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		99.86 1	99.86 1	99.87 1	99.99 1	99.77 1	99.77 1	99.80 1	99.97 1	99.97 1	99.95 1	99.74 1	99.98 1	99.56 1	100.00 1	99.85 4

Anonymous20231211		99.29 2	99.41 2	98.91 22	99.94 2	97.08 37	99.47 3	99.51 5	99.56 2	99.83 3	99.80 2	99.13 3	99.90 13	97.55 51	99.93 21	99.75 13

pmmvs6		99.07 4	99.24 4	98.56 45	99.81 3	96.38 57	98.87 10	99.30 9	99.01 15	99.63 9	99.66 4	99.27 2	99.68 104	97.75 43	99.89 33	99.62 32

OurMVSNet-221017-0		98.61 19	98.61 26	98.63 41	99.77 4	96.35 58	99.17 6	99.05 38	98.05 42	99.61 11	99.52 5	93.72 165	99.88 19	98.72 20	99.88 34	99.65 25

Gipuma		98.07 42	98.31 38	97.36 128	99.76 5	96.28 61	98.51 22	99.10 25	98.76 20	96.79 187	99.34 20	96.61 66	98.82 294	96.38 85	99.50 113	96.98 293
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
MIMVSNet1		98.51 24	98.45 32	98.67 38	99.72 6	96.71 46	98.76 11	98.89 78	98.49 25	99.38 18	99.14 42	95.44 107	99.84 28	96.47 83	99.80 47	99.47 65

LTVRE_ROB	96.88 1	99.18 3	99.34 3	98.72 35	99.71 7	96.99 40	99.69 2	99.57 3	99.02 14	99.62 10	99.36 16	98.53 8	99.52 165	98.58 24	99.95 13	99.66 24
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
mvs_tets		98.90 5	98.94 8	98.75 30	99.69 8	96.48 55	98.54 21	99.22 10	96.23 112	99.71 5	99.48 7	98.77 7	99.93 2	98.89 10	99.95 13	99.84 6

PS-MVSNAJss		98.53 23	98.63 22	98.21 70	99.68 9	94.82 106	98.10 45	99.21 11	96.91 89	99.75 4	99.45 9	95.82 91	99.92 4	98.80 13	99.96 11	99.89 1

jajsoiax		98.77 11	98.79 15	98.74 32	99.66 10	96.48 55	98.45 26	99.12 22	95.83 127	99.67 7	99.37 14	98.25 11	99.92 4	98.77 14	99.94 19	99.82 7

pcd1.5k->3k		41.47 335	44.19 336	33.29 348	99.65 11	0.00 366	0.00 357	99.07 34	0.00 361	0.00 362	0.00 363	99.04 4	0.00 364	0.00 361	99.96 11	99.87 2

v7n		98.73 13	98.99 7	97.95 83	99.64 12	94.20 127	98.67 13	99.14 20	99.08 9	99.42 16	99.23 29	96.53 69	99.91 12	99.27 4	99.93 21	99.73 16

test_djsdf		98.73 13	98.74 18	98.69 37	99.63 13	96.30 60	98.67 13	99.02 51	96.50 101	99.32 21	99.44 10	97.43 30	99.92 4	98.73 17	99.95 13	99.86 3

anonymousdsp		98.72 16	98.63 22	98.99 10	99.62 14	97.29 34	98.65 16	99.19 14	95.62 133	99.35 20	99.37 14	97.38 32	99.90 13	98.59 23	99.91 27	99.77 9

v748		98.58 20	98.89 10	97.67 100	99.61 15	93.53 151	98.59 17	98.90 76	98.97 17	99.43 15	99.15 41	96.53 69	99.85 24	98.88 11	99.91 27	99.64 28

v52		98.85 8	99.01 5	98.37 56	99.61 15	95.53 84	99.01 7	99.04 45	98.48 26	99.31 22	99.41 11	96.82 56	99.87 21	99.44 2	99.95 13	99.70 19

V4		98.85 8	99.01 5	98.37 56	99.61 15	95.53 84	99.01 7	99.04 45	98.48 26	99.31 22	99.41 11	96.81 57	99.87 21	99.44 2	99.95 13	99.70 19

PEN-MVS		98.75 12	98.85 13	98.44 50	99.58 18	95.67 77	98.45 26	99.15 19	99.33 4	99.30 24	99.00 49	97.27 37	99.92 4	97.64 46	99.92 24	99.75 13

Baseline_NR-MVSNet		97.72 74	97.79 60	97.50 113	99.56 19	93.29 156	95.44 191	98.86 84	98.20 38	98.37 77	99.24 27	94.69 127	99.55 157	95.98 99	99.79 48	99.65 25

SixPastTwentyTwo		97.49 91	97.57 81	97.26 134	99.56 19	92.33 169	98.28 32	96.97 261	98.30 34	99.45 14	99.35 18	88.43 258	99.89 17	98.01 33	99.76 51	99.54 46

PS-CasMVS		98.73 13	98.85 13	98.39 55	99.55 21	95.47 86	98.49 23	99.13 21	99.22 7	99.22 29	98.96 53	97.35 33	99.92 4	97.79 41	99.93 21	99.79 8

DTE-MVSNet		98.79 10	98.86 11	98.59 43	99.55 21	96.12 64	98.48 25	99.10 25	99.36 3	99.29 25	99.06 48	97.27 37	99.93 2	97.71 45	99.91 27	99.70 19

HPM-MVS_fast		98.32 31	98.13 45	98.88 23	99.54 23	97.48 27	98.35 29	99.03 50	95.88 124	97.88 134	98.22 111	98.15 13	99.74 60	96.50 82	99.62 80	99.42 87

TDRefinement		98.90 5	98.86 11	99.02 8	99.54 23	98.06 6	99.34 5	99.44 7	98.85 19	99.00 41	99.20 31	97.42 31	99.59 145	97.21 64	99.76 51	99.40 92

Anonymous20240521		98.58 20	98.65 21	98.36 59	99.52 25	95.60 79	98.96 9	98.95 72	98.36 30	99.25 27	99.17 39	95.28 113	99.80 37	98.46 25	99.88 34	99.68 23

pm-mvs1		98.47 25	98.67 19	97.86 87	99.52 25	94.58 114	98.28 32	99.00 62	97.57 65	99.27 26	99.22 30	98.32 10	99.50 177	97.09 70	99.75 55	99.50 51

TransMVSNet (Re)		98.38 29	98.67 19	97.51 110	99.51 27	93.39 155	98.20 40	98.87 82	98.23 36	99.48 12	99.27 25	98.47 9	99.55 157	96.52 80	99.53 106	99.60 35

WR-MVS_H		98.65 18	98.62 24	98.75 30	99.51 27	96.61 51	98.55 20	99.17 15	99.05 12	99.17 32	98.79 61	95.47 105	99.89 17	97.95 34	99.91 27	99.75 13

PMVS	89.60 17	96.71 144	96.97 121	95.95 214	99.51 27	97.81 13	97.42 89	97.49 241	97.93 46	95.95 223	98.58 76	96.88 52	96.91 348	89.59 259	99.36 156	93.12 347
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
MP-MVS-pluss		97.69 77	97.36 91	98.70 36	99.50 30	96.84 43	95.38 200	98.99 65	92.45 240	98.11 103	98.31 99	97.25 39	99.77 48	96.60 77	99.62 80	99.48 62
MP-MVS-pluss: MP-MVS-pluss. MP-MVS-pluss
FC-MVSNet-test		98.16 37	98.37 34	97.56 105	99.49 31	93.10 159	98.35 29	99.21 11	98.43 28	98.89 45	98.83 60	94.30 144	99.81 33	97.87 37	99.91 27	99.77 9

VPNet		97.26 106	97.49 87	96.59 168	99.47 32	90.58 204	96.27 142	98.53 146	97.77 50	98.46 72	98.41 89	94.59 133	99.68 104	94.61 152	99.29 172	99.52 49

CP-MVSNet		98.42 27	98.46 30	98.30 65	99.46 33	95.22 94	98.27 34	98.84 88	99.05 12	99.01 39	98.65 74	95.37 108	99.90 13	97.57 50	99.91 27	99.77 9

XXY-MVS		97.54 88	97.70 66	97.07 142	99.46 33	92.21 173	97.22 96	99.00 62	94.93 166	98.58 63	98.92 57	97.31 35	99.41 210	94.44 156	99.43 141	99.59 36

zzz-MVS		98.01 47	97.66 70	99.06 5	99.44 35	97.90 8	95.66 180	98.73 114	97.69 59	97.90 130	97.96 141	95.81 95	99.82 31	96.13 90	99.61 85	99.45 72

MTAPA		98.14 38	97.84 58	99.06 5	99.44 35	97.90 8	97.25 93	98.73 114	97.69 59	97.90 130	97.96 141	95.81 95	99.82 31	96.13 90	99.61 85	99.45 72

wuykxyi23d		98.68 17	98.53 27	99.13 3	99.44 35	97.97 7	96.85 119	99.02 51	95.81 128	99.88 2	99.38 13	98.14 14	99.69 98	98.32 29	99.95 13	99.73 16

SteuartSystems-ACMMP		98.02 45	97.76 63	98.79 28	99.43 38	97.21 36	97.15 97	98.90 76	96.58 100	98.08 109	97.87 152	97.02 47	99.76 49	95.25 126	99.59 90	99.40 92
Skip Steuart: Steuart Systems R&D Blog.
ACMH	93.61 9	98.44 26	98.76 16	97.51 110	99.43 38	93.54 150	98.23 35	99.05 38	97.40 81	99.37 19	99.08 47	98.79 6	99.47 184	97.74 44	99.71 64	99.50 51
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
HPM-MVS		98.11 41	97.83 59	98.92 19	99.42 40	97.46 28	98.57 18	99.05 38	95.43 142	97.41 158	97.50 183	97.98 17	99.79 39	95.58 116	99.57 96	99.50 51
Chunlin Ren, Qingshan Xu, Shikun Zhang, Jiaqi Yang: Hierarchical Prior Mining for Non-local Multi-View Stereo. ICCV 2023
K. test v3		96.44 156	96.28 157	96.95 148	99.41 41	91.53 190	97.65 72	90.31 337	98.89 18	98.93 44	99.36 16	84.57 284	99.92 4	97.81 39	99.56 98	99.39 95

VDDNet		96.98 117	96.84 128	97.41 125	99.40 42	93.26 157	97.94 54	95.31 289	99.26 6	98.39 76	99.18 35	87.85 265	99.62 129	95.13 136	99.09 192	99.35 106

ACMH+	93.58 10	98.23 36	98.31 38	97.98 82	99.39 43	95.22 94	97.55 82	99.20 13	98.21 37	99.25 27	98.51 83	98.21 12	99.40 213	94.79 147	99.72 60	99.32 108

TSAR-MVS + MP.		97.42 94	97.23 103	98.00 81	99.38 44	95.00 100	97.63 74	98.20 191	93.00 226	98.16 98	98.06 132	95.89 86	99.72 71	95.67 107	99.10 191	99.28 119

FIs		97.93 55	98.07 48	97.48 117	99.38 44	92.95 161	98.03 51	99.11 23	98.04 43	98.62 58	98.66 72	93.75 164	99.78 40	97.23 63	99.84 41	99.73 16

lessismore_v0				97.05 143	99.36 46	92.12 177		84.07 357		98.77 52	98.98 51	85.36 278	99.74 60	97.34 61	99.37 153	99.30 112

ACMMP_Plus		97.89 60	97.63 75	98.67 38	99.35 47	96.84 43	96.36 138	98.79 103	95.07 162	97.88 134	98.35 94	97.24 40	99.72 71	96.05 93	99.58 93	99.45 72

Vis-MVSNet		98.27 33	98.34 36	98.07 75	99.33 48	95.21 96	98.04 49	99.46 6	97.32 84	97.82 143	99.11 44	96.75 59	99.86 23	97.84 38	99.36 156	99.15 135
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
ANet_high		98.31 32	98.94 8	96.41 180	99.33 48	89.64 217	97.92 56	99.56 4	99.27 5	99.66 8	99.50 6	97.67 25	99.83 30	97.55 51	99.98 3	99.77 9

MP-MVS		97.64 80	97.18 107	99.00 9	99.32 50	97.77 14	97.49 85	98.73 114	96.27 109	95.59 236	97.75 162	96.30 79	99.78 40	93.70 184	99.48 123	99.45 72
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
PVSNet_Blended_VisFu		95.95 172	95.80 174	96.42 179	99.28 51	90.62 203	95.31 206	99.08 30	88.40 282	96.97 181	98.17 118	92.11 208	99.78 40	93.64 185	99.21 179	98.86 186

tfpnnormal		97.72 74	97.97 52	96.94 149	99.26 52	92.23 172	97.83 61	98.45 153	98.25 35	99.13 33	98.66 72	96.65 63	99.69 98	93.92 178	99.62 80	98.91 175

HSP-MVS		97.37 98	96.85 127	98.92 19	99.26 52	97.70 15	97.66 71	98.23 187	95.65 131	98.51 67	96.46 245	92.15 206	99.81 33	95.14 135	98.58 239	99.26 123

testgi		96.07 168	96.50 150	94.80 252	99.26 52	87.69 271	95.96 164	98.58 144	95.08 161	98.02 116	96.25 256	97.92 18	97.60 344	88.68 274	98.74 225	99.11 146

IS-MVSNet		96.93 122	96.68 137	97.70 96	99.25 55	94.00 133	98.57 18	96.74 269	98.36 30	98.14 101	97.98 140	88.23 259	99.71 81	93.10 195	99.72 60	99.38 97

ACMMP		98.05 43	97.75 64	98.93 18	99.23 56	97.60 19	98.09 46	98.96 70	95.75 130	97.91 129	98.06 132	96.89 50	99.76 49	95.32 124	99.57 96	99.43 85
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
nrg030		98.54 22	98.62 24	98.32 62	99.22 57	95.66 78	97.90 57	99.08 30	98.31 33	99.02 38	98.74 66	97.68 24	99.61 135	97.77 42	99.85 40	99.70 19

region2R		97.92 56	97.59 79	98.92 19	99.22 57	97.55 23	97.60 78	98.84 88	96.00 119	97.22 163	97.62 172	96.87 53	99.76 49	95.48 117	99.43 141	99.46 67

mPP-MVS		97.91 58	97.53 83	99.04 7	99.22 57	97.87 11	97.74 68	98.78 106	96.04 117	97.10 170	97.73 165	96.53 69	99.78 40	95.16 133	99.50 113	99.46 67

COLMAP_ROB	94.48 6	98.25 35	98.11 46	98.64 40	99.21 60	97.35 32	97.96 53	99.16 16	98.34 32	98.78 49	98.52 82	97.32 34	99.45 193	94.08 170	99.67 74	99.13 138
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
ACMMPR		97.95 51	97.62 77	98.94 15	99.20 61	97.56 22	97.59 79	98.83 96	96.05 115	97.46 156	97.63 171	96.77 58	99.76 49	95.61 113	99.46 127	99.49 59

PGM-MVS		97.88 61	97.52 84	98.96 13	99.20 61	97.62 18	97.09 106	99.06 36	95.45 140	97.55 147	97.94 145	97.11 42	99.78 40	94.77 149	99.46 127	99.48 62

test_0402		97.84 64	97.97 52	97.47 118	99.19 63	94.07 130	96.71 126	98.73 114	98.66 22	98.56 64	98.41 89	96.84 55	99.69 98	94.82 144	99.81 44	98.64 203

EPP-MVSNet		96.84 131	96.58 141	97.65 101	99.18 64	93.78 142	98.68 12	96.34 272	97.91 47	97.30 161	98.06 132	88.46 257	99.85 24	93.85 180	99.40 151	99.32 108

abl_6		98.42 27	98.19 42	99.09 4	99.16 65	98.10 5	97.73 70	99.11 23	97.76 51	98.62 58	98.27 106	97.88 21	99.80 37	95.67 107	99.50 113	99.38 97

XVG-ACMP-BASELINE		97.58 85	97.28 96	98.49 47	99.16 65	96.90 42	96.39 133	98.98 67	95.05 163	98.06 112	98.02 136	95.86 87	99.56 153	94.37 161	99.64 78	99.00 159

CHOSEN 1792x2688		94.10 239	93.41 244	96.18 197	99.16 65	90.04 210	92.15 313	98.68 126	79.90 340	96.22 215	97.83 153	87.92 264	99.42 199	89.18 265	99.65 77	99.08 151

HFP-MVS		97.94 53	97.64 73	98.83 25	99.15 68	97.50 25	97.59 79	98.84 88	96.05 115	97.49 151	97.54 178	97.07 45	99.70 89	95.61 113	99.46 127	99.30 112

#test#		97.62 81	97.22 104	98.83 25	99.15 68	97.50 25	96.81 121	98.84 88	94.25 190	97.49 151	97.54 178	97.07 45	99.70 89	94.37 161	99.46 127	99.30 112

XVS		97.96 49	97.63 75	98.94 15	99.15 68	97.66 16	97.77 63	98.83 96	97.42 73	96.32 208	97.64 170	96.49 72	99.72 71	95.66 109	99.37 153	99.45 72

X-MVStestdata		92.86 261	90.83 295	98.94 15	99.15 68	97.66 16	97.77 63	98.83 96	97.42 73	96.32 208	36.50 358	96.49 72	99.72 71	95.66 109	99.37 153	99.45 72

LPG-MVS_test		97.94 53	97.67 69	98.74 32	99.15 68	97.02 38	97.09 106	99.02 51	95.15 155	98.34 81	98.23 108	97.91 19	99.70 89	94.41 158	99.73 57	99.50 51

LGP-MVS_train				98.74 32	99.15 68	97.02 38		99.02 51	95.15 155	98.34 81	98.23 108	97.91 19	99.70 89	94.41 158	99.73 57	99.50 51

RPSCF		97.87 62	97.51 85	98.95 14	99.15 68	98.43 3	97.56 81	99.06 36	96.19 113	98.48 70	98.70 69	94.72 125	99.24 247	94.37 161	99.33 166	99.17 131

ACMM	93.33 11	98.05 43	97.79 60	98.85 24	99.15 68	97.55 23	96.68 127	98.83 96	95.21 150	98.36 79	98.13 123	98.13 16	99.62 129	96.04 94	99.54 104	99.39 95
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
FMVSNet1		97.95 51	98.08 47	97.56 105	99.14 76	93.67 144	98.23 35	98.66 131	97.41 80	99.00 41	99.19 32	95.47 105	99.73 65	95.83 103	99.76 51	99.30 112

Vis-MVSNet (Re-imp)		95.11 205	94.85 203	95.87 219	99.12 77	89.17 233	97.54 84	94.92 291	96.50 101	96.58 192	97.27 198	83.64 285	99.48 182	88.42 277	99.67 74	98.97 163

OPM-MVS		97.54 88	97.25 97	98.41 52	99.11 78	96.61 51	95.24 212	98.46 152	94.58 178	98.10 106	98.07 129	97.09 44	99.39 219	95.16 133	99.44 132	99.21 126

UA-Net		98.88 7	98.76 16	99.22 2	99.11 78	97.89 10	99.47 3	99.32 8	99.08 9	97.87 139	99.67 3	96.47 74	99.92 4	97.88 36	99.98 3	99.85 4

AllTest		97.20 110	96.92 125	98.06 76	99.08 80	96.16 62	97.14 99	99.16 16	94.35 187	97.78 144	98.07 129	95.84 88	99.12 256	91.41 218	99.42 144	98.91 175

TestCases				98.06 76	99.08 80	96.16 62		99.16 16	94.35 187	97.78 144	98.07 129	95.84 88	99.12 256	91.41 218	99.42 144	98.91 175

TranMVSNet+NR-MVSNet		98.33 30	98.30 40	98.43 51	99.07 82	95.87 70	96.73 125	99.05 38	98.67 21	98.84 46	98.45 87	97.58 27	99.88 19	96.45 84	99.86 39	99.54 46

VPA-MVSNet		98.27 33	98.46 30	97.70 96	99.06 83	93.80 140	97.76 65	99.00 62	98.40 29	99.07 36	98.98 51	96.89 50	99.75 55	97.19 67	99.79 48	99.55 45

114514_t		93.96 243	93.22 248	96.19 196	99.06 83	90.97 197	95.99 158	98.94 73	73.88 354	93.43 302	96.93 216	92.38 204	99.37 227	89.09 266	99.28 173	98.25 240

EG-PatchMatch MVS		97.69 77	97.79 60	97.40 126	99.06 83	93.52 152	95.96 164	98.97 69	94.55 179	98.82 47	98.76 64	97.31 35	99.29 241	97.20 66	99.44 132	99.38 97

ACMP	92.54 13	97.47 92	97.10 114	98.55 46	99.04 86	96.70 48	96.24 146	98.89 78	93.71 209	97.97 121	97.75 162	97.44 29	99.63 123	93.22 192	99.70 67	99.32 108
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
test_part2					99.03 87	96.07 65				98.08 109

ESAPD		97.22 109	96.82 130	98.40 54	99.03 87	96.07 65	95.64 184	98.84 88	94.84 167	98.08 109	97.60 174	96.69 61	99.76 49	91.22 224	99.44 132	99.37 102

v13		98.02 45	98.52 28	96.51 173	99.02 89	90.14 208	98.07 47	99.09 29	98.10 41	99.13 33	99.35 18	94.84 123	99.74 60	99.12 5	99.98 3	99.65 25

XVG-OURS-SEG-HR		97.38 97	97.07 117	98.30 65	99.01 90	97.41 31	94.66 238	99.02 51	95.20 151	98.15 100	97.52 181	98.83 5	98.43 322	94.87 142	96.41 316	99.07 153

XVG-OURS		97.12 111	96.74 135	98.26 67	98.99 91	97.45 29	93.82 274	99.05 38	95.19 152	98.32 84	97.70 168	95.22 115	98.41 323	94.27 166	98.13 256	98.93 171

CP-MVS		97.92 56	97.56 82	98.99 10	98.99 91	97.82 12	97.93 55	98.96 70	96.11 114	96.89 185	97.45 186	96.85 54	99.78 40	95.19 129	99.63 79	99.38 97

v12		97.97 48	98.47 29	96.46 177	98.98 93	90.01 212	97.97 52	99.08 30	98.00 44	99.11 35	99.34 20	94.70 126	99.73 65	99.07 6	99.98 3	99.64 28

CSCG		97.40 96	97.30 93	97.69 98	98.95 94	94.83 105	97.28 92	98.99 65	96.35 108	98.13 102	95.95 269	95.99 84	99.66 116	94.36 164	99.73 57	98.59 208

v11		97.82 68	98.36 35	96.17 198	98.93 95	89.16 234	97.79 62	99.08 30	97.64 62	99.19 30	99.32 22	94.28 145	99.72 71	99.07 6	99.97 8	99.63 30

V9		97.90 59	98.40 33	96.40 181	98.93 95	89.86 214	97.86 59	99.07 34	97.88 48	99.05 37	99.30 23	94.53 137	99.72 71	99.01 8	99.98 3	99.63 30

HyFIR lowres test		93.72 247	92.65 258	96.91 152	98.93 95	91.81 187	91.23 327	98.52 147	82.69 328	96.46 199	96.52 243	80.38 294	99.90 13	90.36 250	98.79 220	99.03 157

testing_2		97.43 93	97.71 65	96.60 165	98.91 98	90.85 198	96.01 157	98.54 145	94.78 171	98.78 49	98.96 53	96.35 78	99.54 159	97.25 62	99.82 43	99.40 92

PM-MVS		97.36 101	97.10 114	98.14 73	98.91 98	96.77 45	96.20 148	98.63 138	93.82 206	98.54 65	98.33 97	93.98 155	99.05 266	95.99 98	99.45 131	98.61 207

CPTT-MVS		96.69 145	96.08 163	98.49 47	98.89 100	96.64 50	97.25 93	98.77 107	92.89 233	96.01 222	97.13 203	92.23 205	99.67 111	92.24 204	99.34 161	99.17 131

V14		97.83 65	98.33 37	96.35 182	98.88 101	89.72 215	97.75 66	99.05 38	97.74 52	99.01 39	99.27 25	94.35 142	99.71 81	98.95 9	99.97 8	99.62 32

SMA-MVS		97.55 86	97.19 106	98.61 42	98.83 102	96.71 46	96.74 123	98.81 102	91.81 252	98.78 49	98.36 93	96.63 65	99.68 104	95.17 131	99.59 90	99.45 72

v15		97.77 71	98.26 41	96.30 187	98.81 103	89.59 222	97.62 75	99.04 45	97.59 64	98.97 43	99.24 27	94.19 149	99.70 89	98.88 11	99.97 8	99.61 34

UniMVSNet (Re)		97.83 65	97.65 71	98.35 61	98.80 104	95.86 71	95.92 168	99.04 45	97.51 70	98.22 93	97.81 157	94.68 129	99.78 40	97.14 69	99.75 55	99.41 89

APD-MVS_3200maxsize		98.13 40	97.90 55	98.79 28	98.79 105	97.31 33	97.55 82	98.92 74	97.72 56	98.25 91	98.13 123	97.10 43	99.75 55	95.44 119	99.24 177	99.32 108

DeepC-MVS	95.41 4	97.82 68	97.70 66	98.16 71	98.78 106	95.72 74	96.23 147	99.02 51	93.92 199	98.62 58	98.99 50	97.69 23	99.62 129	96.18 89	99.87 37	99.15 135
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
v17		97.70 76	98.17 43	96.28 190	98.77 107	89.59 222	97.62 75	99.01 60	97.54 67	98.72 55	99.18 35	94.06 153	99.68 104	98.74 16	99.92 24	99.58 37

v16		97.69 77	98.16 44	96.29 189	98.75 108	89.60 220	97.62 75	99.01 60	97.53 69	98.69 57	99.18 35	94.05 154	99.68 104	98.73 17	99.88 34	99.58 37

MCST-MVS		96.24 162	95.80 174	97.56 105	98.75 108	94.13 129	94.66 238	98.17 196	90.17 268	96.21 216	96.10 263	95.14 116	99.43 198	94.13 169	98.85 219	99.13 138

DU-MVS		97.79 70	97.60 78	98.36 59	98.73 110	95.78 72	95.65 182	98.87 82	97.57 65	98.31 86	97.83 153	94.69 127	99.85 24	97.02 72	99.71 64	99.46 67

NR-MVSNet		97.96 49	97.86 57	98.26 67	98.73 110	95.54 82	98.14 43	98.73 114	97.79 49	99.42 16	97.83 153	94.40 141	99.78 40	95.91 102	99.76 51	99.46 67

Anonymous20231206		95.27 201	95.06 195	95.88 218	98.72 112	89.37 230	95.70 176	97.85 217	88.00 289	96.98 176	97.62 172	91.95 214	99.34 231	89.21 264	99.53 106	98.94 167

APDe-MVS		98.14 38	98.03 51	98.47 49	98.72 112	96.04 67	98.07 47	99.10 25	95.96 121	98.59 62	98.69 70	96.94 48	99.81 33	96.64 76	99.58 93	99.57 41

UniMVSNet_NR-MVSNet		97.83 65	97.65 71	98.37 56	98.72 112	95.78 72	95.66 180	99.02 51	98.11 40	98.31 86	97.69 169	94.65 131	99.85 24	97.02 72	99.71 64	99.48 62

v8		97.60 83	98.06 49	96.23 191	98.71 115	89.44 226	97.43 88	98.82 100	97.29 85	98.74 53	99.10 45	93.86 157	99.68 104	98.61 22	99.94 19	99.56 42

v6		96.97 118	97.24 99	96.15 199	98.71 115	89.44 226	95.97 160	98.33 173	95.25 147	97.89 132	98.15 119	93.86 157	99.61 135	97.51 54	99.50 113	99.42 87

v1neww		96.97 118	97.24 99	96.15 199	98.70 117	89.44 226	95.97 160	98.33 173	95.25 147	97.88 134	98.15 119	93.83 160	99.61 135	97.50 55	99.50 113	99.41 89

v7new		96.97 118	97.24 99	96.15 199	98.70 117	89.44 226	95.97 160	98.33 173	95.25 147	97.88 134	98.15 119	93.83 160	99.61 135	97.50 55	99.50 113	99.41 89

HQP_MVS		96.66 147	96.33 156	97.68 99	98.70 117	94.29 122	96.50 130	98.75 111	96.36 106	96.16 218	96.77 227	91.91 218	99.46 189	92.59 200	99.20 180	99.28 119

plane_prior7					98.70 117	94.67 112

VDD-MVS		97.37 98	97.25 97	97.74 94	98.69 121	94.50 117	97.04 109	95.61 287	98.59 23	98.51 67	98.72 67	92.54 198	99.58 147	96.02 96	99.49 120	99.12 143

v18		97.60 83	98.06 49	96.23 191	98.68 122	89.46 225	97.48 86	98.98 67	97.33 83	98.60 61	99.13 43	93.86 157	99.67 111	98.62 21	99.87 37	99.56 42

HPM-MVS++		96.99 114	96.38 152	98.81 27	98.64 123	97.59 20	95.97 160	98.20 191	95.51 138	95.06 244	96.53 241	94.10 152	99.70 89	94.29 165	99.15 184	99.13 138

ab-mvs		96.59 149	96.59 140	96.60 165	98.64 123	92.21 173	98.35 29	97.67 229	94.45 180	96.99 175	98.79 61	94.96 121	99.49 179	90.39 249	99.07 195	98.08 250

F-COLMAP		95.30 199	94.38 224	98.05 79	98.64 123	96.04 67	95.61 188	98.66 131	89.00 276	93.22 307	96.40 251	92.90 186	99.35 230	87.45 296	97.53 291	98.77 195

ITE_SJBPF				97.85 88	98.64 123	96.66 49		98.51 149	95.63 132	97.22 163	97.30 197	95.52 102	98.55 317	90.97 229	98.90 210	98.34 231

v148		96.58 150	96.97 121	95.42 232	98.63 127	87.57 272	95.09 218	97.90 214	95.91 123	98.24 92	97.96 141	93.42 171	99.39 219	96.04 94	99.52 110	99.29 118

UnsupCasMVSNet_bld		94.72 219	94.26 226	96.08 204	98.62 128	90.54 207	93.38 290	98.05 210	90.30 266	97.02 173	96.80 225	89.54 246	99.16 255	88.44 276	96.18 319	98.56 210

DP-MVS		97.87 62	97.89 56	97.81 90	98.62 128	94.82 106	97.13 100	98.79 103	98.98 16	98.74 53	98.49 84	95.80 97	99.49 179	95.04 140	99.44 132	99.11 146

v10		97.55 86	97.97 52	96.31 186	98.60 130	89.64 217	97.44 87	99.02 51	96.60 98	98.72 55	99.16 40	93.48 169	99.72 71	98.76 15	99.92 24	99.58 37

Test_1112_low_res		93.53 253	92.86 253	95.54 229	98.60 130	88.86 243	92.75 302	98.69 124	82.66 329	92.65 316	96.92 217	84.75 282	99.56 153	90.94 230	97.76 269	98.19 245

v7		96.93 122	97.17 108	96.23 191	98.59 132	89.64 217	95.96 164	98.66 131	94.41 183	97.87 139	98.38 92	93.47 170	99.64 120	97.93 35	99.24 177	99.43 85

V42		97.04 112	97.16 109	96.68 163	98.59 132	91.05 195	96.33 140	98.36 168	94.60 175	97.99 117	98.30 102	93.32 175	99.62 129	97.40 60	99.53 106	99.38 97

1112_ss		94.12 238	93.42 243	96.23 191	98.59 132	90.85 198	94.24 251	98.85 85	85.49 311	92.97 309	94.94 289	86.01 275	99.64 120	91.78 212	97.92 265	98.20 244

v2v482		96.78 138	97.06 118	95.95 214	98.57 135	88.77 247	95.36 201	98.26 185	95.18 153	97.85 141	98.23 108	92.58 195	99.63 123	97.80 40	99.69 68	99.45 72

WR-MVS		96.90 126	96.81 131	97.16 136	98.56 136	92.20 175	94.33 246	98.12 202	97.34 82	98.20 95	97.33 196	92.81 187	99.75 55	94.79 147	99.81 44	99.54 46

v1141		96.86 128	97.14 111	96.04 206	98.55 137	89.06 237	95.44 191	98.33 173	95.14 157	97.93 127	98.19 113	93.36 173	99.62 129	97.61 47	99.69 68	99.44 81

v1		96.86 128	97.14 111	96.04 206	98.55 137	89.06 237	95.44 191	98.33 173	95.14 157	97.94 124	98.18 117	93.39 172	99.61 135	97.61 47	99.69 68	99.44 81

APD-MVS		97.00 113	96.53 147	98.41 52	98.55 137	96.31 59	96.32 141	98.77 107	92.96 232	97.44 157	97.58 177	95.84 88	99.74 60	91.96 206	99.35 159	99.19 128
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
Patchmatch-RL test		94.66 221	94.49 218	95.19 238	98.54 140	88.91 241	92.57 306	98.74 113	91.46 255	98.32 84	97.75 162	77.31 308	98.81 296	96.06 92	99.61 85	97.85 265

divwei89l23v2f112		96.86 128	97.14 111	96.04 206	98.54 140	89.06 237	95.44 191	98.33 173	95.14 157	97.93 127	98.19 113	93.36 173	99.61 135	97.61 47	99.68 72	99.44 81

1111		88.78 315	89.39 308	86.96 340	98.53 142	62.84 359	91.49 322	97.48 243	94.45 180	96.56 194	96.45 246	43.83 365	98.87 290	86.33 303	99.40 151	99.18 130

.test1245		73.49 333	79.27 334	56.15 346	98.53 142	62.84 359	91.49 322	97.48 243	94.45 180	96.56 194	96.45 246	43.83 365	98.87 290	86.33 303	8.32 360	6.75 360

IterMVS-LS		96.92 124	97.29 94	95.79 221	98.51 144	88.13 257	95.10 216	98.66 131	96.99 86	98.46 72	98.68 71	92.55 196	99.74 60	96.91 74	99.79 48	99.50 51
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
DP-MVS Recon		95.55 183	95.13 191	96.80 155	98.51 144	93.99 134	94.60 240	98.69 124	90.20 267	95.78 230	96.21 259	92.73 190	98.98 276	90.58 243	98.86 217	97.42 280

test20.03		96.58 150	96.61 139	96.48 176	98.49 146	91.72 188	95.68 179	97.69 228	96.81 94	98.27 90	97.92 148	94.18 150	98.71 304	90.78 236	99.66 76	99.00 159

plane_prior1					98.49 146

MDA-MVSNet-bldmvs		95.69 177	95.67 177	95.74 222	98.48 148	88.76 248	92.84 299	97.25 249	96.00 119	97.59 146	97.95 144	91.38 226	99.46 189	93.16 194	96.35 317	98.99 162

UnsupCasMVSNet_eth		95.91 173	95.73 176	96.44 178	98.48 148	91.52 191	95.31 206	98.45 153	95.76 129	97.48 154	97.54 178	89.53 248	98.69 306	94.43 157	94.61 333	99.13 138

view600		92.56 265	92.11 265	93.91 278	98.45 150	84.76 307	97.10 102	90.23 338	97.42 73	96.98 176	94.48 298	73.62 324	99.60 141	82.49 328	98.28 248	97.36 281

view800		92.56 265	92.11 265	93.91 278	98.45 150	84.76 307	97.10 102	90.23 338	97.42 73	96.98 176	94.48 298	73.62 324	99.60 141	82.49 328	98.28 248	97.36 281

conf0.05thres1000		92.56 265	92.11 265	93.91 278	98.45 150	84.76 307	97.10 102	90.23 338	97.42 73	96.98 176	94.48 298	73.62 324	99.60 141	82.49 328	98.28 248	97.36 281

tfpn		92.56 265	92.11 265	93.91 278	98.45 150	84.76 307	97.10 102	90.23 338	97.42 73	96.98 176	94.48 298	73.62 324	99.60 141	82.49 328	98.28 248	97.36 281

v1144		96.84 131	97.08 116	96.13 203	98.42 154	89.28 232	95.41 198	98.67 129	94.21 192	97.97 121	98.31 99	93.06 180	99.65 117	98.06 32	99.62 80	99.45 72

plane_prior6					98.38 155	94.37 121						91.91 218

FPMVS		89.92 309	88.63 316	93.82 283	98.37 156	96.94 41	91.58 321	93.34 308	88.00 289	90.32 337	97.10 206	70.87 339	91.13 357	71.91 353	96.16 320	93.39 346

PAPM_NR		94.61 224	94.17 231	95.96 212	98.36 157	91.23 193	95.93 167	97.95 212	92.98 227	93.42 303	94.43 303	90.53 234	98.38 327	87.60 293	96.29 318	98.27 238

MVS_111021_HR		96.73 141	96.54 146	97.27 132	98.35 158	93.66 147	93.42 288	98.36 168	94.74 172	96.58 192	96.76 229	96.54 68	98.99 274	94.87 142	99.27 175	99.15 135

TAMVS		95.49 186	94.94 198	97.16 136	98.31 159	93.41 154	95.07 221	96.82 266	91.09 258	97.51 149	97.82 156	89.96 243	99.42 199	88.42 277	99.44 132	98.64 203

OMC-MVS		96.48 154	96.00 166	97.91 85	98.30 160	96.01 69	94.86 232	98.60 141	91.88 250	97.18 165	97.21 200	96.11 82	99.04 267	90.49 247	99.34 161	98.69 201

新几何1				97.25 135	98.29 161	94.70 111		97.73 225	77.98 347	94.83 251	96.67 234	92.08 210	99.45 193	88.17 281	98.65 233	97.61 274

jason		94.39 229	94.04 234	95.41 234	98.29 161	87.85 268	92.74 304	96.75 268	85.38 316	95.29 240	96.15 260	88.21 260	99.65 117	94.24 167	99.34 161	98.74 197
jason: jason.
no-one		94.84 214	94.76 207	95.09 242	98.29 161	87.49 274	91.82 319	97.49 241	88.21 285	97.84 142	98.75 65	91.51 223	99.27 243	88.96 269	99.99 2	98.52 213

v1192		96.83 134	97.06 118	96.15 199	98.28 164	89.29 231	95.36 201	98.77 107	93.73 208	98.11 103	98.34 95	93.02 185	99.67 111	98.35 27	99.58 93	99.50 51

CDPH-MVS		95.45 192	94.65 210	97.84 89	98.28 164	94.96 102	93.73 277	98.33 173	85.03 318	95.44 237	96.60 237	95.31 111	99.44 197	90.01 254	99.13 187	99.11 146

conf0.01		91.90 281	90.98 286	94.67 256	98.27 166	88.03 259	96.98 112	88.58 346	93.90 200	94.64 255	91.45 334	69.62 343	99.52 165	87.62 287	97.74 270	96.46 312

conf0.002		91.90 281	90.98 286	94.67 256	98.27 166	88.03 259	96.98 112	88.58 346	93.90 200	94.64 255	91.45 334	69.62 343	99.52 165	87.62 287	97.74 270	96.46 312

thresconf0.02		91.72 288	90.98 286	93.97 274	98.27 166	88.03 259	96.98 112	88.58 346	93.90 200	94.64 255	91.45 334	69.62 343	99.52 165	87.62 287	97.74 270	94.35 338

tfpn_n400		91.72 288	90.98 286	93.97 274	98.27 166	88.03 259	96.98 112	88.58 346	93.90 200	94.64 255	91.45 334	69.62 343	99.52 165	87.62 287	97.74 270	94.35 338

tfpnconf		91.72 288	90.98 286	93.97 274	98.27 166	88.03 259	96.98 112	88.58 346	93.90 200	94.64 255	91.45 334	69.62 343	99.52 165	87.62 287	97.74 270	94.35 338

tfpnview11		91.72 288	90.98 286	93.97 274	98.27 166	88.03 259	96.98 112	88.58 346	93.90 200	94.64 255	91.45 334	69.62 343	99.52 165	87.62 287	97.74 270	94.35 338

MVS_111021_LR		96.82 135	96.55 144	97.62 102	98.27 166	95.34 89	93.81 275	98.33 173	94.59 177	96.56 194	96.63 236	96.61 66	98.73 302	94.80 146	99.34 161	98.78 194

CLD-MVS		95.47 189	95.07 193	96.69 162	98.27 166	92.53 166	91.36 325	98.67 129	91.22 257	95.78 230	94.12 307	95.65 100	98.98 276	90.81 234	99.72 60	98.57 209
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
1121		94.26 230	93.26 246	97.27 132	98.26 174	94.73 108	95.86 169	97.71 227	77.96 348	94.53 263	96.71 231	91.93 216	99.40 213	87.71 283	98.64 234	97.69 271

Anonymous202405211		96.34 160	95.98 168	97.43 123	98.25 175	93.85 138	96.74 123	94.41 296	97.72 56	98.37 77	98.03 135	87.15 270	99.53 161	94.06 171	99.07 195	98.92 174

pmmvs-eth3d		96.49 153	96.18 159	97.42 124	98.25 175	94.29 122	94.77 237	98.07 208	89.81 271	97.97 121	98.33 97	93.11 179	99.08 263	95.46 118	99.84 41	98.89 179

v144192		96.69 145	96.90 126	96.03 209	98.25 175	88.92 240	95.49 189	98.77 107	93.05 225	98.09 107	98.29 103	92.51 200	99.70 89	98.11 30	99.56 98	99.47 65

ambc				96.56 172	98.23 178	91.68 189	97.88 58	98.13 201		98.42 75	98.56 79	94.22 148	99.04 267	94.05 174	99.35 159	98.95 165

testmv		95.51 184	95.33 186	96.05 205	98.23 178	89.51 224	93.50 286	98.63 138	94.25 190	98.22 93	97.73 165	92.51 200	99.47 184	85.22 313	99.72 60	99.17 131

tfpn111		91.92 280	91.39 275	93.49 290	98.21 180	84.50 312	96.39 133	90.39 333	96.87 90	96.33 204	93.08 315	73.44 330	99.51 175	79.87 336	97.94 264	96.46 312

conf200view11		91.81 285	91.26 280	93.46 291	98.21 180	84.50 312	96.39 133	90.39 333	96.87 90	96.33 204	93.08 315	73.44 330	99.42 199	78.85 341	97.74 270	96.46 312

thres100view900		91.76 287	91.26 280	93.26 295	98.21 180	84.50 312	96.39 133	90.39 333	96.87 90	96.33 204	93.08 315	73.44 330	99.42 199	78.85 341	97.74 270	95.85 322

v1921920		96.72 142	96.96 123	95.99 210	98.21 180	88.79 246	95.42 196	98.79 103	93.22 217	98.19 96	98.26 107	92.68 191	99.70 89	98.34 28	99.55 102	99.49 59

thres600view7		92.03 278	91.43 274	93.82 283	98.19 184	84.61 311	96.27 142	90.39 333	96.81 94	96.37 203	93.11 313	73.44 330	99.49 179	80.32 335	97.95 261	97.36 281

PatchMatch-RL		94.61 224	93.81 238	97.02 147	98.19 184	95.72 74	93.66 279	97.23 250	88.17 286	94.94 248	95.62 277	91.43 225	98.57 314	87.36 297	97.68 283	96.76 303

LF4IMVS		96.07 168	95.63 179	97.36 128	98.19 184	95.55 81	95.44 191	98.82 100	92.29 242	95.70 234	96.55 239	92.63 194	98.69 306	91.75 215	99.33 166	97.85 265

v1240		96.74 139	97.02 120	95.91 217	98.18 187	88.52 249	95.39 199	98.88 80	93.15 223	98.46 72	98.40 91	92.80 188	99.71 81	98.45 26	99.49 120	99.49 59

TAPA-MVS	93.32 12	94.93 212	94.23 227	97.04 144	98.18 187	94.51 115	95.22 213	98.73 114	81.22 335	96.25 214	95.95 269	93.80 163	98.98 276	89.89 255	98.87 215	97.62 273
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
test222					98.17 189	93.24 158	92.74 304	97.61 239	75.17 352	94.65 254	96.69 233	90.96 231			98.66 232	97.66 272

MIMVSNet		93.42 254	92.86 253	95.10 241	98.17 189	88.19 254	98.13 44	93.69 301	92.07 243	95.04 246	98.21 112	80.95 292	99.03 270	81.42 333	98.06 258	98.07 252

原ACMM1				96.58 169	98.16 191	92.12 177		98.15 199	85.90 308	93.49 298	96.43 248	92.47 202	99.38 224	87.66 286	98.62 235	98.23 241

testdata				95.70 225	98.16 191	90.58 204		97.72 226	80.38 338	95.62 235	97.02 210	92.06 212	98.98 276	89.06 268	98.52 240	97.54 277

MVP-Stereo		95.69 177	95.28 187	96.92 150	98.15 193	93.03 160	95.64 184	98.20 191	90.39 265	96.63 191	97.73 165	91.63 221	99.10 261	91.84 211	97.31 300	98.63 205
Qingsong Yan: MVP-Stereo: A Parallel Multi-View Patchmatch Stereo Method with Dilation Matching for Photogrammetric Application.
SD-MVS		97.37 98	97.70 66	96.35 182	98.14 194	95.13 97	96.54 128	98.92 74	95.94 122	99.19 30	98.08 128	97.74 22	95.06 354	95.24 127	99.54 104	98.87 185

EU-MVSNet		94.25 231	94.47 219	93.60 287	98.14 194	82.60 323	97.24 95	92.72 316	85.08 317	98.48 70	98.94 55	82.59 288	98.76 300	97.47 58	99.53 106	99.44 81

NP-MVS					98.14 194	93.72 143					95.08 285

LCM-MVSNet-Re		97.33 102	97.33 92	97.32 130	98.13 197	93.79 141	96.99 111	99.65 2	96.74 96	99.47 13	98.93 56	96.91 49	99.84 28	90.11 252	99.06 198	98.32 232

3Dnovator+	96.13 3	97.73 73	97.59 79	98.15 72	98.11 198	95.60 79	98.04 49	98.70 123	98.13 39	96.93 183	98.45 87	95.30 112	99.62 129	95.64 111	98.96 204	99.24 124

Test4		95.39 194	95.24 188	95.82 220	98.07 199	89.60 220	94.40 244	98.49 150	91.39 256	97.40 159	96.32 254	87.32 269	99.41 210	95.09 139	98.71 230	98.44 219

VNet		96.84 131	96.83 129	96.88 153	98.06 200	92.02 180	96.35 139	97.57 240	97.70 58	97.88 134	97.80 158	92.40 203	99.54 159	94.73 151	98.96 204	99.08 151

DI_MVS_plusplus_test		95.46 190	95.43 184	95.55 228	98.05 201	88.84 244	94.18 256	95.75 283	91.92 249	97.32 160	96.94 214	91.44 224	99.39 219	94.81 145	98.48 243	98.43 220

LFMVS		95.32 198	94.88 202	96.62 164	98.03 202	91.47 192	97.65 72	90.72 332	99.11 8	97.89 132	98.31 99	79.20 297	99.48 182	93.91 179	99.12 190	98.93 171

tfpn200view9		91.55 293	91.00 284	93.21 297	98.02 203	84.35 316	95.70 176	90.79 330	96.26 110	95.90 227	92.13 330	73.62 324	99.42 199	78.85 341	97.74 270	95.85 322

thres400		91.68 292	91.00 284	93.71 285	98.02 203	84.35 316	95.70 176	90.79 330	96.26 110	95.90 227	92.13 330	73.62 324	99.42 199	78.85 341	97.74 270	97.36 281

xiu_mvs_v1_base_debu		95.62 180	95.96 169	94.60 260	98.01 205	88.42 250	93.99 266	98.21 188	92.98 227	95.91 224	94.53 295	96.39 75	99.72 71	95.43 121	98.19 253	95.64 326

xiu_mvs_v1_base		95.62 180	95.96 169	94.60 260	98.01 205	88.42 250	93.99 266	98.21 188	92.98 227	95.91 224	94.53 295	96.39 75	99.72 71	95.43 121	98.19 253	95.64 326

xiu_mvs_v1_base_debi		95.62 180	95.96 169	94.60 260	98.01 205	88.42 250	93.99 266	98.21 188	92.98 227	95.91 224	94.53 295	96.39 75	99.72 71	95.43 121	98.19 253	95.64 326

CNVR-MVS		96.92 124	96.55 144	98.03 80	98.00 208	95.54 82	94.87 231	98.17 196	94.60 175	96.38 202	97.05 208	95.67 99	99.36 228	95.12 137	99.08 193	99.19 128

PLC	91.02 16	94.05 242	92.90 252	97.51 110	98.00 208	95.12 98	94.25 250	98.25 186	86.17 304	91.48 328	95.25 283	91.01 229	99.19 251	85.02 315	96.69 312	98.22 242
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
test_normal		95.51 184	95.46 183	95.68 226	97.97 210	89.12 236	93.73 277	95.86 281	91.98 246	97.17 166	96.94 214	91.55 222	99.42 199	95.21 128	98.73 228	98.51 214

tfpn1000		91.88 284	91.20 282	93.89 282	97.96 211	87.13 282	97.13 100	88.16 353	94.41 183	94.87 250	92.77 322	68.34 350	99.47 184	89.24 263	97.95 261	95.06 332

GBi-Net		96.99 114	96.80 132	97.56 105	97.96 211	93.67 144	98.23 35	98.66 131	95.59 135	97.99 117	99.19 32	89.51 249	99.73 65	94.60 153	99.44 132	99.30 112

test1		96.99 114	96.80 132	97.56 105	97.96 211	93.67 144	98.23 35	98.66 131	95.59 135	97.99 117	99.19 32	89.51 249	99.73 65	94.60 153	99.44 132	99.30 112

FMVSNet2		96.72 142	96.67 138	96.87 154	97.96 211	91.88 184	97.15 97	98.06 209	95.59 135	98.50 69	98.62 75	89.51 249	99.65 117	94.99 141	99.60 88	99.07 153

BH-untuned		94.69 220	94.75 208	94.52 265	97.95 215	87.53 273	94.07 263	97.01 259	93.99 197	97.10 170	95.65 275	92.65 193	98.95 281	87.60 293	96.74 311	97.09 289

QAPM		95.88 175	95.57 180	96.80 155	97.90 216	91.84 186	98.18 42	98.73 114	88.41 281	96.42 200	98.13 123	94.73 124	99.75 55	88.72 272	98.94 208	98.81 190

TinyColmap		96.00 171	96.34 155	94.96 246	97.90 216	87.91 266	94.13 261	98.49 150	94.41 183	98.16 98	97.76 159	96.29 80	98.68 309	90.52 244	99.42 144	98.30 235

HQP-NCC					97.85 218		94.26 247		93.18 219	92.86 311

ACMP_Plane					97.85 218		94.26 247		93.18 219	92.86 311

N_pmnet		95.18 203	94.23 227	98.06 76	97.85 218	96.55 53	92.49 308	91.63 324	89.34 273	98.09 107	97.41 188	90.33 237	99.06 265	91.58 217	99.31 168	98.56 210

HQP-MVS		95.17 204	94.58 215	96.92 150	97.85 218	92.47 167	94.26 247	98.43 157	93.18 219	92.86 311	95.08 285	90.33 237	99.23 249	90.51 245	98.74 225	99.05 156

TEST9					97.84 222	95.23 91	93.62 281	98.39 164	86.81 299	93.78 285	95.99 264	94.68 129	99.52 165

train_agg		95.46 190	94.66 209	97.88 86	97.84 222	95.23 91	93.62 281	98.39 164	87.04 297	93.78 285	95.99 264	94.58 134	99.52 165	91.76 213	98.90 210	98.89 179

MSLP-MVS++		96.42 159	96.71 136	95.57 227	97.82 224	90.56 206	95.71 175	98.84 88	94.72 173	96.71 189	97.39 192	94.91 122	98.10 338	95.28 125	99.02 200	98.05 255

test_8					97.81 225	95.07 99	93.54 284	98.38 166	87.04 297	93.71 289	95.96 268	94.58 134	99.52 165

NCCC		96.52 152	95.99 167	98.10 74	97.81 225	95.68 76	95.00 227	98.20 191	95.39 143	95.40 239	96.36 252	93.81 162	99.45 193	93.55 187	98.42 245	99.17 131

WTY-MVS		93.55 252	93.00 251	95.19 238	97.81 225	87.86 267	93.89 271	96.00 276	89.02 275	94.07 276	95.44 281	86.27 273	99.33 234	87.69 285	96.82 308	98.39 223

CNLPA		95.04 208	94.47 219	96.75 158	97.81 225	95.25 90	94.12 262	97.89 215	94.41 183	94.57 261	95.69 273	90.30 240	98.35 330	86.72 302	98.76 223	96.64 307

agg_prior3		95.30 199	94.46 222	97.80 91	97.80 229	95.00 100	93.63 280	98.34 172	86.33 303	93.40 305	95.84 271	94.15 151	99.50 177	91.76 213	98.90 210	98.89 179

agg_prior1		95.39 194	94.60 213	97.75 93	97.80 229	94.96 102	93.39 289	98.36 168	87.20 295	93.49 298	95.97 267	94.65 131	99.53 161	91.69 216	98.86 217	98.77 195

agg_prior					97.80 229	94.96 102		98.36 168		93.49 298			99.53 161

旧先验1					97.80 229	93.87 136		97.75 223			97.04 209	93.57 168			98.68 231	98.72 199

PCF-MVS	89.43 18	92.12 277	90.64 298	96.57 171	97.80 229	93.48 153	89.88 341	98.45 153	74.46 353	96.04 221	95.68 274	90.71 233	99.31 236	73.73 349	99.01 202	96.91 297
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
test_prior3		95.91 173	95.39 185	97.46 119	97.79 234	94.26 125	93.33 292	98.42 160	94.21 192	94.02 278	96.25 256	93.64 166	99.34 231	91.90 207	98.96 204	98.79 192

test_prior				97.46 119	97.79 234	94.26 125		98.42 160					99.34 231			98.79 192

PVSNet_BlendedMVS		95.02 210	94.93 200	95.27 236	97.79 234	87.40 277	94.14 260	98.68 126	88.94 277	94.51 264	98.01 137	93.04 181	99.30 238	89.77 257	99.49 120	99.11 146

PVSNet_Blended		93.96 243	93.65 240	94.91 247	97.79 234	87.40 277	91.43 324	98.68 126	84.50 322	94.51 264	94.48 298	93.04 181	99.30 238	89.77 257	98.61 236	98.02 260

USDC		94.56 226	94.57 217	94.55 264	97.78 238	86.43 290	92.75 302	98.65 137	85.96 306	96.91 184	97.93 147	90.82 232	98.74 301	90.71 239	99.59 90	98.47 216

alignmvs		96.01 170	95.52 181	97.50 113	97.77 239	94.71 110	96.07 153	96.84 264	97.48 71	96.78 188	94.28 306	85.50 277	99.40 213	96.22 88	98.73 228	98.40 221

TSAR-MVS + GP.		96.47 155	96.12 160	97.49 116	97.74 240	95.23 91	94.15 259	96.90 263	93.26 216	98.04 114	96.70 232	94.41 140	98.89 286	94.77 149	99.14 185	98.37 225

3Dnovator	96.53 2	97.61 82	97.64 73	97.50 113	97.74 240	93.65 148	98.49 23	98.88 80	96.86 93	97.11 169	98.55 80	95.82 91	99.73 65	95.94 100	99.42 144	99.13 138

tfpn_ndepth		90.98 299	90.24 304	93.20 299	97.72 242	87.18 281	96.52 129	88.20 352	92.63 236	93.69 291	90.70 347	68.22 351	99.42 199	86.98 299	97.47 295	93.00 348

sss		94.22 232	93.72 239	95.74 222	97.71 243	89.95 213	93.84 273	96.98 260	88.38 284	93.75 287	95.74 272	87.94 261	98.89 286	91.02 227	98.10 257	98.37 225

DeepC-MVS_fast	94.34 7	96.74 139	96.51 149	97.44 122	97.69 244	94.15 128	96.02 156	98.43 157	93.17 222	97.30 161	97.38 194	95.48 104	99.28 242	93.74 183	99.34 161	98.88 183
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
semantic-postprocess				94.85 251	97.68 245	85.53 294		97.63 237	96.99 86	98.36 79	98.54 81	87.44 267	99.75 55	97.07 71	99.08 193	99.27 122

MVSFormer		96.14 167	96.36 154	95.49 231	97.68 245	87.81 269	98.67 13	99.02 51	96.50 101	94.48 266	96.15 260	86.90 271	99.92 4	98.73 17	99.13 187	98.74 197

lupinMVS		93.77 245	93.28 245	95.24 237	97.68 245	87.81 269	92.12 314	96.05 275	84.52 321	94.48 266	95.06 287	86.90 271	99.63 123	93.62 186	99.13 187	98.27 238

Fast-Effi-MVS+		95.49 186	95.07 193	96.75 158	97.67 248	92.82 162	94.22 253	98.60 141	91.61 253	93.42 303	92.90 320	96.73 60	99.70 89	92.60 199	97.89 268	97.74 270

canonicalmvs		97.23 108	97.21 105	97.30 131	97.65 249	94.39 119	97.84 60	99.05 38	97.42 73	96.68 190	93.85 309	97.63 26	99.33 234	96.29 87	98.47 244	98.18 247

CDS-MVSNet		94.88 213	94.12 232	97.14 138	97.64 250	93.57 149	93.96 269	97.06 258	90.05 269	96.30 211	96.55 239	86.10 274	99.47 184	90.10 253	99.31 168	98.40 221
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
pmmvs5		94.63 223	94.34 225	95.50 230	97.63 251	88.34 253	94.02 264	97.13 255	87.15 296	95.22 242	97.15 202	87.50 266	99.27 243	93.99 176	99.26 176	98.88 183

test12				97.46 119	97.61 252	94.07 130		97.78 222		93.57 296		93.31 176	99.42 199		98.78 221	98.89 179

PMMVS2		93.66 249	94.07 233	92.45 312	97.57 253	80.67 331	86.46 348	96.00 276	93.99 197	97.10 170	97.38 194	89.90 244	97.82 341	88.76 271	99.47 125	98.86 186

BH-RMVSNet		94.56 226	94.44 223	94.91 247	97.57 253	87.44 276	93.78 276	96.26 273	93.69 210	96.41 201	96.50 244	92.10 209	99.00 273	85.96 305	97.71 280	98.31 233

PVSNet	86.72 19	91.10 296	90.97 292	91.49 319	97.56 255	78.04 340	87.17 346	94.60 294	84.65 320	92.34 320	92.20 329	87.37 268	98.47 320	85.17 314	97.69 282	97.96 262

DELS-MVS		96.17 166	96.23 158	95.99 210	97.55 256	90.04 210	92.38 311	98.52 147	94.13 195	96.55 197	97.06 207	94.99 120	99.58 147	95.62 112	99.28 173	98.37 225
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
IterMVS		95.42 193	95.83 173	94.20 272	97.52 257	83.78 320	92.41 310	97.47 246	95.49 139	98.06 112	98.49 84	87.94 261	99.58 147	96.02 96	99.02 200	99.23 125
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
casdiffmvs		96.43 158	96.38 152	96.60 165	97.51 258	91.95 183	97.08 108	98.41 162	93.69 210	93.95 282	98.34 95	93.03 183	99.45 193	98.09 31	97.30 301	98.39 223

new-patchmatchnet		95.67 179	96.58 141	92.94 305	97.48 259	80.21 332	92.96 298	98.19 195	94.83 169	98.82 47	98.79 61	93.31 176	99.51 175	95.83 103	99.04 199	99.12 143

MDA-MVSNet_test_wron		94.73 217	94.83 206	94.42 266	97.48 259	85.15 300	90.28 336	95.87 280	92.52 237	97.48 154	97.76 159	91.92 217	99.17 254	93.32 188	96.80 310	98.94 167

PHI-MVS		96.96 121	96.53 147	98.25 69	97.48 259	96.50 54	96.76 122	98.85 85	93.52 213	96.19 217	96.85 220	95.94 85	99.42 199	93.79 182	99.43 141	98.83 189

DeepPCF-MVS	94.58 5	96.90 126	96.43 151	98.31 64	97.48 259	97.23 35	92.56 307	98.60 141	92.84 234	98.54 65	97.40 189	96.64 64	98.78 298	94.40 160	99.41 150	98.93 171

thres200		91.00 298	90.42 302	92.77 307	97.47 263	83.98 319	94.01 265	91.18 328	95.12 160	95.44 237	91.21 342	73.93 320	99.31 236	77.76 345	97.63 288	95.01 333

YYNet1		94.73 217	94.84 204	94.41 267	97.47 263	85.09 302	90.29 335	95.85 282	92.52 237	97.53 148	97.76 159	91.97 213	99.18 252	93.31 189	96.86 307	98.95 165

Effi-MVS+		96.19 165	96.01 165	96.71 160	97.43 265	92.19 176	96.12 152	99.10 25	95.45 140	93.33 306	94.71 293	97.23 41	99.56 153	93.21 193	97.54 290	98.37 225

pmmvs4		94.82 216	94.19 230	96.70 161	97.42 266	92.75 164	92.09 316	96.76 267	86.80 300	95.73 233	97.22 199	89.28 252	98.89 286	93.28 190	99.14 185	98.46 218

MSDG		95.33 197	95.13 191	95.94 216	97.40 267	91.85 185	91.02 328	98.37 167	95.30 145	96.31 210	95.99 264	94.51 138	98.38 327	89.59 259	97.65 286	97.60 275

EI-MVSNet-Vis-set		97.32 103	97.39 90	97.11 139	97.36 268	92.08 179	95.34 203	97.65 233	97.74 52	98.29 89	98.11 126	95.05 117	99.68 104	97.50 55	99.50 113	99.56 42

PS-MVSNAJ		94.10 239	94.47 219	93.00 303	97.35 269	84.88 304	91.86 318	97.84 218	91.96 247	94.17 271	92.50 327	95.82 91	99.71 81	91.27 221	97.48 293	94.40 337

Regformer-3		97.25 107	97.29 94	97.11 139	97.35 269	92.32 170	95.26 210	97.62 238	97.67 61	98.17 97	97.89 150	95.05 117	99.56 153	97.16 68	99.42 144	99.46 67

Regformer-4		97.53 90	97.47 88	97.71 95	97.35 269	93.91 135	95.26 210	98.14 200	97.97 45	98.34 81	97.89 150	95.49 103	99.71 81	97.41 59	99.42 144	99.51 50

EI-MVSNet-UG-set		97.32 103	97.40 89	97.09 141	97.34 272	92.01 181	95.33 204	97.65 233	97.74 52	98.30 88	98.14 122	95.04 119	99.69 98	97.55 51	99.52 110	99.58 37

AdaColmap		95.11 205	94.62 212	96.58 169	97.33 273	94.45 118	94.92 229	98.08 206	93.15 223	93.98 281	95.53 280	94.34 143	99.10 261	85.69 308	98.61 236	96.20 319

xiu_mvs_v2_base		94.22 232	94.63 211	92.99 304	97.32 274	84.84 305	92.12 314	97.84 218	91.96 247	94.17 271	93.43 310	96.07 83	99.71 81	91.27 221	97.48 293	94.42 336

OpenMVS_ROB	91.80 14	93.64 250	93.05 249	95.42 232	97.31 275	91.21 194	95.08 220	96.68 271	81.56 332	96.88 186	96.41 249	90.44 236	99.25 246	85.39 312	97.67 284	95.80 324

EI-MVSNet		96.63 148	96.93 124	95.74 222	97.26 276	88.13 257	95.29 208	97.65 233	96.99 86	97.94 124	98.19 113	92.55 196	99.58 147	96.91 74	99.56 98	99.50 51

CVMVSNet		92.33 273	92.79 255	90.95 325	97.26 276	75.84 347	95.29 208	92.33 319	81.86 330	96.27 212	98.19 113	81.44 290	98.46 321	94.23 168	98.29 247	98.55 212

Regformer-1		97.27 105	97.16 109	97.61 103	97.21 278	93.86 137	94.85 233	98.04 211	97.62 63	98.03 115	97.50 183	95.34 109	99.63 123	96.52 80	99.31 168	99.35 106

Regformer-2		97.41 95	97.24 99	97.93 84	97.21 278	94.72 109	94.85 233	98.27 183	97.74 52	98.11 103	97.50 183	95.58 101	99.69 98	96.57 79	99.31 168	99.37 102

Fast-Effi-MVS+-dtu		96.44 156	96.12 160	97.39 127	97.18 280	94.39 119	95.46 190	98.73 114	96.03 118	94.72 252	94.92 291	96.28 81	99.69 98	93.81 181	97.98 260	98.09 249

OpenMVS	94.22 8	95.48 188	95.20 189	96.32 185	97.16 281	91.96 182	97.74 68	98.84 88	87.26 293	94.36 268	98.01 137	93.95 156	99.67 111	90.70 240	98.75 224	97.35 287

BH-w/o		92.14 276	91.94 269	92.73 308	97.13 282	85.30 297	92.46 309	95.64 286	89.33 274	94.21 270	92.74 324	89.60 245	98.24 333	81.68 332	94.66 332	94.66 335

MG-MVS		94.08 241	94.00 236	94.32 269	97.09 283	85.89 291	93.19 296	95.96 278	92.52 237	94.93 249	97.51 182	89.54 246	98.77 299	87.52 295	97.71 280	98.31 233

MVS_0304		96.22 163	95.94 172	97.04 144	97.07 284	92.54 165	94.19 255	99.04 45	95.17 154	93.74 288	96.92 217	91.77 220	99.73 65	95.76 105	99.81 44	98.85 188

MVS-HIRNet		88.40 319	90.20 305	82.99 343	97.01 285	60.04 361	93.11 297	85.61 355	84.45 323	88.72 345	99.09 46	84.72 283	98.23 334	82.52 327	96.59 314	90.69 354

GA-MVS		92.83 262	92.15 264	94.87 250	96.97 286	87.27 280	90.03 337	96.12 274	91.83 251	94.05 277	94.57 294	76.01 315	98.97 280	92.46 202	97.34 299	98.36 230

test1235678		92.95 260	92.40 260	94.61 259	96.95 287	86.87 285	90.75 330	97.75 223	91.00 260	96.33 204	95.38 282	85.21 279	98.92 282	79.00 339	99.20 180	98.03 258

MVS_Test		96.27 161	96.79 134	94.73 255	96.94 288	86.63 288	96.18 149	98.33 173	94.94 164	96.07 220	98.28 104	95.25 114	99.26 245	97.21 64	97.90 267	98.30 235

MAR-MVS		94.21 235	93.03 250	97.76 92	96.94 288	97.44 30	96.97 118	97.15 254	87.89 291	92.00 323	92.73 325	92.14 207	99.12 256	83.92 321	97.51 292	96.73 304
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
Effi-MVS+-dtu		96.81 136	96.09 162	98.99 10	96.90 290	98.69 2	96.42 132	98.09 204	95.86 125	95.15 243	95.54 279	94.26 146	99.81 33	94.06 171	98.51 242	98.47 216

mvs-test1		96.20 164	95.50 182	98.32 62	96.90 290	98.16 4	95.07 221	98.09 204	95.86 125	93.63 292	94.32 305	94.26 146	99.71 81	94.06 171	97.27 303	97.07 290

MS-PatchMatch		94.83 215	94.91 201	94.57 263	96.81 292	87.10 283	94.23 252	97.34 247	88.74 279	97.14 167	97.11 205	91.94 215	98.23 334	92.99 197	97.92 265	98.37 225

UGNet		96.81 136	96.56 143	97.58 104	96.64 293	93.84 139	97.75 66	97.12 256	96.47 104	93.62 293	98.88 59	93.22 178	99.53 161	95.61 113	99.69 68	99.36 105
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
API-MVS		95.09 207	95.01 196	95.31 235	96.61 294	94.02 132	96.83 120	97.18 253	95.60 134	95.79 229	94.33 304	94.54 136	98.37 329	85.70 307	98.52 240	93.52 344

PAPM		87.64 326	85.84 329	93.04 301	96.54 295	84.99 303	88.42 345	95.57 288	79.52 341	83.82 354	93.05 319	80.57 293	98.41 323	62.29 357	92.79 339	95.71 325

diffmvs		95.00 211	95.00 197	95.01 245	96.53 296	87.96 265	95.73 173	98.32 182	90.67 263	91.89 325	97.43 187	92.07 211	98.90 283	95.44 119	96.88 306	98.16 248

FMVSNet3		95.26 202	94.94 198	96.22 195	96.53 296	90.06 209	95.99 158	97.66 231	94.11 196	97.99 117	97.91 149	80.22 295	99.63 123	94.60 153	99.44 132	98.96 164

HY-MVS	91.43 15	92.58 264	91.81 272	94.90 249	96.49 298	88.87 242	97.31 90	94.62 293	85.92 307	90.50 336	96.84 221	85.05 280	99.40 213	83.77 324	95.78 324	96.43 316

TR-MVS		92.54 269	92.20 263	93.57 288	96.49 298	86.66 287	93.51 285	94.73 292	89.96 270	94.95 247	93.87 308	90.24 242	98.61 312	81.18 334	94.88 330	95.45 330

CANet		95.86 176	95.65 178	96.49 175	96.41 300	90.82 200	94.36 245	98.41 162	94.94 164	92.62 318	96.73 230	92.68 191	99.71 81	95.12 137	99.60 88	98.94 167

mvs_anonymous		95.36 196	96.07 164	93.21 297	96.29 301	81.56 326	94.60 240	97.66 231	93.30 215	96.95 182	98.91 58	93.03 183	99.38 224	96.60 77	97.30 301	98.69 201

Patchmatch-test1		93.38 256	93.59 241	92.73 308	96.24 302	81.40 327	93.24 294	94.00 299	91.58 254	94.57 261	96.67 234	87.94 261	99.03 270	90.42 248	97.66 285	97.77 269

LS3D		97.77 71	97.50 86	98.57 44	96.24 302	97.58 21	98.45 26	98.85 85	98.58 24	97.51 149	97.94 145	95.74 98	99.63 123	95.19 129	98.97 203	98.51 214

new_pmnet		92.34 272	91.69 273	94.32 269	96.23 304	89.16 234	92.27 312	92.88 313	84.39 324	95.29 240	96.35 253	85.66 276	96.74 351	84.53 318	97.56 289	97.05 291

MVE	73.61 22	86.48 328	85.92 328	88.18 337	96.23 304	85.28 298	81.78 356	75.79 359	86.01 305	82.53 356	91.88 332	92.74 189	87.47 359	71.42 354	94.86 331	91.78 350
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
DSMNet-mixed		92.19 275	91.83 271	93.25 296	96.18 306	83.68 321	96.27 142	93.68 303	76.97 351	92.54 319	99.18 35	89.20 254	98.55 317	83.88 322	98.60 238	97.51 278

our_test_3		94.20 237	94.58 215	93.07 300	96.16 307	81.20 328	90.42 334	96.84 264	90.72 262	97.14 167	97.13 203	90.47 235	99.11 259	94.04 175	98.25 252	98.91 175

ppachtmachnet_test		94.49 228	94.84 204	93.46 291	96.16 307	82.10 325	90.59 332	97.48 243	90.53 264	97.01 174	97.59 176	91.01 229	99.36 228	93.97 177	99.18 183	98.94 167

Patchmatch-test		93.60 251	93.25 247	94.63 258	96.14 309	87.47 275	96.04 155	94.50 295	93.57 212	96.47 198	96.97 212	76.50 311	98.61 312	90.67 241	98.41 246	97.81 268

testus		90.90 301	90.51 300	92.06 316	96.07 310	79.45 334	88.99 342	98.44 156	85.46 313	94.15 273	90.77 344	89.12 255	98.01 340	73.66 350	97.95 261	98.71 200

PNet_i23d		83.82 331	83.39 331	85.10 342	96.07 310	65.16 357	81.87 355	94.37 297	90.87 261	93.92 283	92.89 321	52.80 363	96.44 353	77.52 347	70.22 357	93.70 343

wuyk23d		93.25 258	95.20 189	87.40 339	96.07 310	95.38 87	97.04 109	94.97 290	95.33 144	99.70 6	98.11 126	98.14 14	91.94 356	77.76 345	99.68 72	74.89 356

CANet_DTU		94.65 222	94.21 229	95.96 212	95.90 313	89.68 216	93.92 270	97.83 220	93.19 218	90.12 339	95.64 276	88.52 256	99.57 152	93.27 191	99.47 125	98.62 206

test12356		87.98 323	88.41 318	86.69 341	95.84 314	63.49 358	87.15 347	97.32 248	87.21 294	91.78 327	93.36 311	70.66 341	98.39 325	74.70 348	97.64 287	98.19 245

MVSTER		94.21 235	93.93 237	95.05 244	95.83 315	86.46 289	95.18 214	97.65 233	92.41 241	97.94 124	98.00 139	72.39 334	99.58 147	96.36 86	99.56 98	99.12 143

FMVSNet5		93.39 255	92.35 261	96.50 174	95.83 315	90.81 202	97.31 90	98.27 183	92.74 235	96.27 212	98.28 104	62.23 355	99.67 111	90.86 232	99.36 156	99.03 157

PVSNet_0	81.89 21	84.49 330	83.21 333	88.34 336	95.76 317	74.97 350	83.49 352	92.70 317	78.47 346	87.94 348	86.90 355	83.38 286	96.63 352	73.44 351	66.86 358	93.40 345

PAPR		92.22 274	91.27 279	95.07 243	95.73 318	88.81 245	91.97 317	97.87 216	85.80 309	90.91 330	92.73 325	91.16 227	98.33 331	79.48 337	95.76 325	98.08 250

CHOSEN 280x420		89.98 307	89.19 313	92.37 313	95.60 319	81.13 329	86.22 349	97.09 257	81.44 334	87.44 350	93.15 312	73.99 319	99.47 184	88.69 273	99.07 195	96.52 311

ADS-MVSNet2		91.47 294	90.51 300	94.36 268	95.51 320	85.63 292	95.05 224	95.70 284	83.46 326	92.69 314	96.84 221	79.15 298	99.41 210	85.66 309	90.52 343	98.04 256

ADS-MVSNet		90.95 300	90.26 303	93.04 301	95.51 320	82.37 324	95.05 224	93.41 307	83.46 326	92.69 314	96.84 221	79.15 298	98.70 305	85.66 309	90.52 343	98.04 256

CR-MVSNet		93.29 257	92.79 255	94.78 253	95.44 322	88.15 255	96.18 149	97.20 251	84.94 319	94.10 274	98.57 77	77.67 303	99.39 219	95.17 131	95.81 321	96.81 301

RPMNet		94.22 232	94.03 235	94.78 253	95.44 322	88.15 255	96.18 149	93.73 300	97.43 72	94.10 274	98.49 84	79.40 296	99.39 219	95.69 106	95.81 321	96.81 301

1314		92.38 271	92.30 262	92.64 310	95.42 324	85.15 300	95.86 169	96.97 261	85.40 315	90.62 332	93.06 318	91.12 228	97.80 342	86.74 301	95.49 329	94.97 334

tpm		91.08 297	90.85 294	91.75 318	95.33 325	78.09 338	95.03 226	91.27 327	88.75 278	93.53 297	97.40 189	71.24 337	99.30 238	91.25 223	93.87 335	97.87 264

IB-MVS	85.98 20	88.63 316	86.95 325	93.68 286	95.12 326	84.82 306	90.85 329	90.17 342	87.55 292	88.48 346	91.34 341	58.01 357	99.59 145	87.24 298	93.80 336	96.63 309
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
PatchT		93.75 246	93.57 242	94.29 271	95.05 327	87.32 279	96.05 154	92.98 311	97.54 67	94.25 269	98.72 67	75.79 316	99.24 247	95.92 101	95.81 321	96.32 317

tpm2		88.47 317	87.69 320	90.79 326	94.98 328	77.34 343	95.09 218	91.83 322	77.51 350	89.40 342	96.41 249	67.83 352	98.73 302	83.58 326	92.60 341	96.29 318

Patchmtry		95.03 209	94.59 214	96.33 184	94.83 329	90.82 200	96.38 137	97.20 251	96.59 99	97.49 151	98.57 77	77.67 303	99.38 224	92.95 198	99.62 80	98.80 191

MVS		90.02 305	89.20 312	92.47 311	94.71 330	86.90 284	95.86 169	96.74 269	64.72 356	90.62 332	92.77 322	92.54 198	98.39 325	79.30 338	95.56 328	92.12 349

CostFormer		89.75 310	89.25 309	91.26 322	94.69 331	78.00 341	95.32 205	91.98 321	81.50 333	90.55 334	96.96 213	71.06 338	98.89 286	88.59 275	92.63 340	96.87 298

PatchmatchNet		91.98 279	91.87 270	92.30 314	94.60 332	79.71 333	95.12 215	93.59 306	89.52 272	93.61 294	97.02 210	77.94 301	99.18 252	90.84 233	94.57 334	98.01 261
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
tpmp4_e23		88.46 318	87.54 321	91.22 323	94.56 333	78.08 339	95.63 187	93.17 309	79.08 344	85.85 352	96.80 225	65.86 354	98.85 293	84.10 320	92.85 338	96.72 305

LP		93.12 259	92.78 257	94.14 273	94.50 334	85.48 295	95.73 173	95.68 285	92.97 231	95.05 245	97.17 201	81.93 289	99.40 213	93.06 196	88.96 348	97.55 276

tpm cat1		88.01 322	87.33 322	90.05 331	94.48 335	76.28 346	94.47 243	94.35 298	73.84 355	89.26 343	95.61 278	73.64 323	98.30 332	84.13 319	86.20 352	95.57 329

MDTV_nov1_ep13			91.28 278		94.31 336	73.51 351	94.80 235	93.16 310	86.75 301	93.45 301	97.40 189	76.37 312	98.55 317	88.85 270	96.43 315

cascas		91.89 283	91.35 277	93.51 289	94.27 337	85.60 293	88.86 344	98.61 140	79.32 342	92.16 322	91.44 340	89.22 253	98.12 337	90.80 235	97.47 295	96.82 300

test-LLR		89.97 308	89.90 306	90.16 329	94.24 338	74.98 348	89.89 338	89.06 343	92.02 244	89.97 340	90.77 344	73.92 321	98.57 314	91.88 209	97.36 297	96.92 295

test-mter		87.92 324	87.17 323	90.16 329	94.24 338	74.98 348	89.89 338	89.06 343	86.44 302	89.97 340	90.77 344	54.96 361	98.57 314	91.88 209	97.36 297	96.92 295

pmmvs3		90.00 306	88.90 315	93.32 293	94.20 340	85.34 296	91.25 326	92.56 318	78.59 345	93.82 284	95.17 284	67.36 353	98.69 306	89.08 267	98.03 259	95.92 320

tpmrst		90.31 303	90.61 299	89.41 332	94.06 341	72.37 354	95.06 223	93.69 301	88.01 288	92.32 321	96.86 219	77.45 305	98.82 294	91.04 226	87.01 351	97.04 292

test0.0.03 1		90.11 304	89.21 311	92.83 306	93.89 342	86.87 285	91.74 320	88.74 345	92.02 244	94.71 253	91.14 343	73.92 321	94.48 355	83.75 325	92.94 337	97.16 288

JIA-IIPM		91.79 286	90.69 297	95.11 240	93.80 343	90.98 196	94.16 258	91.78 323	96.38 105	90.30 338	99.30 23	72.02 336	98.90 283	88.28 279	90.17 345	95.45 330

TESTMET0.1,1		87.20 327	86.57 327	89.07 333	93.62 344	72.84 353	89.89 338	87.01 354	85.46 313	89.12 344	90.20 349	56.00 360	97.72 343	90.91 231	96.92 304	96.64 307

PatchFormer-LS_test		89.62 311	89.12 314	91.11 324	93.62 344	78.42 337	94.57 242	93.62 305	88.39 283	90.54 335	88.40 352	72.33 335	99.03 270	92.41 203	88.20 349	95.89 321

CMPMVS	73.10 23	92.74 263	91.39 275	96.77 157	93.57 346	94.67 112	94.21 254	97.67 229	80.36 339	93.61 294	96.60 237	82.85 287	97.35 345	84.86 316	98.78 221	98.29 237
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
DWT-MVSNet_test		87.92 324	86.77 326	91.39 320	93.18 347	78.62 336	95.10 216	91.42 325	85.58 310	88.00 347	88.73 351	60.60 356	98.90 283	90.60 242	87.70 350	96.65 306

E-PMN		89.52 312	89.78 307	88.73 334	93.14 348	77.61 342	83.26 353	92.02 320	94.82 170	93.71 289	93.11 313	75.31 317	96.81 349	85.81 306	96.81 309	91.77 351

PMMVS		92.39 270	91.08 283	96.30 187	93.12 349	92.81 163	90.58 333	95.96 278	79.17 343	91.85 326	92.27 328	90.29 241	98.66 311	89.85 256	96.68 313	97.43 279

EMVS		89.06 314	89.22 310	88.61 335	93.00 350	77.34 343	82.91 354	90.92 329	94.64 174	92.63 317	91.81 333	76.30 313	97.02 347	83.83 323	96.90 305	91.48 352

dp		88.08 321	88.05 319	88.16 338	92.85 351	68.81 356	94.17 257	92.88 313	85.47 312	91.38 329	96.14 262	68.87 349	98.81 296	86.88 300	83.80 355	96.87 298

gg-mvs-nofinetune		88.28 320	86.96 324	92.23 315	92.84 352	84.44 315	98.19 41	74.60 360	99.08 9	87.01 351	99.47 8	56.93 358	98.23 334	78.91 340	95.61 327	94.01 342

tpmvs		90.79 302	90.87 293	90.57 328	92.75 353	76.30 345	95.79 172	93.64 304	91.04 259	91.91 324	96.26 255	77.19 309	98.86 292	89.38 262	89.85 346	96.56 310

EPMVS		89.26 313	88.55 317	91.39 320	92.36 354	79.11 335	95.65 182	79.86 358	88.60 280	93.12 308	96.53 241	70.73 340	98.10 338	90.75 237	89.32 347	96.98 293

gm-plane-assit					91.79 355	71.40 355			81.67 331		90.11 350		98.99 274	84.86 316

test2356		85.45 329	83.26 332	92.01 317	91.12 356	80.76 330	85.16 350	92.90 312	83.90 325	90.63 331	87.71 354	53.10 362	97.24 346	69.20 355	95.65 326	98.03 258

GG-mvs-BLEND				90.60 327	91.00 357	84.21 318	98.23 35	72.63 363		82.76 355	84.11 356	56.14 359	96.79 350	72.20 352	92.09 342	90.78 353

DeepMVS_CX				77.17 345	90.94 358	85.28 298		74.08 362	52.51 357	80.87 358	88.03 353	75.25 318	70.63 360	59.23 358	84.94 353	75.62 355

EPNet_dtu		91.39 295	90.75 296	93.31 294	90.48 359	82.61 322	94.80 235	92.88 313	93.39 214	81.74 357	94.90 292	81.36 291	99.11 259	88.28 279	98.87 215	98.21 243
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
testpf		82.70 332	84.35 330	77.74 344	88.97 360	73.23 352	93.85 272	84.33 356	88.10 287	85.06 353	90.42 348	52.62 364	91.05 358	91.00 228	84.82 354	68.93 357

EPNet		93.72 247	92.62 259	97.03 146	87.61 361	92.25 171	96.27 142	91.28 326	96.74 96	87.65 349	97.39 192	85.00 281	99.64 120	92.14 205	99.48 123	99.20 127
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
tmp_tt		57.23 334	62.50 335	41.44 347	34.77 362	49.21 363	83.93 351	60.22 364	15.31 358	71.11 359	79.37 357	70.09 342	44.86 361	64.76 356	82.93 356	30.25 358

test123		12.59 337	15.49 338	3.87 349	6.07 363	2.55 364	90.75 330	2.59 366	2.52 359	5.20 361	13.02 360	4.96 367	1.85 363	5.20 359	9.09 359	7.23 359

testmvs		12.33 338	15.23 339	3.64 350	5.77 364	2.23 365	88.99 342	3.62 365	2.30 360	5.29 360	13.09 359	4.52 368	1.95 362	5.16 360	8.32 360	6.75 360

cdsmvs_eth3d_5k		24.22 336	32.30 337	0.00 351	0.00 365	0.00 366	0.00 357	98.10 203	0.00 361	0.00 362	95.06 287	97.54 28	0.00 364	0.00 361	0.00 362	0.00 362

pcd_1.5k_mvsjas		7.98 339	10.65 340	0.00 351	0.00 365	0.00 366	0.00 357	0.00 367	0.00 361	0.00 362	0.00 363	95.82 91	0.00 364	0.00 361	0.00 362	0.00 362

sosnet-low-res		0.00 341	0.00 342	0.00 351	0.00 365	0.00 366	0.00 357	0.00 367	0.00 361	0.00 362	0.00 363	0.00 369	0.00 364	0.00 361	0.00 362	0.00 362

sosnet		0.00 341	0.00 342	0.00 351	0.00 365	0.00 366	0.00 357	0.00 367	0.00 361	0.00 362	0.00 363	0.00 369	0.00 364	0.00 361	0.00 362	0.00 362

uncertanet		0.00 341	0.00 342	0.00 351	0.00 365	0.00 366	0.00 357	0.00 367	0.00 361	0.00 362	0.00 363	0.00 369	0.00 364	0.00 361	0.00 362	0.00 362

Regformer		0.00 341	0.00 342	0.00 351	0.00 365	0.00 366	0.00 357	0.00 367	0.00 361	0.00 362	0.00 363	0.00 369	0.00 364	0.00 361	0.00 362	0.00 362

ab-mvs-re		7.91 340	10.55 341	0.00 351	0.00 365	0.00 366	0.00 357	0.00 367	0.00 361	0.00 362	94.94 289	0.00 369	0.00 364	0.00 361	0.00 362	0.00 362

uanet		0.00 341	0.00 342	0.00 351	0.00 365	0.00 366	0.00 357	0.00 367	0.00 361	0.00 362	0.00 363	0.00 369	0.00 364	0.00 361	0.00 362	0.00 362

GSMVS																98.06 253

test_part3							95.64 184		94.84 167		97.60 174		99.76 49	91.22 224

test_part1								98.84 88				96.69 61			99.44 132	99.37 102

sam_mvs1												77.80 302				98.06 253

sam_mvs												77.38 306

MTGPA								98.73 114

test_post1							94.98 228				10.37 362	76.21 314	99.04 267	89.47 261

test_post											10.87 361	76.83 310	99.07 264

patchmatchnet-post											96.84 221	77.36 307	99.42 199

MTMP								74.60 360

test9_res														91.29 220	98.89 214	99.00 159

agg_prior2														90.34 251	98.90 210	99.10 150

test_prior4						95.38 87	93.61 283

test_prior2							93.33 292		94.21 192	94.02 278	96.25 256	93.64 166		91.90 207	98.96 204

旧先验2							93.35 291		77.95 349	95.77 232			98.67 310	90.74 238

新几何2							93.43 287

无先验							93.20 295	97.91 213	80.78 336				99.40 213	87.71 283		97.94 263

原ACMM2							92.82 300

testdata2													99.46 189	87.84 282

segment_acmp												95.34 109

testdata1							92.77 301		93.78 207

plane_prior5								98.75 111					99.46 189	92.59 200	99.20 180	99.28 119

plane_prior4											96.77 227

plane_prior3						94.51 115			95.29 146	96.16 218

plane_prior2							96.50 130		96.36 106

plane_prior						94.29 122	95.42 196		94.31 189						98.93 209

n2								0.00 367

nn								0.00 367

door-mid								98.17 196

test11								98.08 206

door								97.81 221

HQP5-MVS						92.47 167

BP-MVS														90.51 245

HQP4-MVS										92.87 310			99.23 249			99.06 155

HQP3-MVS								98.43 157							98.74 225

HQP2-MVS												90.33 237

MDTV_nov1_ep13_2view						57.28 362	94.89 230		80.59 337	94.02 278		78.66 300		85.50 311		97.82 267

ACMMP++_ref															99.52 110

ACMMP++															99.55 102

Test By Simon												94.51 138

High-resolution multi-view benchmark