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.

pcd_1.5k_mvsjas		16.61 334	22.14 335	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	100.00 1	99.28 39	0.00 359	0.00 356	0.00 357	0.00 357

pcd1.5k->3k		49.97 330	55.52 331	33.31 343	99.95 13	0.00 361	0.00 352	99.81 56	0.00 356	0.00 357	100.00 1	99.96 1	0.00 359	0.00 356	100.00 1	99.92 3

sosnet-low-res		8.33 335	11.11 336	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	100.00 1	0.00 364	0.00 359	0.00 356	0.00 357	0.00 357

sosnet		8.33 335	11.11 336	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	100.00 1	0.00 364	0.00 359	0.00 356	0.00 357	0.00 357

uncertanet		8.33 335	11.11 336	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	100.00 1	0.00 364	0.00 359	0.00 356	0.00 357	0.00 357

Regformer		8.33 335	11.11 336	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	100.00 1	0.00 364	0.00 359	0.00 356	0.00 357	0.00 357

uanet		8.33 335	11.11 336	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	100.00 1	0.00 364	0.00 359	0.00 356	0.00 357	0.00 357

Anonymous20231211		99.83 11	99.81 10	99.89 6	99.97 4	99.95 2	99.88 4	99.93 6	99.87 13	99.94 20	99.98 8	99.55 21	99.95 41	99.21 79	99.98 36	99.78 31

LCM-MVSNet		99.95 1	99.95 1	99.95 1	99.99 1	99.99 1	99.95 2	99.97 3	99.99 1	100.00 1	99.98 8	99.78 8	100.00 1	99.92 3	100.00 1	99.87 10

ANet_high		99.88 4	99.87 4	99.91 2	99.99 1	99.91 3	99.65 54	100.00 1	99.90 6	100.00 1	99.97 10	99.61 17	99.97 16	99.75 31	100.00 1	99.84 15

MVS-HIRNet		97.86 267	98.22 240	96.76 326	99.28 277	91.53 348	98.38 271	92.60 355	99.13 150	99.31 215	99.96 11	97.18 238	99.68 310	98.34 156	99.83 143	99.07 273

pmmvs6		99.86 6	99.86 6	99.83 24	99.94 15	99.90 4	99.83 8	99.91 11	99.85 19	99.94 20	99.95 12	99.73 10	99.90 109	99.65 35	99.97 47	99.69 56

gg-mvs-nofinetune		95.87 322	95.17 325	97.97 299	98.19 345	96.95 300	99.69 38	89.23 358	99.89 10	96.24 348	99.94 13	81.19 352	99.51 342	93.99 329	98.20 332	97.44 339

anonymousdsp		99.80 13	99.77 14	99.90 4	99.96 5	99.88 8	99.73 21	99.85 29	99.70 49	99.92 31	99.93 14	99.45 23	99.97 16	99.36 61	100.00 1	99.85 14

mvs_tets		99.90 2	99.90 2	99.90 4	99.96 5	99.79 33	99.72 25	99.88 18	99.92 5	99.98 3	99.93 14	99.94 2	99.98 7	99.77 30	100.00 1	99.92 3

OurMVSNet-221017-0		99.75 19	99.71 25	99.84 21	99.96 5	99.83 22	99.83 8	99.85 29	99.80 31	99.93 26	99.93 14	98.54 140	99.93 66	99.59 39	99.98 36	99.76 37

PS-MVSNAJss		99.84 9	99.82 9	99.89 6	99.96 5	99.77 37	99.68 41	99.85 29	99.95 3	99.98 3	99.92 17	99.28 39	99.98 7	99.75 31	100.00 1	99.94 2

test_djsdf		99.84 9	99.81 10	99.91 2	99.94 15	99.84 18	99.77 13	99.80 60	99.73 42	99.97 6	99.92 17	99.77 9	99.98 7	99.43 53	100.00 1	99.90 5

TDRefinement		99.72 25	99.70 28	99.77 39	99.90 25	99.85 13	99.86 7	99.92 7	99.69 53	99.78 82	99.92 17	99.37 30	99.88 139	98.93 121	99.95 65	99.60 124

UA-Net		99.78 15	99.76 18	99.86 18	99.72 130	99.71 52	99.91 3	99.95 5	99.96 2	99.71 106	99.91 20	99.15 53	99.97 16	99.50 48	100.00 1	99.90 5

v7n		99.82 12	99.80 12	99.88 12	99.96 5	99.84 18	99.82 10	99.82 48	99.84 23	99.94 20	99.91 20	99.13 57	99.96 33	99.83 20	99.99 20	99.83 18

v13		99.76 17	99.77 14	99.73 63	99.86 35	99.55 96	99.77 13	99.86 22	99.79 33	99.96 8	99.91 20	98.90 84	99.87 158	99.91 5	100.00 1	99.78 31

jajsoiax		99.89 3	99.89 3	99.89 6	99.96 5	99.78 35	99.70 29	99.86 22	99.89 10	99.98 3	99.90 23	99.94 2	99.98 7	99.75 31	100.00 1	99.90 5

v12		99.75 19	99.77 14	99.72 68	99.85 39	99.53 99	99.75 17	99.86 22	99.78 34	99.96 8	99.90 23	98.88 87	99.86 178	99.91 5	100.00 1	99.77 34

v11		99.75 19	99.76 18	99.71 72	99.85 39	99.49 102	99.73 21	99.84 37	99.75 39	99.95 16	99.90 23	98.93 80	99.86 178	99.92 3	100.00 1	99.77 34

v52		99.85 7	99.84 7	99.89 6	99.96 5	99.89 6	99.87 5	99.81 56	99.85 19	99.96 8	99.90 23	99.27 42	99.95 41	99.93 1	99.99 20	99.82 23

V4		99.85 7	99.84 7	99.88 12	99.96 5	99.89 6	99.87 5	99.81 56	99.85 19	99.96 8	99.90 23	99.27 42	99.95 41	99.93 1	100.00 1	99.82 23

V9		99.74 23	99.75 20	99.71 72	99.84 42	99.50 100	99.74 19	99.86 22	99.76 38	99.96 8	99.90 23	98.83 90	99.85 194	99.91 5	100.00 1	99.77 34

SixPastTwentyTwo		99.42 84	99.30 101	99.76 42	99.92 19	99.67 68	99.70 29	99.14 280	99.65 65	99.89 38	99.90 23	96.20 262	99.94 55	99.42 57	99.92 88	99.67 69

DeepC-MVS	98.90 4	99.62 43	99.61 41	99.67 85	99.72 130	99.44 117	99.24 140	99.71 104	99.27 123	99.93 26	99.90 23	99.70 12	99.93 66	98.99 108	99.99 20	99.64 95
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
V14		99.73 24	99.74 21	99.69 79	99.83 46	99.48 105	99.72 25	99.85 29	99.74 40	99.96 8	99.89 31	98.79 98	99.85 194	99.91 5	100.00 1	99.76 37

TransMVSNet (Re)		99.78 15	99.77 14	99.81 27	99.91 21	99.85 13	99.75 17	99.86 22	99.70 49	99.91 33	99.89 31	99.60 19	99.87 158	99.59 39	99.74 190	99.71 49

MIMVSNet1		99.66 36	99.62 38	99.80 29	99.94 15	99.87 9	99.69 38	99.77 73	99.78 34	99.93 26	99.89 31	97.94 190	99.92 83	99.65 35	99.98 36	99.62 113

v15		99.72 25	99.73 24	99.68 82	99.82 53	99.44 117	99.70 29	99.85 29	99.72 45	99.95 16	99.88 34	98.76 105	99.84 210	99.90 9	100.00 1	99.75 40

Baseline_NR-MVSNet		99.49 68	99.37 87	99.82 25	99.91 21	99.84 18	98.83 224	99.86 22	99.68 56	99.65 126	99.88 34	97.67 210	99.87 158	99.03 105	99.86 126	99.76 37

K. test v3		98.87 201	98.60 208	99.69 79	99.93 18	99.46 110	99.74 19	94.97 353	99.78 34	99.88 46	99.88 34	93.66 285	99.97 16	99.61 38	99.95 65	99.64 95

new-patchmatchnet		99.35 102	99.57 48	98.71 275	99.82 53	96.62 304	98.55 251	99.75 84	99.50 90	99.88 46	99.87 37	99.31 35	99.88 139	99.43 53	100.00 1	99.62 113

v748		99.76 17	99.74 21	99.84 21	99.95 13	99.83 22	99.82 10	99.80 60	99.82 27	99.95 16	99.87 37	98.72 112	99.93 66	99.72 34	99.98 36	99.75 40

pm-mvs1		99.79 14	99.79 13	99.78 37	99.91 21	99.83 22	99.76 16	99.87 20	99.73 42	99.89 38	99.87 37	99.63 15	99.87 158	99.54 44	99.92 88	99.63 99

v17		99.70 28	99.71 25	99.67 85	99.81 61	99.44 117	99.70 29	99.83 40	99.69 53	99.94 20	99.87 37	98.70 113	99.84 210	99.88 14	99.99 20	99.73 43

v16		99.70 28	99.71 25	99.67 85	99.81 61	99.43 123	99.70 29	99.83 40	99.70 49	99.94 20	99.87 37	98.69 115	99.84 210	99.88 14	99.99 20	99.73 43

v10		99.69 32	99.69 29	99.66 93	99.81 61	99.39 135	99.66 49	99.75 84	99.60 80	99.92 31	99.87 37	98.75 108	99.86 178	99.90 9	99.99 20	99.73 43

JIA-IIPM		98.06 263	97.92 260	98.50 282	98.59 337	97.02 299	98.80 229	98.51 307	99.88 12	97.89 327	99.87 37	91.89 299	99.90 109	98.16 174	97.68 344	98.59 299

LTVRE_ROB	99.19 1	99.88 4	99.87 4	99.88 12	99.91 21	99.90 4	99.96 1	99.92 7	99.90 6	99.97 6	99.87 37	99.81 7	99.95 41	99.54 44	99.99 20	99.80 25
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
v18		99.68 33	99.69 29	99.65 97	99.79 82	99.40 132	99.68 41	99.83 40	99.66 62	99.93 26	99.85 45	98.65 124	99.84 210	99.87 18	99.99 20	99.71 49

v8		99.68 33	99.69 29	99.65 97	99.80 69	99.40 132	99.66 49	99.76 79	99.64 67	99.93 26	99.85 45	98.66 122	99.84 210	99.88 14	99.99 20	99.71 49

EU-MVSNet		99.39 93	99.62 38	98.72 274	99.88 28	96.44 306	99.56 70	99.85 29	99.90 6	99.90 35	99.85 45	98.09 179	99.83 226	99.58 41	99.95 65	99.90 5

DSMNet-mixed		99.48 70	99.65 34	98.95 249	99.71 133	97.27 295	99.50 74	99.82 48	99.59 82	99.41 189	99.85 45	99.62 16	100.00 1	99.53 46	99.89 106	99.59 135

ACMH	98.42 6	99.59 45	99.54 53	99.72 68	99.86 35	99.62 83	99.56 70	99.79 68	98.77 187	99.80 74	99.85 45	99.64 14	99.85 194	98.70 137	99.89 106	99.70 53
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
XXY-MVS		99.71 27	99.67 32	99.81 27	99.89 27	99.72 51	99.59 65	99.82 48	99.39 111	99.82 65	99.84 50	99.38 28	99.91 92	99.38 58	99.93 85	99.80 25

FC-MVSNet-test		99.70 28	99.65 34	99.86 18	99.88 28	99.86 12	99.72 25	99.78 70	99.90 6	99.82 65	99.83 51	98.45 153	99.87 158	99.51 47	99.97 47	99.86 12

lessismore_v0				99.64 104	99.86 35	99.38 141		90.66 356		99.89 38	99.83 51	94.56 279	99.97 16	99.56 43	99.92 88	99.57 143

no-one		99.28 117	99.23 117	99.45 172	99.87 32	99.08 201	98.95 208	99.52 200	98.88 173	99.77 86	99.83 51	97.78 202	99.90 109	98.46 149	99.99 20	99.38 212

GBi-Net		99.42 84	99.31 96	99.73 63	99.49 216	99.77 37	99.68 41	99.70 107	99.44 101	99.62 138	99.83 51	97.21 234	99.90 109	98.96 115	99.90 100	99.53 157

test1		99.42 84	99.31 96	99.73 63	99.49 216	99.77 37	99.68 41	99.70 107	99.44 101	99.62 138	99.83 51	97.21 234	99.90 109	98.96 115	99.90 100	99.53 157

FMVSNet1		99.66 36	99.63 37	99.73 63	99.78 88	99.77 37	99.68 41	99.70 107	99.67 58	99.82 65	99.83 51	98.98 74	99.90 109	99.24 78	99.97 47	99.53 157

TAMVS		99.49 68	99.45 73	99.63 108	99.48 222	99.42 127	99.45 79	99.57 175	99.66 62	99.78 82	99.83 51	97.85 197	99.86 178	99.44 52	99.96 59	99.61 118

SD-MVS		99.01 178	99.30 101	98.15 295	99.50 211	99.40 132	98.94 211	99.61 147	99.22 137	99.75 90	99.82 58	99.54 22	95.51 356	97.48 214	99.87 119	99.54 154

ab-mvs		99.33 110	99.28 108	99.47 165	99.57 183	99.39 135	99.78 12	99.43 225	98.87 174	99.57 149	99.82 58	98.06 182	99.87 158	98.69 138	99.73 196	99.15 250

PMVS	92.94 21	98.82 206	98.81 194	98.85 259	99.84 42	97.99 276	99.20 150	99.47 214	99.71 47	99.42 183	99.82 58	98.09 179	99.47 344	93.88 330	99.85 129	99.07 273
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
VPA-MVSNet		99.66 36	99.62 38	99.79 34	99.68 149	99.75 44	99.62 56	99.69 113	99.85 19	99.80 74	99.81 61	98.81 91	99.91 92	99.47 50	99.88 112	99.70 53

UGNet		99.38 95	99.34 92	99.49 160	98.90 312	98.90 220	99.70 29	99.35 246	99.86 16	98.57 296	99.81 61	98.50 149	99.93 66	99.38 58	99.98 36	99.66 79
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
ambc				99.20 229	99.35 254	98.53 241	99.17 159	99.46 217		99.67 116	99.80 63	98.46 152	99.70 296	97.92 185	99.70 204	99.38 212

VDDNet		98.97 184	98.82 193	99.42 179	99.71 133	98.81 229	99.62 56	98.68 301	99.81 28	99.38 201	99.80 63	94.25 281	99.85 194	98.79 129	99.32 269	99.59 135

mvs_anonymous		99.28 117	99.39 82	98.94 250	99.19 289	97.81 283	99.02 194	99.55 181	99.78 34	99.85 57	99.80 63	98.24 170	99.86 178	99.57 42	99.50 243	99.15 250

QAPM		98.40 243	97.99 254	99.65 97	99.39 246	99.47 106	99.67 46	99.52 200	91.70 344	98.78 279	99.80 63	98.55 138	99.95 41	94.71 320	99.75 183	99.53 157

wuykxyi23d		99.65 41	99.64 36	99.69 79	99.92 19	99.20 186	98.89 213	99.99 2	98.73 195	99.95 16	99.80 63	99.84 4	99.99 4	99.64 37	99.98 36	99.89 9

3Dnovator	99.15 2	99.43 81	99.36 90	99.65 97	99.39 246	99.42 127	99.70 29	99.56 178	99.23 134	99.35 205	99.80 63	99.17 51	99.95 41	98.21 166	99.84 133	99.59 135

CMPMVS	77.52 23	98.50 231	98.19 245	99.41 186	98.33 343	99.56 93	99.01 196	99.59 165	95.44 323	99.57 149	99.80 63	95.64 269	99.46 347	96.47 268	99.92 88	99.21 240
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
FIs		99.65 41	99.58 45	99.84 21	99.84 42	99.85 13	99.66 49	99.75 84	99.86 16	99.74 98	99.79 70	98.27 168	99.85 194	99.37 60	99.93 85	99.83 18

testing_2		99.58 46	99.56 51	99.62 116	99.81 61	99.44 117	99.14 172	99.43 225	99.69 53	99.82 65	99.79 70	99.14 54	99.79 261	99.31 70	99.95 65	99.63 99

testmv		99.53 65	99.51 66	99.59 127	99.73 120	99.31 157	98.48 260	99.92 7	99.57 84	99.87 51	99.79 70	99.12 58	99.91 92	99.16 91	99.99 20	99.55 147

LCM-MVSNet-Re		99.28 117	99.15 123	99.67 85	99.33 268	99.76 41	99.34 106	99.97 3	98.93 168	99.91 33	99.79 70	98.68 117	99.93 66	96.80 250	99.56 229	99.30 230

CHOSEN 1792x2688		99.39 93	99.30 101	99.65 97	99.88 28	99.25 173	98.78 233	99.88 18	98.66 199	99.96 8	99.79 70	97.45 221	99.93 66	99.34 63	99.99 20	99.78 31

LP		98.34 250	98.44 220	98.05 297	98.88 319	95.31 329	99.28 130	98.74 298	99.12 151	98.98 254	99.79 70	93.40 287	99.93 66	98.38 152	99.41 259	98.90 285

CR-MVSNet		98.35 248	98.20 242	98.83 263	99.05 306	98.12 268	99.30 121	99.67 119	97.39 283	99.16 237	99.79 70	91.87 300	99.91 92	98.78 132	98.77 303	98.44 307

Patchmtry		98.78 211	98.54 216	99.49 160	98.89 316	99.19 188	99.32 111	99.67 119	99.65 65	99.72 102	99.79 70	91.87 300	99.95 41	98.00 182	99.97 47	99.33 224

wuyk23d		97.58 274	99.13 127	92.93 340	99.69 142	99.49 102	99.52 72	99.77 73	97.97 251	99.96 8	99.79 70	99.84 4	99.94 55	95.85 291	99.82 152	79.36 353

DTE-MVSNet		99.68 33	99.61 41	99.88 12	99.80 69	99.87 9	99.67 46	99.71 104	99.72 45	99.84 60	99.78 79	98.67 120	99.97 16	99.30 71	99.95 65	99.80 25

EG-PatchMatch MVS		99.57 47	99.56 51	99.62 116	99.77 98	99.33 154	99.26 134	99.76 79	99.32 119	99.80 74	99.78 79	99.29 37	99.87 158	99.15 92	99.91 98	99.66 79

RPSCF		99.18 147	99.02 160	99.64 104	99.83 46	99.85 13	99.44 81	99.82 48	98.33 234	99.50 171	99.78 79	97.90 192	99.65 327	96.78 251	99.83 143	99.44 196

3Dnovator+	98.92 3	99.35 102	99.24 115	99.67 85	99.35 254	99.47 106	99.62 56	99.50 205	99.44 101	99.12 242	99.78 79	98.77 104	99.94 55	97.87 188	99.72 201	99.62 113

Gipuma		99.57 47	99.59 43	99.49 160	99.98 3	99.71 52	99.72 25	99.84 37	99.81 28	99.94 20	99.78 79	98.91 83	99.71 295	98.41 151	99.95 65	99.05 275
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
COLMAP_ROB	98.06 12	99.45 79	99.37 87	99.70 78	99.83 46	99.70 59	99.38 92	99.78 70	99.53 87	99.67 116	99.78 79	99.19 49	99.86 178	97.32 222	99.87 119	99.55 147
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
USDC		98.96 187	98.93 175	99.05 243	99.54 197	97.99 276	97.07 339	99.80 60	98.21 240	99.75 90	99.77 85	98.43 154	99.64 329	97.90 186	99.88 112	99.51 168

EPP-MVSNet		99.17 150	99.00 165	99.66 93	99.80 69	99.43 123	99.70 29	99.24 271	99.48 92	99.56 156	99.77 85	94.89 275	99.93 66	98.72 136	99.89 106	99.63 99

OpenMVS	98.12 10	98.23 255	97.89 264	99.26 219	99.19 289	99.26 169	99.65 54	99.69 113	91.33 345	98.14 318	99.77 85	98.28 167	99.96 33	95.41 310	99.55 235	98.58 301

PatchT		98.45 237	98.32 235	98.83 263	98.94 310	98.29 259	99.24 140	98.82 294	99.84 23	99.08 245	99.76 88	91.37 303	99.94 55	98.82 128	99.00 291	98.26 314

MIMVSNet		98.43 238	98.20 242	99.11 235	99.53 200	98.38 250	99.58 67	98.61 303	98.96 164	99.33 211	99.76 88	90.92 307	99.81 253	97.38 220	99.76 180	99.15 250

DP-MVS		99.48 70	99.39 82	99.74 55	99.57 183	99.62 83	99.29 129	99.61 147	99.87 13	99.74 98	99.76 88	98.69 115	99.87 158	98.20 167	99.80 166	99.75 40

ACMH+	98.40 8	99.50 66	99.43 78	99.71 72	99.86 35	99.76 41	99.32 111	99.77 73	99.53 87	99.77 86	99.76 88	99.26 45	99.78 269	97.77 194	99.88 112	99.60 124

v1240		99.56 50	99.58 45	99.51 156	99.80 69	99.00 206	99.00 198	99.65 132	99.15 147	99.90 35	99.75 92	99.09 61	99.88 139	99.90 9	99.96 59	99.67 69

Vis-MVSNet		99.75 19	99.74 21	99.79 34	99.88 28	99.66 71	99.69 38	99.92 7	99.67 58	99.77 86	99.75 92	99.61 17	99.98 7	99.35 62	99.98 36	99.72 46
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
RPMNet		98.53 229	98.44 220	98.83 263	99.05 306	98.12 268	99.30 121	98.78 296	99.86 16	99.16 237	99.74 94	92.53 296	99.91 92	98.75 133	98.77 303	98.44 307

FMVSNet2		99.35 102	99.28 108	99.55 146	99.49 216	99.35 151	99.45 79	99.57 175	99.44 101	99.70 108	99.74 94	97.21 234	99.87 158	99.03 105	99.94 77	99.44 196

IterMVS		98.97 184	99.16 121	98.42 284	99.74 117	95.64 324	98.06 300	99.83 40	99.83 26	99.85 57	99.74 94	96.10 265	99.99 4	99.27 77	100.00 1	99.63 99
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
OpenMVS_ROB	97.31 17	97.36 279	96.84 289	98.89 258	99.29 275	99.45 115	98.87 217	99.48 210	86.54 350	99.44 177	99.74 94	97.34 228	99.86 178	91.61 334	99.28 273	97.37 341

semantic-postprocess				98.51 279	99.75 111	95.90 317		99.84 37	99.84 23	99.89 38	99.73 98	95.96 267	99.99 4	99.33 65	100.00 1	99.63 99

ACMMP_Plus		99.28 117	99.11 132	99.79 34	99.75 111	99.81 28	98.95 208	99.53 190	98.27 238	99.53 165	99.73 98	98.75 108	99.87 158	97.70 198	99.83 143	99.68 62

v1144		99.54 59	99.53 61	99.59 127	99.79 82	99.28 163	99.10 181	99.61 147	99.20 138	99.84 60	99.73 98	98.67 120	99.84 210	99.86 19	99.98 36	99.64 95

PM-MVS		99.36 100	99.29 106	99.58 131	99.83 46	99.66 71	98.95 208	99.86 22	98.85 176	99.81 71	99.73 98	98.40 158	99.92 83	98.36 154	99.83 143	99.17 248

PEN-MVS		99.66 36	99.59 43	99.89 6	99.83 46	99.87 9	99.66 49	99.73 92	99.70 49	99.84 60	99.73 98	98.56 134	99.96 33	99.29 74	99.94 77	99.83 18

PVSNet_Blended_VisFu		99.40 90	99.38 84	99.44 174	99.90 25	98.66 236	98.94 211	99.91 11	97.97 251	99.79 79	99.73 98	99.05 69	99.97 16	99.15 92	99.99 20	99.68 62

Patchmatch-RL test		98.60 222	98.36 231	99.33 203	99.77 98	99.07 203	98.27 278	99.87 20	98.91 171	99.74 98	99.72 104	90.57 314	99.79 261	98.55 145	99.85 129	99.11 259

v144192		99.55 54	99.54 53	99.58 131	99.78 88	99.20 186	99.11 180	99.62 143	99.18 140	99.89 38	99.72 104	98.66 122	99.87 158	99.88 14	99.97 47	99.66 79

v1192		99.57 47	99.57 48	99.57 137	99.77 98	99.22 180	99.04 191	99.60 161	99.18 140	99.87 51	99.72 104	99.08 64	99.85 194	99.89 13	99.98 36	99.66 79

AllTest		99.21 140	99.07 146	99.63 108	99.78 88	99.64 77	99.12 179	99.83 40	98.63 202	99.63 131	99.72 104	98.68 117	99.75 283	96.38 269	99.83 143	99.51 168

TestCases				99.63 108	99.78 88	99.64 77		99.83 40	98.63 202	99.63 131	99.72 104	98.68 117	99.75 283	96.38 269	99.83 143	99.51 168

v7		99.56 50	99.54 53	99.61 119	99.80 69	99.39 135	99.30 121	99.59 165	99.14 149	99.82 65	99.72 104	98.75 108	99.84 210	99.83 20	99.94 77	99.61 118

ACMM	98.09 11	99.46 77	99.38 84	99.72 68	99.80 69	99.69 63	99.13 177	99.65 132	98.99 162	99.64 128	99.72 104	99.39 24	99.86 178	98.23 164	99.81 161	99.60 124
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
v1921920		99.56 50	99.57 48	99.55 146	99.75 111	99.11 195	99.05 189	99.61 147	99.15 147	99.88 46	99.71 111	99.08 64	99.87 158	99.90 9	99.97 47	99.66 79

APDe-MVS		99.48 70	99.36 90	99.85 20	99.55 196	99.81 28	99.50 74	99.69 113	98.99 162	99.75 90	99.71 111	98.79 98	99.93 66	98.46 149	99.85 129	99.80 25

PS-CasMVS		99.66 36	99.58 45	99.89 6	99.80 69	99.85 13	99.66 49	99.73 92	99.62 71	99.84 60	99.71 111	98.62 128	99.96 33	99.30 71	99.96 59	99.86 12

XVG-ACMP-BASELINE		99.23 128	99.10 139	99.63 108	99.82 53	99.58 90	98.83 224	99.72 101	98.36 224	99.60 145	99.71 111	98.92 81	99.91 92	97.08 237	99.84 133	99.40 207

PVSNet_BlendedMVS		99.03 172	99.01 163	99.09 237	99.54 197	97.99 276	98.58 246	99.82 48	97.62 270	99.34 209	99.71 111	98.52 146	99.77 275	97.98 183	99.97 47	99.52 165

IS-MVSNet		99.03 172	98.85 187	99.55 146	99.80 69	99.25 173	99.73 21	99.15 279	99.37 113	99.61 143	99.71 111	94.73 277	99.81 253	97.70 198	99.88 112	99.58 139

LS3D		99.24 127	99.11 132	99.61 119	98.38 342	99.79 33	99.57 68	99.68 116	99.61 75	99.15 239	99.71 111	98.70 113	99.91 92	97.54 211	99.68 207	99.13 257

TSAR-MVS + MP.		99.34 107	99.24 115	99.63 108	99.82 53	99.37 144	99.26 134	99.35 246	98.77 187	99.57 149	99.70 118	99.27 42	99.88 139	97.71 197	99.75 183	99.65 89

V42		99.56 50	99.54 53	99.63 108	99.79 82	99.46 110	99.39 86	99.59 165	99.24 132	99.86 56	99.70 118	98.55 138	99.82 234	99.79 26	99.95 65	99.60 124

MDA-MVSNet-bldmvs		99.06 166	99.05 153	99.07 241	99.80 69	97.83 282	98.89 213	99.72 101	99.29 120	99.63 131	99.70 118	96.47 255	99.89 124	98.17 173	99.82 152	99.50 174

CDS-MVSNet		99.22 137	99.13 127	99.50 158	99.35 254	99.11 195	98.96 207	99.54 185	99.46 99	99.61 143	99.70 118	96.31 259	99.83 226	99.34 63	99.88 112	99.55 147
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
DeepPCF-MVS	98.42 6	99.18 147	99.02 160	99.67 85	99.22 284	99.75 44	97.25 337	99.47 214	98.72 196	99.66 120	99.70 118	99.29 37	99.63 331	98.07 179	99.81 161	99.62 113

TinyColmap		98.97 184	98.93 175	99.07 241	99.46 231	98.19 264	97.75 322	99.75 84	98.79 184	99.54 162	99.70 118	98.97 76	99.62 332	96.63 260	99.83 143	99.41 206

tfpnnormal		99.43 81	99.38 84	99.60 125	99.87 32	99.75 44	99.59 65	99.78 70	99.71 47	99.90 35	99.69 124	98.85 89	99.90 109	97.25 228	99.78 174	99.15 250

tmp_tt		95.75 324	95.42 321	96.76 326	89.90 357	94.42 333	98.86 218	97.87 324	78.01 351	99.30 219	99.69 124	97.70 205	95.89 355	99.29 74	98.14 336	99.95 1

v1141		99.54 59	99.52 63	99.57 137	99.78 88	99.27 167	99.15 167	99.61 147	99.26 127	99.89 38	99.69 124	98.56 134	99.82 234	99.82 23	99.97 47	99.63 99

v1neww		99.55 54	99.54 53	99.61 119	99.80 69	99.39 135	99.32 111	99.61 147	99.18 140	99.87 51	99.69 124	98.64 126	99.82 234	99.79 26	99.94 77	99.60 124

v7new		99.55 54	99.54 53	99.61 119	99.80 69	99.39 135	99.32 111	99.61 147	99.18 140	99.87 51	99.69 124	98.64 126	99.82 234	99.79 26	99.94 77	99.60 124

divwei89l23v2f112		99.54 59	99.52 63	99.57 137	99.78 88	99.27 167	99.15 167	99.61 147	99.26 127	99.89 38	99.69 124	98.56 134	99.82 234	99.82 23	99.96 59	99.63 99

VDD-MVS		99.20 142	99.11 132	99.44 174	99.43 238	98.98 208	99.50 74	98.32 315	99.80 31	99.56 156	99.69 124	96.99 244	99.85 194	98.99 108	99.73 196	99.50 174

WR-MVS_H		99.61 44	99.53 61	99.87 16	99.80 69	99.83 22	99.67 46	99.75 84	99.58 83	99.85 57	99.69 124	98.18 177	99.94 55	99.28 76	99.95 65	99.83 18

LPG-MVS_test		99.22 137	99.05 153	99.74 55	99.82 53	99.63 81	99.16 165	99.73 92	97.56 274	99.64 128	99.69 124	99.37 30	99.89 124	96.66 258	99.87 119	99.69 56

LGP-MVS_train				99.74 55	99.82 53	99.63 81		99.73 92	97.56 274	99.64 128	99.69 124	99.37 30	99.89 124	96.66 258	99.87 119	99.69 56

FMVSNet5		97.80 268	97.25 279	99.42 179	98.83 322	98.97 210	99.38 92	99.80 60	98.87 174	99.25 223	99.69 124	80.60 356	99.91 92	98.96 115	99.90 100	99.38 212

ACMMP		99.25 124	99.08 142	99.74 55	99.79 82	99.68 66	99.50 74	99.65 132	98.07 245	99.52 167	99.69 124	98.57 133	99.92 83	97.18 234	99.79 169	99.63 99
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
MVP-Stereo		99.16 152	99.08 142	99.43 177	99.48 222	99.07 203	99.08 186	99.55 181	98.63 202	99.31 215	99.68 136	98.19 176	99.78 269	98.18 171	99.58 228	99.45 191
Qingsong Yan: MVP-Stereo: A Parallel Multi-View Patchmatch Stereo Method with Dilation Matching for Photogrammetric Application.
nrg030		99.70 28	99.66 33	99.82 25	99.76 103	99.84 18	99.61 60	99.70 107	99.93 4	99.78 82	99.68 136	99.10 59	99.78 269	99.45 51	99.96 59	99.83 18

v6		99.55 54	99.54 53	99.61 119	99.80 69	99.39 135	99.32 111	99.60 161	99.18 140	99.87 51	99.68 136	98.65 124	99.82 234	99.79 26	99.95 65	99.61 118

v1		99.54 59	99.52 63	99.58 131	99.77 98	99.28 163	99.15 167	99.61 147	99.26 127	99.88 46	99.68 136	98.56 134	99.82 234	99.82 23	99.97 47	99.63 99

XVG-OURS		99.21 140	99.06 148	99.65 97	99.82 53	99.62 83	97.87 319	99.74 89	98.36 224	99.66 120	99.68 136	99.71 11	99.90 109	96.84 248	99.88 112	99.43 202

N_pmnet		98.73 216	98.53 217	99.35 200	99.72 130	98.67 235	98.34 274	94.65 354	98.35 229	99.79 79	99.68 136	98.03 183	99.93 66	98.28 162	99.92 88	99.44 196

EI-MVSNet		99.38 95	99.44 75	99.21 227	99.58 174	98.09 272	99.26 134	99.46 217	99.62 71	99.75 90	99.67 142	98.54 140	99.85 194	99.15 92	99.92 88	99.68 62

CVMVSNet		98.61 221	98.88 183	97.80 308	99.58 174	93.60 336	99.26 134	99.64 137	99.66 62	99.72 102	99.67 142	93.26 288	99.93 66	99.30 71	99.81 161	99.87 10

MVS_Test		99.28 117	99.31 96	99.19 230	99.35 254	98.79 231	99.36 98	99.49 209	99.17 145	99.21 231	99.67 142	98.78 101	99.66 320	99.09 101	99.66 216	99.10 263

SteuartSystems-ACMMP		99.30 114	99.14 124	99.76 42	99.87 32	99.66 71	99.18 152	99.60 161	98.55 208	99.57 149	99.67 142	99.03 71	99.94 55	97.01 240	99.80 166	99.69 56
Skip Steuart: Steuart Systems R&D Blog.
pmmvs-eth3d		99.48 70	99.47 69	99.51 156	99.77 98	99.41 131	98.81 228	99.66 123	99.42 108	99.75 90	99.66 146	99.20 48	99.76 277	98.98 110	99.99 20	99.36 219

EI-MVSNet-UG-set		99.48 70	99.50 67	99.42 179	99.57 183	98.65 238	99.24 140	99.46 217	99.68 56	99.80 74	99.66 146	98.99 73	99.89 124	99.19 83	99.90 100	99.72 46

YYNet1		98.95 190	98.99 168	98.84 261	99.64 158	97.14 298	98.22 282	99.32 251	98.92 170	99.59 146	99.66 146	97.40 223	99.83 226	98.27 163	99.90 100	99.55 147

MDA-MVSNet_test_wron		98.95 190	98.99 168	98.85 259	99.64 158	97.16 297	98.23 281	99.33 249	98.93 168	99.56 156	99.66 146	97.39 225	99.83 226	98.29 161	99.88 112	99.55 147

MVSTER		98.47 235	98.22 240	99.24 224	99.06 305	98.35 252	99.08 186	99.46 217	99.27 123	99.75 90	99.66 146	88.61 324	99.85 194	99.14 98	99.92 88	99.52 165

EI-MVSNet-Vis-set		99.47 76	99.49 68	99.42 179	99.57 183	98.66 236	99.24 140	99.46 217	99.67 58	99.79 79	99.65 151	98.97 76	99.89 124	99.15 92	99.89 106	99.71 49

SMA-MVS		99.23 128	99.06 148	99.74 55	99.46 231	99.76 41	99.13 177	99.58 173	97.62 270	99.68 112	99.64 152	99.02 72	99.83 226	97.61 207	99.82 152	99.63 99

APD-MVS_3200maxsize		99.31 113	99.16 121	99.74 55	99.53 200	99.75 44	99.27 133	99.61 147	99.19 139	99.57 149	99.64 152	98.76 105	99.90 109	97.29 224	99.62 221	99.56 144

ADS-MVSNet2		97.78 269	97.66 275	98.12 296	99.14 293	95.36 327	99.22 146	98.75 297	96.97 294	98.25 310	99.64 152	90.90 308	99.94 55	96.51 265	99.56 229	99.08 269

ADS-MVSNet		97.72 271	97.67 274	97.86 306	99.14 293	94.65 332	99.22 146	98.86 291	96.97 294	98.25 310	99.64 152	90.90 308	99.84 210	96.51 265	99.56 229	99.08 269

CP-MVSNet		99.54 59	99.43 78	99.87 16	99.76 103	99.82 27	99.57 68	99.61 147	99.54 85	99.80 74	99.64 152	97.79 201	99.95 41	99.21 79	99.94 77	99.84 15

FMVSNet3		98.80 209	98.63 207	99.32 207	99.13 295	98.72 232	99.10 181	99.48 210	99.23 134	99.62 138	99.64 152	92.57 294	99.86 178	98.96 115	99.90 100	99.39 209

IterMVS-LS		99.41 87	99.47 69	99.25 222	99.81 61	98.09 272	98.85 221	99.76 79	99.62 71	99.83 64	99.64 152	98.54 140	99.97 16	99.15 92	99.99 20	99.68 62
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
DeepC-MVS_fast	98.47 5	99.23 128	99.12 130	99.56 143	99.28 277	99.22 180	98.99 201	99.40 234	99.08 157	99.58 147	99.64 152	98.90 84	99.83 226	97.44 216	99.75 183	99.63 99
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
OPM-MVS		99.26 123	99.13 127	99.63 108	99.70 140	99.61 87	98.58 246	99.48 210	98.50 212	99.52 167	99.63 160	99.14 54	99.76 277	97.89 187	99.77 178	99.51 168

zzz-MVS		99.30 114	99.14 124	99.80 29	99.81 61	99.81 28	98.73 238	99.53 190	99.27 123	99.42 183	99.63 160	98.21 173	99.95 41	97.83 191	99.79 169	99.65 89

MTAPA		99.35 102	99.20 120	99.80 29	99.81 61	99.81 28	99.33 108	99.53 190	99.27 123	99.42 183	99.63 160	98.21 173	99.95 41	97.83 191	99.79 169	99.65 89

abl_6		99.36 100	99.23 117	99.75 51	99.71 133	99.74 49	99.33 108	99.76 79	99.07 158	99.65 126	99.63 160	99.09 61	99.92 83	97.13 236	99.76 180	99.58 139

APD-MVS		98.87 201	98.59 209	99.71 72	99.50 211	99.62 83	99.01 196	99.57 175	96.80 299	99.54 162	99.63 160	98.29 166	99.91 92	95.24 313	99.71 202	99.61 118
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
MG-MVS		98.52 230	98.39 227	98.94 250	99.15 292	97.39 294	98.18 284	99.21 274	98.89 172	99.23 227	99.63 160	97.37 227	99.74 287	94.22 326	99.61 225	99.69 56

FPMVS		96.32 313	95.50 320	98.79 266	99.60 168	98.17 266	98.46 265	98.80 295	97.16 289	96.28 346	99.63 160	82.19 351	99.09 350	88.45 344	98.89 295	99.10 263

pmmvs5		99.19 145	99.11 132	99.42 179	99.76 103	98.88 223	98.55 251	99.73 92	98.82 180	99.72 102	99.62 167	96.56 251	99.82 234	99.32 68	99.95 65	99.56 144

patchmatchnet-post											99.62 167	90.58 313	99.94 55

v2v482		99.50 66	99.47 69	99.58 131	99.78 88	99.25 173	99.14 172	99.58 173	99.25 130	99.81 71	99.62 167	98.24 170	99.84 210	99.83 20	99.97 47	99.64 95

test20.03		99.55 54	99.54 53	99.58 131	99.79 82	99.37 144	99.02 194	99.89 15	99.60 80	99.82 65	99.62 167	98.81 91	99.89 124	99.43 53	99.86 126	99.47 185

TSAR-MVS + GP.		99.12 158	99.04 158	99.38 192	99.34 264	99.16 190	98.15 287	99.29 259	98.18 242	99.63 131	99.62 167	99.18 50	99.68 310	98.20 167	99.74 190	99.30 230

EPNet		98.13 259	97.77 270	99.18 233	94.57 356	97.99 276	99.24 140	97.96 320	99.74 40	97.29 341	99.62 167	93.13 289	99.97 16	98.59 142	99.83 143	99.58 139
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
OMC-MVS		98.90 197	98.72 200	99.44 174	99.39 246	99.42 127	98.58 246	99.64 137	97.31 286	99.44 177	99.62 167	98.59 132	99.69 302	96.17 276	99.79 169	99.22 237

v148		99.40 90	99.41 80	99.39 190	99.76 103	98.94 213	99.09 185	99.59 165	99.17 145	99.81 71	99.61 174	98.41 156	99.69 302	99.32 68	99.94 77	99.53 157

DELS-MVS		99.34 107	99.30 101	99.48 163	99.51 206	99.36 147	98.12 291	99.53 190	99.36 115	99.41 189	99.61 174	99.22 47	99.87 158	99.21 79	99.68 207	99.20 241
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
MDTV_nov1_ep13			97.73 271		98.70 335	90.83 351	99.15 167	98.02 318	98.51 211	98.82 273	99.61 174	90.98 306	99.66 320	96.89 246	98.92 292

Regformer-3		99.41 87	99.41 80	99.40 187	99.52 202	98.70 233	99.17 159	99.44 222	99.62 71	99.75 90	99.60 177	98.90 84	99.85 194	98.89 123	99.84 133	99.65 89

Regformer-4		99.45 79	99.44 75	99.50 158	99.52 202	98.94 213	99.17 159	99.53 190	99.64 67	99.76 89	99.60 177	98.96 79	99.90 109	98.91 122	99.84 133	99.67 69

PGM-MVS		99.20 142	99.01 163	99.77 39	99.75 111	99.71 52	99.16 165	99.72 101	97.99 249	99.42 183	99.60 177	98.81 91	99.93 66	96.91 244	99.74 190	99.66 79

HyFIR lowres test		98.91 195	98.64 206	99.73 63	99.85 39	99.47 106	98.07 299	99.83 40	98.64 201	99.89 38	99.60 177	92.57 294	100.00 1	99.33 65	99.97 47	99.72 46

CSCG		99.37 97	99.29 106	99.60 125	99.71 133	99.46 110	99.43 82	99.85 29	98.79 184	99.41 189	99.60 177	98.92 81	99.92 83	98.02 180	99.92 88	99.43 202

ACMP	97.51 14	99.05 169	98.84 189	99.67 85	99.78 88	99.55 96	98.88 215	99.66 123	97.11 293	99.47 174	99.60 177	99.07 66	99.89 124	96.18 275	99.85 129	99.58 139
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
dp		96.86 297	97.07 281	96.24 336	98.68 336	90.30 355	99.19 151	98.38 314	97.35 285	98.23 312	99.59 183	87.23 329	99.82 234	96.27 273	98.73 309	98.59 299

EPMVS		96.53 307	96.32 305	97.17 324	98.18 346	92.97 339	99.39 86	89.95 357	98.21 240	98.61 292	99.59 183	86.69 338	99.72 290	96.99 241	99.23 281	98.81 292

MP-MVS-pluss		99.14 155	98.92 178	99.80 29	99.83 46	99.83 22	98.61 242	99.63 140	96.84 297	99.44 177	99.58 185	98.81 91	99.91 92	97.70 198	99.82 152	99.67 69
MP-MVS-pluss: MP-MVS-pluss. MP-MVS-pluss
MS-PatchMatch		99.00 181	98.97 172	99.09 237	99.11 300	98.19 264	98.76 234	99.33 249	98.49 213	99.44 177	99.58 185	98.21 173	99.69 302	98.20 167	99.62 221	99.39 209

LFMVS		98.46 236	98.19 245	99.26 219	99.24 282	98.52 242	99.62 56	96.94 338	99.87 13	99.31 215	99.58 185	91.04 305	99.81 253	98.68 140	99.42 257	99.45 191

VPNet		99.46 77	99.37 87	99.71 72	99.82 53	99.59 88	99.48 78	99.70 107	99.81 28	99.69 110	99.58 185	97.66 214	99.86 178	99.17 88	99.44 250	99.67 69

diffmvs		98.94 193	98.87 184	99.13 234	99.37 251	98.90 220	99.25 138	99.64 137	97.55 276	99.04 250	99.58 185	97.23 233	99.64 329	98.73 135	99.44 250	98.86 288

PMMVS2		99.48 70	99.45 73	99.57 137	99.76 103	98.99 207	98.09 295	99.90 14	98.95 165	99.78 82	99.58 185	99.57 20	99.93 66	99.48 49	99.95 65	99.79 30

PatchmatchNet		97.65 272	97.80 267	97.18 323	98.82 325	92.49 340	99.17 159	98.39 313	98.12 243	98.79 277	99.58 185	90.71 312	99.89 124	97.23 229	99.41 259	99.16 249
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
test_part3							98.74 235		97.71 265		99.57 192		99.90 109	94.47 322

ESAPD		98.87 201	98.58 211	99.74 55	99.62 165	99.67 68	98.74 235	99.53 190	97.71 265	99.55 159	99.57 192	98.40 158	99.90 109	94.47 322	99.68 207	99.66 79

Patchmatch-test		98.10 261	97.98 256	98.48 283	99.27 279	96.48 305	99.40 85	99.07 283	98.81 181	99.23 227	99.57 192	90.11 318	99.87 158	96.69 256	99.64 219	99.09 266

VNet		99.18 147	99.06 148	99.56 143	99.24 282	99.36 147	99.33 108	99.31 255	99.67 58	99.47 174	99.57 192	96.48 254	99.84 210	99.15 92	99.30 271	99.47 185

PNet_i23d		97.02 292	97.87 265	94.49 339	99.69 142	84.81 358	95.18 351	99.85 29	97.83 261	99.32 213	99.57 192	95.53 272	99.47 344	96.09 277	97.74 343	99.18 246

MSLP-MVS++		99.05 169	99.09 141	98.91 253	99.21 285	98.36 251	98.82 227	99.47 214	98.85 176	98.90 268	99.56 197	98.78 101	99.09 350	98.57 143	99.68 207	99.26 234

TranMVSNet+NR-MVSNet		99.54 59	99.47 69	99.76 42	99.58 174	99.64 77	99.30 121	99.63 140	99.61 75	99.71 106	99.56 197	98.76 105	99.96 33	99.14 98	99.92 88	99.68 62

114514_t		98.49 233	98.11 248	99.64 104	99.73 120	99.58 90	99.24 140	99.76 79	89.94 347	99.42 183	99.56 197	97.76 203	99.86 178	97.74 196	99.82 152	99.47 185

Vis-MVSNet (Re-imp)		98.77 212	98.58 211	99.34 201	99.78 88	98.88 223	99.61 60	99.56 178	99.11 152	99.24 226	99.56 197	93.00 292	99.78 269	97.43 217	99.89 106	99.35 221

test_0402		99.22 137	99.14 124	99.45 172	99.79 82	99.43 123	99.28 130	99.68 116	99.54 85	99.40 193	99.56 197	99.07 66	99.82 234	96.01 283	99.96 59	99.11 259

tpmvs		97.39 277	97.69 272	96.52 332	98.41 341	91.76 345	99.30 121	98.94 290	97.74 263	97.85 330	99.55 202	92.40 297	99.73 289	96.25 274	98.73 309	98.06 323

MSDG		99.08 164	98.98 171	99.37 196	99.60 168	99.13 193	97.54 327	99.74 89	98.84 179	99.53 165	99.55 202	99.10 59	99.79 261	97.07 238	99.86 126	99.18 246

tpmrst		97.73 270	98.07 250	96.73 328	98.71 334	92.00 342	99.10 181	98.86 291	98.52 210	98.92 265	99.54 204	91.90 298	99.82 234	98.02 180	99.03 289	98.37 309

new_pmnet		98.88 200	98.89 182	98.84 261	99.70 140	97.62 289	98.15 287	99.50 205	97.98 250	99.62 138	99.54 204	98.15 178	99.94 55	97.55 210	99.84 133	98.95 281

Anonymous20231206		99.35 102	99.31 96	99.47 165	99.74 117	99.06 205	99.28 130	99.74 89	99.23 134	99.72 102	99.53 206	97.63 216	99.88 139	99.11 100	99.84 133	99.48 181

ITE_SJBPF				99.38 192	99.63 160	99.44 117		99.73 92	98.56 207	99.33 211	99.53 206	98.88 87	99.68 310	96.01 283	99.65 218	99.02 278

CHOSEN 280x420		98.41 241	98.41 225	98.40 286	99.34 264	95.89 318	96.94 340	99.44 222	98.80 183	99.25 223	99.52 208	93.51 286	99.98 7	98.94 120	99.98 36	99.32 228

Patchmatch-test1		98.13 259	98.40 226	97.31 322	99.20 288	92.99 338	98.17 286	98.49 309	98.24 239	99.10 244	99.52 208	96.01 266	99.83 226	97.22 230	99.62 221	99.12 258

Regformer-1		99.32 112	99.27 110	99.47 165	99.41 242	98.95 212	98.99 201	99.48 210	99.48 92	99.66 120	99.52 208	98.78 101	99.87 158	98.36 154	99.74 190	99.60 124

Regformer-2		99.34 107	99.27 110	99.53 151	99.41 242	99.10 198	98.99 201	99.53 190	99.47 96	99.66 120	99.52 208	98.80 95	99.89 124	98.31 159	99.74 190	99.60 124

CANet_DTU		98.91 195	98.85 187	99.09 237	98.79 327	98.13 267	98.18 284	99.31 255	99.48 92	98.86 271	99.51 212	96.56 251	99.95 41	99.05 104	99.95 65	99.19 243

pmmvs3		98.08 262	97.80 267	98.91 253	99.41 242	97.69 287	97.87 319	99.66 123	95.87 315	99.50 171	99.51 212	90.35 316	99.97 16	98.55 145	99.47 247	99.08 269

HY-MVS	98.23 9	98.21 257	97.95 258	98.99 247	99.03 308	98.24 260	99.61 60	98.72 299	96.81 298	98.73 282	99.51 212	94.06 282	99.86 178	96.91 244	98.20 332	98.86 288

mPP-MVS		99.19 145	99.00 165	99.76 42	99.76 103	99.68 66	99.38 92	99.54 185	98.34 233	99.01 252	99.50 215	98.53 144	99.93 66	97.18 234	99.78 174	99.66 79

HPM-MVS_fast		99.43 81	99.30 101	99.80 29	99.83 46	99.81 28	99.52 72	99.70 107	98.35 229	99.51 169	99.50 215	99.31 35	99.88 139	98.18 171	99.84 133	99.69 56

MP-MVS		99.06 166	98.83 192	99.76 42	99.76 103	99.71 52	99.32 111	99.50 205	98.35 229	98.97 255	99.48 217	98.37 161	99.92 83	95.95 289	99.75 183	99.63 99

test1235678		98.93 194	98.84 189	99.19 230	99.46 231	98.55 240	97.53 329	99.77 73	98.76 190	99.69 110	99.48 217	96.69 248	99.90 109	98.30 160	99.91 98	99.11 259

MVS_111021_LR		99.13 156	99.03 159	99.42 179	99.58 174	99.32 156	97.91 318	99.73 92	98.68 198	99.31 215	99.48 217	99.09 61	99.66 320	97.70 198	99.77 178	99.29 233

XVS		99.27 122	99.11 132	99.75 51	99.71 133	99.71 52	99.37 96	99.61 147	99.29 120	98.76 280	99.47 220	98.47 150	99.88 139	97.62 205	99.73 196	99.67 69

EPNet_dtu		97.62 273	97.79 269	97.11 325	96.67 355	92.31 341	98.51 257	98.04 317	99.24 132	95.77 350	99.47 220	93.78 284	99.66 320	98.98 110	99.62 221	99.37 216
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
MVS_111021_HR		99.12 158	99.02 160	99.40 187	99.50 211	99.11 195	97.92 316	99.71 104	98.76 190	99.08 245	99.47 220	99.17 51	99.54 340	97.85 190	99.76 180	99.54 154

tpm cat1		96.78 302	96.98 285	96.16 337	98.85 321	90.59 354	99.08 186	99.32 251	92.37 342	97.73 337	99.46 223	91.15 304	99.69 302	96.07 279	98.80 300	98.21 318

PHI-MVS		99.11 161	98.95 174	99.59 127	99.13 295	99.59 88	99.17 159	99.65 132	97.88 255	99.25 223	99.46 223	98.97 76	99.80 258	97.26 227	99.82 152	99.37 216

pmmvs4		99.13 156	99.06 148	99.36 199	99.57 183	99.10 198	98.01 303	99.25 268	98.78 186	99.58 147	99.44 225	98.24 170	99.76 277	98.74 134	99.93 85	99.22 237

1111		97.29 280	96.71 299	99.04 244	99.65 156	97.72 284	98.35 272	99.80 60	99.40 109	99.66 120	99.43 226	75.10 360	99.87 158	98.98 110	99.98 36	99.52 165

.test1245		85.84 329	89.27 330	75.54 342	99.65 156	97.72 284	98.35 272	99.80 60	99.40 109	99.66 120	99.43 226	75.10 360	99.87 158	98.98 110	33.07 354	29.03 355

XVG-OURS-SEG-HR		99.16 152	98.99 168	99.66 93	99.84 42	99.64 77	98.25 280	99.73 92	98.39 221	99.63 131	99.43 226	99.70 12	99.90 109	97.34 221	98.64 312	99.44 196

CNVR-MVS		98.99 183	98.80 196	99.56 143	99.25 280	99.43 123	98.54 254	99.27 263	98.58 206	98.80 276	99.43 226	98.53 144	99.70 296	97.22 230	99.59 227	99.54 154

LF4IMVS		99.01 178	98.92 178	99.27 214	99.71 133	99.28 163	98.59 245	99.77 73	98.32 235	99.39 194	99.41 230	98.62 128	99.84 210	96.62 261	99.84 133	98.69 296

testdata				99.42 179	99.51 206	98.93 217		99.30 258	96.20 310	98.87 270	99.40 231	98.33 165	99.89 124	96.29 272	99.28 273	99.44 196

Test_1112_low_res		98.95 190	98.73 199	99.63 108	99.68 149	99.15 192	98.09 295	99.80 60	97.14 290	99.46 176	99.40 231	96.11 264	99.89 124	99.01 107	99.84 133	99.84 15

PCF-MVS	96.03 18	96.73 303	95.86 315	99.33 203	99.44 236	99.16 190	96.87 341	99.44 222	86.58 349	98.95 261	99.40 231	94.38 280	99.88 139	87.93 346	99.80 166	98.95 281
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
旧先验1					99.49 216	99.29 161		99.26 265			99.39 234	97.67 210			99.36 265	99.46 189

ACMMPR		99.23 128	99.06 148	99.76 42	99.74 117	99.69 63	99.31 118	99.59 165	98.36 224	99.35 205	99.38 235	98.61 130	99.93 66	97.43 217	99.75 183	99.67 69

HFP-MVS		99.25 124	99.08 142	99.76 42	99.73 120	99.70 59	99.31 118	99.59 165	98.36 224	99.36 203	99.37 236	98.80 95	99.91 92	97.43 217	99.75 183	99.68 62

#test#		99.12 158	98.90 181	99.76 42	99.73 120	99.70 59	99.10 181	99.59 165	97.60 272	99.36 203	99.37 236	98.80 95	99.91 92	96.84 248	99.75 183	99.68 62

test12356		98.43 238	98.39 227	98.55 278	99.46 231	96.36 307	97.32 336	99.81 56	97.60 272	99.62 138	99.37 236	94.57 278	99.89 124	97.80 193	99.92 88	99.40 207

CPTT-MVS		98.74 214	98.44 220	99.64 104	99.61 167	99.38 141	99.18 152	99.55 181	96.49 307	99.27 220	99.37 236	97.11 240	99.92 83	95.74 296	99.67 213	99.62 113

DP-MVS Recon		98.50 231	98.23 239	99.31 209	99.49 216	99.46 110	98.56 250	99.63 140	94.86 332	98.85 272	99.37 236	97.81 199	99.59 337	96.08 278	99.44 250	98.88 286

region2R		99.23 128	99.05 153	99.77 39	99.76 103	99.70 59	99.31 118	99.59 165	98.41 219	99.32 213	99.36 241	98.73 111	99.93 66	97.29 224	99.74 190	99.67 69

DU-MVS		99.33 110	99.21 119	99.71 72	99.43 238	99.56 93	98.83 224	99.53 190	99.38 112	99.67 116	99.36 241	97.67 210	99.95 41	99.17 88	99.81 161	99.63 99

UniMVSNet (Re)		99.37 97	99.26 112	99.68 82	99.51 206	99.58 90	98.98 205	99.60 161	99.43 106	99.70 108	99.36 241	97.70 205	99.88 139	99.20 82	99.87 119	99.59 135

NR-MVSNet		99.40 90	99.31 96	99.68 82	99.43 238	99.55 96	99.73 21	99.50 205	99.46 99	99.88 46	99.36 241	97.54 217	99.87 158	98.97 114	99.87 119	99.63 99

UnsupCasMVSNet_eth		98.83 204	98.57 213	99.59 127	99.68 149	99.45 115	98.99 201	99.67 119	99.48 92	99.55 159	99.36 241	94.92 274	99.86 178	98.95 119	96.57 348	99.45 191

UnsupCasMVSNet_bld		98.55 228	98.27 237	99.40 187	99.56 194	99.37 144	97.97 311	99.68 116	97.49 279	99.08 245	99.35 246	95.41 273	99.82 234	97.70 198	98.19 334	99.01 279

sss		98.90 197	98.77 197	99.27 214	99.48 222	98.44 244	98.72 239	99.32 251	97.94 253	99.37 202	99.35 246	96.31 259	99.91 92	98.85 125	99.63 220	99.47 185

CostFormer		96.71 304	96.79 292	96.46 333	98.90 312	90.71 352	99.41 83	98.68 301	94.69 335	98.14 318	99.34 248	86.32 345	99.80 258	97.60 208	98.07 337	98.88 286

原ACMM1				99.37 196	99.47 227	98.87 225		99.27 263	96.74 300	98.26 309	99.32 249	97.93 191	99.82 234	95.96 288	99.38 262	99.43 202

tpm		97.15 289	96.95 286	97.75 310	98.91 311	94.24 334	99.32 111	97.96 320	97.71 265	98.29 307	99.32 249	86.72 337	99.92 83	98.10 178	96.24 350	99.09 266

test222					99.51 206	99.08 201	97.83 321	99.29 259	95.21 327	98.68 288	99.31 251	97.28 230			99.38 262	99.43 202

tpmp4_e23		96.11 317	96.06 310	96.27 334	98.90 312	90.70 353	99.34 106	99.03 287	93.72 339	96.56 345	99.31 251	83.63 349	99.75 283	96.06 280	98.02 339	98.35 310

BH-RMVSNet		98.41 241	98.14 247	99.21 227	99.21 285	98.47 243	98.60 244	98.26 316	98.35 229	98.93 263	99.31 251	97.20 237	99.66 320	94.32 324	99.10 285	99.51 168

MVS_0304		99.17 150	99.10 139	99.38 192	99.08 303	98.86 226	98.46 265	99.73 92	99.53 87	99.35 205	99.30 254	97.11 240	99.96 33	99.33 65	99.99 20	99.33 224

MVSFormer		99.41 87	99.44 75	99.31 209	99.57 183	98.40 247	99.77 13	99.80 60	99.73 42	99.63 131	99.30 254	98.02 185	99.98 7	99.43 53	99.69 205	99.55 147

jason		99.16 152	99.11 132	99.32 207	99.75 111	98.44 244	98.26 279	99.39 237	98.70 197	99.74 98	99.30 254	98.54 140	99.97 16	98.48 148	99.82 152	99.55 147
jason: jason.
1121		98.56 226	98.24 238	99.52 153	99.49 216	99.24 176	99.30 121	99.22 273	95.77 318	98.52 298	99.29 257	97.39 225	99.85 194	95.79 294	99.34 266	99.46 189

新几何1				99.52 153	99.50 211	99.22 180		99.26 265	95.66 322	98.60 293	99.28 258	97.67 210	99.89 124	95.95 289	99.32 269	99.45 191

UniMVSNet_NR-MVSNet		99.37 97	99.25 114	99.72 68	99.47 227	99.56 93	98.97 206	99.61 147	99.43 106	99.67 116	99.28 258	97.85 197	99.95 41	99.17 88	99.81 161	99.65 89

CP-MVS		99.23 128	99.05 153	99.75 51	99.66 154	99.66 71	99.38 92	99.62 143	98.38 222	99.06 249	99.27 260	98.79 98	99.94 55	97.51 213	99.82 152	99.66 79

AdaColmap		98.60 222	98.35 232	99.38 192	99.12 297	99.22 180	98.67 241	99.42 227	97.84 260	98.81 274	99.27 260	97.32 229	99.81 253	95.14 314	99.53 240	99.10 263

NCCC		98.82 206	98.57 213	99.58 131	99.21 285	99.31 157	98.61 242	99.25 268	98.65 200	98.43 304	99.26 262	97.86 196	99.81 253	96.55 263	99.27 276	99.61 118

TAPA-MVS	97.92 13	98.03 264	97.55 276	99.46 168	99.47 227	99.44 117	98.50 258	99.62 143	86.79 348	99.07 248	99.26 262	98.26 169	99.62 332	97.28 226	99.73 196	99.31 229
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
MCST-MVS		99.02 174	98.81 194	99.65 97	99.58 174	99.49 102	98.58 246	99.07 283	98.40 220	99.04 250	99.25 264	98.51 148	99.80 258	97.31 223	99.51 242	99.65 89

HQP_MVS		98.90 197	98.68 203	99.55 146	99.58 174	99.24 176	98.80 229	99.54 185	98.94 166	99.14 240	99.25 264	97.24 231	99.82 234	95.84 292	99.78 174	99.60 124

plane_prior4											99.25 264

HPM-MVS		99.25 124	99.07 146	99.78 37	99.81 61	99.75 44	99.61 60	99.67 119	97.72 264	99.35 205	99.25 264	99.23 46	99.92 83	97.21 232	99.82 152	99.67 69
Chunlin Ren, Qingshan Xu, Shikun Zhang, Jiaqi Yang: Hierarchical Prior Mining for Non-local Multi-View Stereo. ICCV 2023
PatchMatch-RL		98.68 218	98.47 218	99.30 211	99.44 236	99.28 163	98.14 289	99.54 185	97.12 292	99.11 243	99.25 264	97.80 200	99.70 296	96.51 265	99.30 271	98.93 283

Effi-MVS+-dtu		99.07 165	98.92 178	99.52 153	98.89 316	99.78 35	99.15 167	99.66 123	99.34 116	98.92 265	99.24 269	97.69 207	99.98 7	98.11 176	99.28 273	98.81 292

test_normal		98.82 206	98.67 204	99.27 214	99.56 194	98.83 228	98.22 282	98.01 319	99.03 160	99.49 173	99.24 269	96.21 261	99.76 277	98.69 138	99.56 229	99.22 237

WTY-MVS		98.59 224	98.37 230	99.26 219	99.43 238	98.40 247	98.74 235	99.13 282	98.10 244	99.21 231	99.24 269	94.82 276	99.90 109	97.86 189	98.77 303	99.49 180

DI_MVS_plusplus_test		98.80 209	98.65 205	99.27 214	99.57 183	98.90 220	98.44 267	97.95 322	99.02 161	99.51 169	99.23 272	96.18 263	99.76 277	98.52 147	99.42 257	99.14 254

CANet		99.11 161	99.05 153	99.28 212	98.83 322	98.56 239	98.71 240	99.41 228	99.25 130	99.23 227	99.22 273	97.66 214	99.94 55	99.19 83	99.97 47	99.33 224

tpm2		96.35 312	96.22 306	96.73 328	98.88 319	91.75 346	99.21 149	98.51 307	93.27 341	97.89 327	99.21 274	84.83 348	99.70 296	96.04 281	98.18 335	98.75 295

WR-MVS		99.11 161	98.93 175	99.66 93	99.30 274	99.42 127	98.42 269	99.37 243	99.04 159	99.57 149	99.20 275	96.89 246	99.86 178	98.66 141	99.87 119	99.70 53

F-COLMAP		98.74 214	98.45 219	99.62 116	99.57 183	99.47 106	98.84 222	99.65 132	96.31 309	98.93 263	99.19 276	97.68 209	99.87 158	96.52 264	99.37 264	99.53 157

1112_ss		99.05 169	98.84 189	99.67 85	99.66 154	99.29 161	98.52 256	99.82 48	97.65 269	99.43 181	99.16 277	96.42 257	99.91 92	99.07 103	99.84 133	99.80 25

ab-mvs-re		8.26 340	11.02 341	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	99.16 277	0.00 364	0.00 359	0.00 356	0.00 357	0.00 357

cdsmvs_eth3d_5k		24.88 333	33.17 333	0.00 346	0.00 360	0.00 361	0.00 352	99.62 143	0.00 356	0.00 357	99.13 279	99.82 6	0.00 359	0.00 356	0.00 357	0.00 357

lupinMVS		98.96 187	98.87 184	99.24 224	99.57 183	98.40 247	98.12 291	99.18 276	98.28 237	99.63 131	99.13 279	98.02 185	99.97 16	98.22 165	99.69 205	99.35 221

PVSNet	97.47 15	98.42 240	98.44 220	98.35 288	99.46 231	96.26 308	96.70 344	99.34 248	97.68 268	99.00 253	99.13 279	97.40 223	99.72 290	97.59 209	99.68 207	99.08 269

CLD-MVS		98.76 213	98.57 213	99.33 203	99.57 183	98.97 210	97.53 329	99.55 181	96.41 308	99.27 220	99.13 279	99.07 66	99.78 269	96.73 255	99.89 106	99.23 236
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
1314		98.00 265	97.90 263	98.27 293	98.90 312	97.45 293	99.30 121	99.06 285	94.98 329	97.21 342	99.12 283	98.43 154	99.67 315	95.58 305	98.56 322	97.71 336

E-PMN		97.14 291	97.43 277	96.27 334	98.79 327	91.62 347	95.54 348	99.01 288	99.44 101	98.88 269	99.12 283	92.78 293	99.68 310	94.30 325	99.03 289	97.50 338

CDPH-MVS		98.56 226	98.20 242	99.61 119	99.50 211	99.46 110	98.32 276	99.41 228	95.22 326	99.21 231	99.10 285	98.34 163	99.82 234	95.09 316	99.66 216	99.56 144

conf0.01		97.19 287	96.74 293	98.51 279	99.73 120	98.35 252	99.35 99	95.78 344	96.54 301	99.39 194	99.08 286	86.57 339	99.88 139	95.69 297	98.57 315	97.30 342

conf0.002		97.19 287	96.74 293	98.51 279	99.73 120	98.35 252	99.35 99	95.78 344	96.54 301	99.39 194	99.08 286	86.57 339	99.88 139	95.69 297	98.57 315	97.30 342

thresconf0.02		97.25 282	96.74 293	98.75 269	99.73 120	98.35 252	99.35 99	95.78 344	96.54 301	99.39 194	99.08 286	86.57 339	99.88 139	95.69 297	98.57 315	98.02 325

tfpn_n400		97.25 282	96.74 293	98.75 269	99.73 120	98.35 252	99.35 99	95.78 344	96.54 301	99.39 194	99.08 286	86.57 339	99.88 139	95.69 297	98.57 315	98.02 325

tfpnconf		97.25 282	96.74 293	98.75 269	99.73 120	98.35 252	99.35 99	95.78 344	96.54 301	99.39 194	99.08 286	86.57 339	99.88 139	95.69 297	98.57 315	98.02 325

tfpnview11		97.25 282	96.74 293	98.75 269	99.73 120	98.35 252	99.35 99	95.78 344	96.54 301	99.39 194	99.08 286	86.57 339	99.88 139	95.69 297	98.57 315	98.02 325

MVS		95.72 325	94.63 328	98.99 247	98.56 338	97.98 281	99.30 121	98.86 291	72.71 353	97.30 340	99.08 286	98.34 163	99.74 287	89.21 342	98.33 329	99.26 234

tfpn1000		97.28 281	96.83 290	98.64 276	99.67 153	97.68 288	99.41 83	95.47 351	97.14 290	99.43 181	99.07 293	85.87 346	99.88 139	96.78 251	98.67 311	98.34 311

HPM-MVS++		98.96 187	98.70 201	99.74 55	99.52 202	99.71 52	98.86 218	99.19 275	98.47 215	98.59 294	99.06 294	98.08 181	99.91 92	96.94 243	99.60 226	99.60 124

Fast-Effi-MVS+-dtu		99.20 142	99.12 130	99.43 177	99.25 280	99.69 63	99.05 189	99.82 48	99.50 90	98.97 255	99.05 295	98.98 74	99.98 7	98.20 167	99.24 279	98.62 297

test_prior3		98.62 220	98.34 233	99.46 168	99.35 254	99.22 180	97.95 312	99.39 237	97.87 256	98.05 320	99.05 295	97.90 192	99.69 302	95.99 285	99.49 245	99.48 181

test_prior2							97.95 312		97.87 256	98.05 320	99.05 295	97.90 192		95.99 285	99.49 245

testgi		99.29 116	99.26 112	99.37 196	99.75 111	98.81 229	98.84 222	99.89 15	98.38 222	99.75 90	99.04 298	99.36 33	99.86 178	99.08 102	99.25 277	99.45 191

HSP-MVS		99.01 178	98.76 198	99.76 42	99.78 88	99.73 50	99.35 99	99.31 255	98.54 209	99.54 162	98.99 299	96.81 247	99.93 66	96.97 242	99.53 240	99.61 118

TEST9					99.35 254	99.35 151	98.11 293	99.41 228	94.83 334	97.92 325	98.99 299	98.02 185	99.85 194

train_agg		98.35 248	97.95 258	99.57 137	99.35 254	99.35 151	98.11 293	99.41 228	94.90 330	97.92 325	98.99 299	98.02 185	99.85 194	95.38 311	99.44 250	99.50 174

agg_prior1		98.33 251	97.92 260	99.57 137	99.35 254	99.36 147	97.99 307	99.39 237	94.85 333	97.76 335	98.98 302	98.03 183	99.85 194	95.49 306	99.44 250	99.51 168

PVSNet_Blended		98.70 217	98.59 209	99.02 246	99.54 197	97.99 276	97.58 326	99.82 48	95.70 320	99.34 209	98.98 302	98.52 146	99.77 275	97.98 183	99.83 143	99.30 230

CNLPA		98.57 225	98.34 233	99.28 212	99.18 291	99.10 198	98.34 274	99.41 228	98.48 214	98.52 298	98.98 302	97.05 242	99.78 269	95.59 304	99.50 243	98.96 280

test_8					99.34 264	99.31 157	98.08 298	99.40 234	94.90 330	97.87 329	98.97 305	98.02 185	99.84 210

GA-MVS		97.99 266	97.68 273	98.93 252	99.52 202	98.04 275	97.19 338	99.05 286	98.32 235	98.81 274	98.97 305	89.89 321	99.41 348	98.33 157	99.05 287	99.34 223

agg_prior3		98.24 253	97.81 266	99.53 151	99.34 264	99.26 169	98.09 295	99.39 237	94.21 338	97.77 334	98.96 307	97.74 204	99.84 210	95.38 311	99.44 250	99.50 174

PLC	97.35 16	98.36 245	97.99 254	99.48 163	99.32 269	99.24 176	98.50 258	99.51 202	95.19 328	98.58 295	98.96 307	96.95 245	99.83 226	95.63 303	99.25 277	99.37 216
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
Test4		98.65 219	98.44 220	99.27 214	99.57 183	98.86 226	98.43 268	99.41 228	98.85 176	99.57 149	98.95 309	93.05 290	99.75 283	98.57 143	99.56 229	99.19 243

xiu_mvs_v1_base_debu		99.23 128	99.34 92	98.91 253	99.59 171	98.23 261	98.47 261	99.66 123	99.61 75	99.68 112	98.94 310	99.39 24	99.97 16	99.18 85	99.55 235	98.51 304

mvs-test1		98.83 204	98.70 201	99.22 226	98.89 316	99.65 75	98.88 215	99.66 123	99.34 116	98.29 307	98.94 310	97.69 207	99.96 33	98.11 176	98.54 323	98.04 324

Effi-MVS+		99.06 166	98.97 172	99.34 201	99.31 270	98.98 208	98.31 277	99.91 11	98.81 181	98.79 277	98.94 310	99.14 54	99.84 210	98.79 129	98.74 307	99.20 241

xiu_mvs_v1_base		99.23 128	99.34 92	98.91 253	99.59 171	98.23 261	98.47 261	99.66 123	99.61 75	99.68 112	98.94 310	99.39 24	99.97 16	99.18 85	99.55 235	98.51 304

xiu_mvs_v1_base_debi		99.23 128	99.34 92	98.91 253	99.59 171	98.23 261	98.47 261	99.66 123	99.61 75	99.68 112	98.94 310	99.39 24	99.97 16	99.18 85	99.55 235	98.51 304

EMVS		96.96 294	97.28 278	95.99 338	98.76 331	91.03 350	95.26 350	98.61 303	99.34 116	98.92 265	98.88 315	93.79 283	99.66 320	92.87 331	99.05 287	97.30 342

NP-MVS					99.40 245	99.13 193					98.83 316

HQP-MVS		98.36 245	98.02 253	99.39 190	99.31 270	98.94 213	97.98 308	99.37 243	97.45 280	98.15 314	98.83 316	96.67 249	99.70 296	94.73 318	99.67 213	99.53 157

MAR-MVS		98.24 253	97.92 260	99.19 230	98.78 329	99.65 75	99.17 159	99.14 280	95.36 324	98.04 322	98.81 318	97.47 220	99.72 290	95.47 308	99.06 286	98.21 318
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
tfpn_ndepth		96.93 296	96.43 304	98.42 284	99.60 168	97.72 284	99.22 146	95.16 352	95.91 314	99.26 222	98.79 319	85.56 347	99.87 158	96.03 282	98.35 328	97.68 337

API-MVS		98.38 244	98.39 227	98.35 288	98.83 322	99.26 169	99.14 172	99.18 276	98.59 205	98.66 289	98.78 320	98.61 130	99.57 339	94.14 327	99.56 229	96.21 349

testpf		94.48 328	95.31 322	91.99 341	97.22 353	89.64 356	98.86 218	96.52 339	94.36 337	96.09 349	98.76 321	82.21 350	98.73 352	97.05 239	96.74 347	87.60 352

BH-untuned		98.22 256	98.09 249	98.58 277	99.38 249	97.24 296	98.55 251	98.98 289	97.81 262	99.20 236	98.76 321	97.01 243	99.65 327	94.83 317	98.33 329	98.86 288

Fast-Effi-MVS+		99.02 174	98.87 184	99.46 168	99.38 249	99.50 100	99.04 191	99.79 68	97.17 288	98.62 291	98.74 323	99.34 34	99.95 41	98.32 158	99.41 259	98.92 284

MVE	92.54 22	96.66 305	96.11 309	98.31 291	99.68 149	97.55 291	97.94 314	95.60 350	99.37 113	90.68 354	98.70 324	96.56 251	98.61 354	86.94 352	99.55 235	98.77 294
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
PAPM		95.61 326	94.71 327	98.31 291	99.12 297	96.63 303	96.66 345	98.46 310	90.77 346	96.25 347	98.68 325	93.01 291	99.69 302	81.60 353	97.86 342	98.62 297

test-LLR		97.15 289	96.95 286	97.74 311	98.18 346	95.02 330	97.38 332	96.10 340	98.00 247	97.81 331	98.58 326	90.04 319	99.91 92	97.69 203	98.78 301	98.31 312

test-mter		96.23 316	95.73 318	97.74 311	98.18 346	95.02 330	97.38 332	96.10 340	97.90 254	97.81 331	98.58 326	79.12 358	99.91 92	97.69 203	98.78 301	98.31 312

PAPM_NR		98.36 245	98.04 252	99.33 203	99.48 222	98.93 217	98.79 232	99.28 262	97.54 277	98.56 297	98.57 328	97.12 239	99.69 302	94.09 328	98.90 294	99.38 212

TESTMET0.1,1		96.24 315	95.84 316	97.41 319	98.24 344	93.84 335	97.38 332	95.84 343	98.43 216	97.81 331	98.56 329	79.77 357	99.89 124	97.77 194	98.77 303	98.52 303

xiu_mvs_v2_base		99.02 174	99.11 132	98.77 267	99.37 251	98.09 272	98.13 290	99.51 202	99.47 96	99.42 183	98.54 330	99.38 28	99.97 16	98.83 126	99.33 268	98.24 316

TR-MVS		97.44 276	97.15 280	98.32 290	98.53 339	97.46 292	98.47 261	97.91 323	96.85 296	98.21 313	98.51 331	96.42 257	99.51 342	92.16 333	97.29 345	97.98 329

PS-MVSNAJ		99.00 181	99.08 142	98.76 268	99.37 251	98.10 271	98.00 305	99.51 202	99.47 96	99.41 189	98.50 332	99.28 39	99.97 16	98.83 126	99.34 266	98.20 320

testus		98.15 258	98.06 251	98.40 286	99.11 300	95.95 313	96.77 342	99.89 15	95.83 316	99.23 227	98.47 333	97.50 219	99.84 210	96.58 262	99.20 282	99.39 209

gm-plane-assit					97.59 350	89.02 357			93.47 340		98.30 334		99.84 210	96.38 269

DeepMVS_CX				97.98 298	99.69 142	96.95 300		99.26 265	75.51 352	95.74 351	98.28 335	96.47 255	99.62 332	91.23 336	97.89 341	97.38 340

PAPR		97.56 275	97.07 281	99.04 244	98.80 326	98.11 270	97.63 324	99.25 268	94.56 336	98.02 323	98.25 336	97.43 222	99.68 310	90.90 337	98.74 307	99.33 224

PMMVS		98.49 233	98.29 236	99.11 235	98.96 309	98.42 246	97.54 327	99.32 251	97.53 278	98.47 303	98.15 337	97.88 195	99.82 234	97.46 215	99.24 279	99.09 266

test0.0.03 1		97.37 278	96.91 288	98.74 273	97.72 349	97.57 290	97.60 325	97.36 337	98.00 247	99.21 231	98.02 338	90.04 319	99.79 261	98.37 153	95.89 351	98.86 288

BH-w/o		97.20 286	97.01 284	97.76 309	99.08 303	95.69 323	98.03 302	98.52 306	95.76 319	97.96 324	98.02 338	95.62 270	99.47 344	92.82 332	97.25 346	98.12 322

alignmvs		98.28 252	97.96 257	99.25 222	99.12 297	98.93 217	99.03 193	98.42 312	99.64 67	98.72 283	97.85 340	90.86 310	99.62 332	98.88 124	99.13 283	99.19 243

test2356		95.99 321	95.26 324	98.18 294	96.93 354	95.53 326	95.31 349	98.71 300	95.67 321	98.48 302	97.83 341	80.72 354	99.88 139	95.47 308	98.21 331	99.11 259

view600		96.86 297	96.52 300	97.88 302	99.69 142	95.87 319	99.39 86	97.68 326	99.11 152	98.96 257	97.82 342	87.40 325	99.79 261	89.78 338	98.83 296	97.98 329

view800		96.86 297	96.52 300	97.88 302	99.69 142	95.87 319	99.39 86	97.68 326	99.11 152	98.96 257	97.82 342	87.40 325	99.79 261	89.78 338	98.83 296	97.98 329

conf0.05thres1000		96.86 297	96.52 300	97.88 302	99.69 142	95.87 319	99.39 86	97.68 326	99.11 152	98.96 257	97.82 342	87.40 325	99.79 261	89.78 338	98.83 296	97.98 329

tfpn		96.86 297	96.52 300	97.88 302	99.69 142	95.87 319	99.39 86	97.68 326	99.11 152	98.96 257	97.82 342	87.40 325	99.79 261	89.78 338	98.83 296	97.98 329

PVSNet_0	95.53 19	95.85 323	95.31 322	97.47 317	98.78 329	93.48 337	95.72 347	99.40 234	96.18 311	97.37 339	97.73 346	95.73 268	99.58 338	95.49 306	81.40 353	99.36 219

canonicalmvs		99.02 174	99.00 165	99.09 237	99.10 302	98.70 233	99.61 60	99.66 123	99.63 70	98.64 290	97.65 347	99.04 70	99.54 340	98.79 129	98.92 292	99.04 276

DWT-MVSNet_test		96.03 320	95.80 317	96.71 330	98.50 340	91.93 343	99.25 138	97.87 324	95.99 313	96.81 344	97.61 348	81.02 353	99.66 320	97.20 233	97.98 340	98.54 302

cascas		96.99 293	96.82 291	97.48 316	97.57 352	95.64 324	96.43 346	99.56 178	91.75 343	97.13 343	97.61 348	95.58 271	98.63 353	96.68 257	99.11 284	98.18 321

IB-MVS	95.41 20	95.30 327	94.46 329	97.84 307	98.76 331	95.33 328	97.33 335	96.07 342	96.02 312	95.37 352	97.41 350	76.17 359	99.96 33	97.54 211	95.44 352	98.22 317
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
thres600view7		96.60 306	96.16 307	97.93 300	99.63 160	96.09 312	99.18 152	97.57 330	98.77 187	98.72 283	97.32 351	87.04 330	99.72 290	88.57 343	98.62 313	97.98 329

tfpn111		96.50 308	96.12 308	97.65 313	99.63 160	95.93 314	99.18 152	97.57 330	98.75 192	98.70 285	97.31 352	87.04 330	99.72 290	88.27 345	98.61 314	97.30 342

conf200view11		96.43 309	96.03 311	97.63 314	99.63 160	95.93 314	99.18 152	97.57 330	98.75 192	98.70 285	97.31 352	87.04 330	99.67 315	87.62 347	98.51 324	97.30 342

thres100view900		96.39 311	96.03 311	97.47 317	99.63 160	95.93 314	99.18 152	97.57 330	98.75 192	98.70 285	97.31 352	87.04 330	99.67 315	87.62 347	98.51 324	96.81 347

GG-mvs-BLEND				97.36 320	97.59 350	96.87 302	99.70 29	88.49 359		94.64 353	97.26 355	80.66 355	99.12 349	91.50 335	96.50 349	96.08 351

PatchFormer-LS_test		96.95 295	97.07 281	96.62 331	98.76 331	91.85 344	99.18 152	98.45 311	97.29 287	97.73 337	97.22 356	88.77 323	99.76 277	98.13 175	98.04 338	98.25 315

tfpn200view9		96.30 314	95.89 313	97.53 315	99.58 174	96.11 310	99.00 198	97.54 335	98.43 216	98.52 298	96.98 357	86.85 334	99.67 315	87.62 347	98.51 324	96.81 347

thres400		96.40 310	95.89 313	97.92 301	99.58 174	96.11 310	99.00 198	97.54 335	98.43 216	98.52 298	96.98 357	86.85 334	99.67 315	87.62 347	98.51 324	97.98 329

thres200		96.09 318	95.68 319	97.33 321	99.48 222	96.22 309	98.53 255	97.57 330	98.06 246	98.37 306	96.73 359	86.84 336	99.61 336	86.99 351	98.57 315	96.16 350

X-MVStestdata		96.09 318	94.87 326	99.75 51	99.71 133	99.71 52	99.37 96	99.61 147	99.29 120	98.76 280	61.30 360	98.47 150	99.88 139	97.62 205	99.73 196	99.67 69

test_post											52.41 361	90.25 317	99.86 178

test_post1							99.14 172				51.63 362	89.54 322	99.82 234	96.86 247

testmvs		28.94 332	33.33 332	15.79 345	26.03 358	9.81 360	96.77 342	15.67 360	11.55 355	23.87 356	50.74 363	19.03 363	8.53 358	23.21 355	33.07 354	29.03 355

test123		29.31 331	33.05 334	18.08 344	25.93 359	12.24 359	97.53 329	10.93 361	11.78 354	24.21 355	50.08 364	21.04 362	8.60 357	23.51 354	32.43 356	33.39 354

GSMVS																99.14 254

test_part2					99.62 165	99.67 68				99.55 159

test_part1								99.53 190				98.40 158			99.68 207	99.66 79

sam_mvs1												90.81 311				99.14 254

sam_mvs												90.52 315

MTGPA								99.53 190

MTMP								98.59 305

test9_res														95.10 315	99.44 250	99.50 174

agg_prior2														94.58 321	99.46 249	99.50 174

agg_prior					99.35 254	99.36 147		99.39 237		97.76 335			99.85 194

test_prior4						99.19 188	98.00 305

test_prior				99.46 168	99.35 254	99.22 180		99.39 237					99.69 302			99.48 181

旧先验2							97.94 314		95.33 325	98.94 262			99.88 139	96.75 253

新几何2							98.04 301

无先验							98.01 303	99.23 272	95.83 316				99.85 194	95.79 294		99.44 196

原ACMM2							97.92 316

testdata2													99.89 124	95.99 285

segment_acmp												98.37 161

testdata1							97.72 323		97.86 259

test12				99.54 150	99.29 275	99.33 154		99.16 278		98.43 304		97.54 217	99.82 234		99.47 247	99.48 181

plane_prior7					99.58 174	99.38 141

plane_prior6					99.47 227	99.26 169						97.24 231

plane_prior5								99.54 185					99.82 234	95.84 292	99.78 174	99.60 124

plane_prior3						99.31 157			98.36 224	99.14 240

plane_prior2							98.80 229		98.94 166

plane_prior1					99.51 206

plane_prior						99.24 176	98.42 269		97.87 256						99.71 202

n2								0.00 362

nn								0.00 362

door-mid								99.83 40

test11								99.29 259

door								99.77 73

HQP5-MVS						98.94 213

HQP-NCC					99.31 270		97.98 308		97.45 280	98.15 314

ACMP_Plane					99.31 270		97.98 308		97.45 280	98.15 314

BP-MVS														94.73 318

HQP4-MVS										98.15 314			99.70 296			99.53 157

HQP3-MVS								99.37 243							99.67 213

HQP2-MVS												96.67 249

MDTV_nov1_ep13_2view						91.44 349	99.14 172		97.37 284	99.21 231		91.78 302		96.75 253		99.03 277

ACMMP++_ref															99.94 77

ACMMP++															99.79 169

Test By Simon												98.41 156

High-resolution multi-view benchmark