0.00 346	0.00 360	0.00 361	0.00 352	95.58 203	0.00 356	0.00 357	91.15 285	93.43 62	0.00 359	0.00 356	0.00 357	0.00 357

pcd_1.5k_mvsjas		7.56 334	10.09 335	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	0.00 358	90.77 119	0.00 359	0.00 356	0.00 357	0.00 357

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		7.56 334	10.08 336	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	90.69 296	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																94.75 268

test_part3							93.92 122		91.83 101		96.39 131		99.44 24	89.00 154

test_part1								98.14 28				94.69 45			99.10 91	98.17 128

sam_mvs1												66.64 300				94.75 268

sam_mvs												66.41 301

MTGPA								97.62 75

test_post1							90.21 245				5.85 357	65.36 305	96.00 294	79.61 259

test_post											6.07 356	65.74 304	95.84 296

patchmatchnet-post											91.71 278	66.22 303	97.59 244

MTMP								54.62 357

test9_res														88.16 169	98.40 150	97.83 154

agg_prior2														87.06 183	98.36 158	97.98 140

test_prior4						89.91 71	90.74 229

test_prior2							90.21 245		89.33 152	90.77 241	94.81 201	90.41 130		88.21 166	98.55 138

旧先验2							90.00 255		68.65 331	92.71 202			96.52 279	85.15 204

新几何2							90.02 254

无先验							89.94 256	95.75 196	70.81 324				98.59 162	81.17 242		94.81 265

原ACMM2							89.34 273

testdata2													98.03 210	80.24 251

segment_acmp												92.14 88

testdata1							88.96 283		88.44 178

plane_prior5								97.81 63					98.95 100	89.26 149	98.51 143	98.60 109

plane_prior4											95.59 169

plane_prior3						88.43 103			90.35 137	93.31 184

plane_prior2							94.56 100		91.74 108

plane_prior						88.12 106	93.01 145		88.98 157						98.06 188

n2								0.00 362

nn								0.00 362

door-mid								92.13 270

test11								96.65 155

door								91.26 276

HQP5-MVS						84.89 153

BP-MVS														86.55 191

HQP4-MVS										88.81 277			98.61 158			98.15 131

HQP3-MVS								97.31 110							97.73 202

HQP2-MVS												84.76 218

MDTV_nov1_ep13_2view						42.48 356	88.45 290		67.22 337	83.56 323		66.80 297		72.86 309		94.06 283

ACMMP++_ref															98.82 120

ACMMP++															99.25 76

Test By Simon												90.61 126

High-resolution multi-view benchmark