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 3

UA-Net		98.88 7	98.76 13	99.22 2	99.11 82	97.89 13	99.47 3	99.32 8	99.08 10	97.87 135	99.67 2	96.47 84	99.92 4	97.88 23	99.98 2	99.85 3

pmmvs6		99.07 4	99.24 4	98.56 49	99.81 2	96.38 61	98.87 7	99.30 9	99.01 16	99.63 9	99.66 3	99.27 2	99.68 119	97.75 30	99.89 22	99.62 25

UniMVSNet_ETH3D		99.12 3	99.28 3	98.65 43	99.77 3	96.34 63	99.18 5	99.20 13	99.67 2	99.73 3	99.65 4	99.15 3	99.86 20	97.22 45	99.92 14	99.77 8

OurMVSNet-221017-0		98.61 16	98.61 23	98.63 45	99.77 3	96.35 62	99.17 6	99.05 40	98.05 40	99.61 11	99.52 5	93.72 174	99.88 18	98.72 9	99.88 23	99.65 23

ANet_high		98.31 28	98.94 6	96.41 200	99.33 43	89.64 243	97.92 52	99.56 4	99.27 6	99.66 8	99.50 6	97.67 25	99.83 28	97.55 36	99.98 2	99.77 8

mvs_tets		98.90 5	98.94 6	98.75 33	99.69 8	96.48 59	98.54 18	99.22 10	96.23 107	99.71 4	99.48 7	98.77 6	99.93 2	98.89 3	99.95 5	99.84 5

gg-mvs-nofinetune		88.28 323	86.96 328	92.23 323	92.84 357	84.44 325	98.19 38	74.60 365	99.08 10	87.01 357	99.47 8	56.93 365	98.23 341	78.91 349	95.61 332	94.01 347

PS-MVSNAJss		98.53 19	98.63 19	98.21 75	99.68 9	94.82 122	98.10 42	99.21 11	96.91 82	99.75 2	99.45 9	95.82 105	99.92 4	98.80 4	99.96 4	99.89 1

test_djsdf		98.73 11	98.74 16	98.69 40	99.63 12	96.30 65	98.67 11	99.02 49	96.50 96	99.32 20	99.44 10	97.43 31	99.92 4	98.73 7	99.95 5	99.86 2

Anonymous20231211		98.55 17	98.76 13	97.94 93	98.79 106	94.37 140	98.84 8	99.15 21	99.37 3	99.67 6	99.43 11	95.61 117	99.72 82	98.12 16	99.86 25	99.73 15

anonymousdsp		98.72 14	98.63 19	98.99 13	99.62 13	97.29 37	98.65 14	99.19 15	95.62 141	99.35 19	99.37 12	97.38 33	99.90 13	98.59 11	99.91 17	99.77 8

jajsoiax		98.77 9	98.79 12	98.74 35	99.66 10	96.48 59	98.45 23	99.12 25	95.83 133	99.67 6	99.37 12	98.25 10	99.92 4	98.77 5	99.94 8	99.82 6

K. test v3		96.44 154	96.28 155	96.95 165	99.41 35	91.53 217	97.65 67	90.31 350	98.89 18	98.93 38	99.36 14	84.57 290	99.92 4	97.81 26	99.56 84	99.39 83

LTVRE_ROB	96.88 1	99.18 2	99.34 2	98.72 38	99.71 7	96.99 44	99.69 2	99.57 3	99.02 15	99.62 10	99.36 14	98.53 7	99.52 174	98.58 12	99.95 5	99.66 22
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
SixPastTwentyTwo		97.49 90	97.57 79	97.26 151	99.56 15	92.33 197	98.28 29	96.97 272	98.30 33	99.45 14	99.35 16	88.43 261	99.89 16	98.01 20	99.76 39	99.54 36

Gipuma		98.07 40	98.31 29	97.36 145	99.76 5	96.28 66	98.51 19	99.10 28	98.76 22	96.79 194	99.34 17	96.61 74	98.82 297	96.38 72	99.50 108	96.98 306
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
JIA-IIPM		91.79 293	90.69 302	95.11 252	93.80 347	90.98 224	94.16 255	91.78 337	96.38 100	90.30 342	99.30 18	72.02 347	98.90 290	88.28 300	90.17 352	95.45 340

TransMVSNet (Re)		98.38 25	98.67 17	97.51 124	99.51 22	93.39 177	98.20 37	98.87 83	98.23 35	99.48 12	99.27 19	98.47 8	99.55 165	96.52 67	99.53 96	99.60 26

Baseline_NR-MVSNet		97.72 74	97.79 53	97.50 127	99.56 15	93.29 178	95.44 186	98.86 85	98.20 37	98.37 73	99.24 20	94.69 144	99.55 165	95.98 90	99.79 35	99.65 23

v7n		98.73 11	98.99 5	97.95 92	99.64 11	94.20 148	98.67 11	99.14 23	99.08 10	99.42 15	99.23 21	96.53 79	99.91 12	99.27 2	99.93 10	99.73 15

pm-mvs1		98.47 21	98.67 17	97.86 99	99.52 21	94.58 132	98.28 29	99.00 57	97.57 60	99.27 24	99.22 22	98.32 9	99.50 179	97.09 53	99.75 43	99.50 43

TDRefinement		98.90 5	98.86 8	99.02 9	99.54 19	98.06 7	99.34 4	99.44 7	98.85 19	99.00 36	99.20 23	97.42 32	99.59 152	97.21 47	99.76 39	99.40 81

GBi-Net		96.99 115	96.80 128	97.56 119	97.96 201	93.67 167	98.23 32	98.66 144	95.59 143	97.99 119	99.19 24	89.51 252	99.73 78	94.60 161	99.44 126	99.30 102

test1		96.99 115	96.80 128	97.56 119	97.96 201	93.67 167	98.23 32	98.66 144	95.59 143	97.99 119	99.19 24	89.51 252	99.73 78	94.60 161	99.44 126	99.30 102

FMVSNet1		97.95 49	98.08 35	97.56 119	99.14 80	93.67 167	98.23 32	98.66 144	97.41 70	99.00 36	99.19 24	95.47 122	99.73 78	95.83 96	99.76 39	99.30 102

VDDNet		96.98 118	96.84 125	97.41 141	99.40 36	93.26 179	97.94 49	95.31 307	99.26 7	98.39 72	99.18 27	87.85 270	99.62 144	95.13 140	99.09 201	99.35 93

DSMNet-mixed		92.19 287	91.83 284	93.25 301	96.18 307	83.68 331	96.27 139	93.68 319	76.97 356	92.54 327	99.18 27	89.20 257	98.55 324	83.88 337	98.60 253	97.51 293

v10		97.55 85	97.97 41	96.31 204	98.60 131	89.64 243	97.44 82	99.02 49	96.60 90	98.72 49	99.16 29	93.48 178	99.72 82	98.76 6	99.92 14	99.58 28

MIMVSNet1		98.51 20	98.45 26	98.67 41	99.72 6	96.71 50	98.76 9	98.89 76	98.49 27	99.38 17	99.14 30	95.44 124	99.84 25	96.47 70	99.80 33	99.47 59

Vis-MVSNet		98.27 29	98.34 28	98.07 83	99.33 43	95.21 113	98.04 45	99.46 6	97.32 73	97.82 140	99.11 31	96.75 68	99.86 20	97.84 25	99.36 152	99.15 133
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
v8		97.60 82	98.06 38	96.23 207	98.71 117	89.44 247	97.43 84	98.82 109	97.29 75	98.74 47	99.10 32	93.86 169	99.68 119	98.61 10	99.94 8	99.56 33

MVS-HIRNet		88.40 322	90.20 308	82.99 344	97.01 283	60.04 366	93.11 293	85.61 361	84.45 330	88.72 350	99.09 33	84.72 289	98.23 341	82.52 342	96.59 319	90.69 358

ACMH	93.61 9	98.44 22	98.76 13	97.51 124	99.43 32	93.54 173	98.23 32	99.05 40	97.40 71	99.37 18	99.08 34	98.79 5	99.47 186	97.74 31	99.71 51	99.50 43
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
test_part1		96.77 135	96.53 144	97.47 132	98.04 191	92.92 187	97.93 50	98.85 89	98.83 20	99.30 21	99.07 35	79.25 310	99.79 38	97.59 34	99.93 10	99.69 20

DTE-MVSNet		98.79 8	98.86 8	98.59 47	99.55 17	96.12 70	98.48 22	99.10 28	99.36 4	99.29 23	99.06 36	97.27 38	99.93 2	97.71 32	99.91 17	99.70 18

Anonymous20240521		97.07 112	97.51 83	95.76 228	99.35 41	88.18 269	97.78 58	98.40 175	97.11 77	98.34 78	99.04 37	89.58 248	99.79 38	98.09 18	99.93 10	99.30 102

PEN-MVS		98.75 10	98.85 10	98.44 55	99.58 14	95.67 87	98.45 23	99.15 21	99.33 5	99.30 21	99.00 38	97.27 38	99.92 4	97.64 33	99.92 14	99.75 13

DeepC-MVS	95.41 4	97.82 67	97.70 60	98.16 76	98.78 108	95.72 82	96.23 144	99.02 49	93.92 204	98.62 51	98.99 39	97.69 23	99.62 144	96.18 78	99.87 24	99.15 133
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
VPA-MVSNet		98.27 29	98.46 24	97.70 110	99.06 87	93.80 162	97.76 61	99.00 57	98.40 29	99.07 33	98.98 40	96.89 60	99.75 65	97.19 50	99.79 35	99.55 35

lessismore_v0				97.05 161	99.36 40	92.12 205		84.07 362		98.77 46	98.98 40	85.36 284	99.74 72	97.34 43	99.37 149	99.30 102

PS-CasMVS		98.73 11	98.85 10	98.39 59	99.55 17	95.47 98	98.49 20	99.13 24	99.22 8	99.22 27	98.96 42	97.35 34	99.92 4	97.79 28	99.93 10	99.79 7

EU-MVSNet		94.25 242	94.47 228	93.60 294	98.14 184	82.60 334	97.24 94	92.72 330	85.08 322	98.48 63	98.94 43	82.59 298	98.76 304	97.47 39	99.53 96	99.44 76

LCM-MVSNet-Re		97.33 102	97.33 94	97.32 147	98.13 187	93.79 163	96.99 108	99.65 2	96.74 87	99.47 13	98.93 44	96.91 59	99.84 25	90.11 273	99.06 207	98.32 241

XXY-MVS		97.54 86	97.70 60	97.07 160	99.46 28	92.21 201	97.22 95	99.00 57	94.93 171	98.58 56	98.92 45	97.31 36	99.41 208	94.44 167	99.43 134	99.59 27

mvs_anonymous		95.36 196	96.07 165	93.21 303	96.29 300	81.56 339	94.60 238	97.66 243	93.30 219	96.95 188	98.91 46	93.03 188	99.38 218	96.60 63	97.30 305	98.69 211

DIV-MVS_2432*1600		97.86 63	98.07 36	97.25 152	99.22 57	92.81 189	97.55 74	98.94 71	97.10 78	98.85 40	98.88 47	95.03 136	99.67 124	97.39 42	99.65 61	99.26 115

UGNet		96.81 132	96.56 140	97.58 118	96.64 292	93.84 161	97.75 62	97.12 266	96.47 99	93.62 300	98.88 47	93.22 183	99.53 170	95.61 106	99.69 55	99.36 91
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
Anonymous20240529		97.96 46	98.04 39	97.71 108	98.69 121	94.28 145	97.86 55	98.31 188	98.79 21	99.23 26	98.86 49	95.76 112	99.61 150	95.49 110	99.36 152	99.23 122

FC-MVSNet-test		98.16 33	98.37 27	97.56 119	99.49 26	93.10 183	98.35 26	99.21 11	98.43 28	98.89 39	98.83 50	94.30 159	99.81 31	97.87 24	99.91 17	99.77 8

new-patchmatchnet		95.67 182	96.58 138	92.94 311	97.48 255	80.21 344	92.96 294	98.19 203	94.83 173	98.82 42	98.79 51	93.31 181	99.51 178	95.83 96	99.04 208	99.12 144

WR-MVS_H		98.65 15	98.62 21	98.75 33	99.51 22	96.61 55	98.55 17	99.17 16	99.05 13	99.17 29	98.79 51	95.47 122	99.89 16	97.95 21	99.91 17	99.75 13

ab-mvs		96.59 147	96.59 137	96.60 185	98.64 123	92.21 201	98.35 26	97.67 241	94.45 185	96.99 184	98.79 51	94.96 139	99.49 180	90.39 270	99.07 204	98.08 260

EG-PatchMatch MVS		97.69 76	97.79 53	97.40 142	99.06 87	93.52 174	95.96 160	98.97 67	94.55 184	98.82 42	98.76 54	97.31 36	99.29 242	97.20 49	99.44 126	99.38 85

nrg030		98.54 18	98.62 21	98.32 64	99.22 57	95.66 88	97.90 53	99.08 34	98.31 32	99.02 34	98.74 55	97.68 24	99.61 150	97.77 29	99.85 27	99.70 18

VDD-MVS		97.37 99	97.25 99	97.74 106	98.69 121	94.50 136	97.04 105	95.61 301	98.59 25	98.51 60	98.72 56	92.54 202	99.58 154	96.02 86	99.49 112	99.12 144

PatchT		93.75 257	93.57 253	94.29 286	95.05 332	87.32 289	96.05 152	92.98 326	97.54 63	94.25 279	98.72 56	75.79 331	99.24 250	95.92 92	95.81 327	96.32 328

RPSCF		97.87 61	97.51 83	98.95 17	99.15 72	98.43 3	97.56 73	99.06 38	96.19 108	98.48 63	98.70 58	94.72 143	99.24 250	94.37 172	99.33 167	99.17 129

APDe-MVS		98.14 34	98.03 40	98.47 54	98.72 114	96.04 73	98.07 44	99.10 28	95.96 122	98.59 55	98.69 59	96.94 55	99.81 31	96.64 62	99.58 78	99.57 32

IterMVS-LS		96.92 122	97.29 96	95.79 227	98.51 141	88.13 272	95.10 211	98.66 144	96.99 79	98.46 66	98.68 60	92.55 200	99.74 72	96.91 59	99.79 35	99.50 43
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
tfpnnormal		97.72 74	97.97 41	96.94 166	99.26 48	92.23 200	97.83 57	98.45 165	98.25 34	99.13 30	98.66 61	96.65 71	99.69 113	93.92 193	99.62 65	98.91 181

FIs		97.93 54	98.07 36	97.48 131	99.38 38	92.95 186	98.03 47	99.11 26	98.04 41	98.62 51	98.66 61	93.75 173	99.78 42	97.23 44	99.84 28	99.73 15

CP-MVSNet		98.42 23	98.46 24	98.30 67	99.46 28	95.22 111	98.27 31	98.84 94	99.05 13	99.01 35	98.65 63	95.37 125	99.90 13	97.57 35	99.91 17	99.77 8

FMVSNet2		96.72 139	96.67 135	96.87 171	97.96 201	91.88 211	97.15 97	98.06 220	95.59 143	98.50 62	98.62 64	89.51 252	99.65 131	94.99 148	99.60 74	99.07 155

PMVS	89.60 17	96.71 141	96.97 118	95.95 220	99.51 22	97.81 16	97.42 85	97.49 253	97.93 43	95.95 235	98.58 65	96.88 62	96.91 353	89.59 281	99.36 152	93.12 352
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
CR-MVSNet		93.29 270	92.79 268	94.78 268	95.44 326	88.15 270	96.18 146	97.20 261	84.94 326	94.10 283	98.57 66	77.67 318	99.39 215	95.17 133	95.81 327	96.81 317

Patchmtry		95.03 211	94.59 223	96.33 202	94.83 334	90.82 227	96.38 134	97.20 261	96.59 91	97.49 151	98.57 66	77.67 318	99.38 218	92.95 216	99.62 65	98.80 197

ambc				96.56 191	98.23 171	91.68 216	97.88 54	98.13 210		98.42 69	98.56 68	94.22 162	99.04 276	94.05 188	99.35 157	98.95 170

3Dnovator	96.53 2	97.61 81	97.64 70	97.50 127	97.74 236	93.65 171	98.49 20	98.88 81	96.86 84	97.11 173	98.55 69	95.82 105	99.73 78	95.94 91	99.42 137	99.13 139

IterMVS-SCA-FT		95.86 177	96.19 158	94.85 264	97.68 240	85.53 309	92.42 306	97.63 249	96.99 79	98.36 75	98.54 70	87.94 265	99.75 65	97.07 55	99.08 202	99.27 114

COLMAP_ROB	94.48 6	98.25 31	98.11 34	98.64 44	99.21 64	97.35 35	97.96 48	99.16 17	98.34 31	98.78 44	98.52 71	97.32 35	99.45 193	94.08 184	99.67 58	99.13 139
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
ACMH+	93.58 10	98.23 32	98.31 29	97.98 91	99.39 37	95.22 111	97.55 74	99.20 13	98.21 36	99.25 25	98.51 72	98.21 11	99.40 210	94.79 154	99.72 48	99.32 96

RPMNet		94.68 228	94.60 221	94.90 261	95.44 326	88.15 270	96.18 146	98.86 85	97.43 66	94.10 283	98.49 73	79.40 309	99.76 57	95.69 99	95.81 327	96.81 317

IterMVS		95.42 194	95.83 174	94.20 287	97.52 253	83.78 330	92.41 307	97.47 256	95.49 147	98.06 113	98.49 73	87.94 265	99.58 154	96.02 86	99.02 209	99.23 122
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
DP-MVS		97.87 61	97.89 46	97.81 102	98.62 128	94.82 122	97.13 100	98.79 111	98.98 17	98.74 47	98.49 73	95.80 111	99.49 180	95.04 144	99.44 126	99.11 148

TranMVSNet+NR-MVSNet		98.33 26	98.30 31	98.43 56	99.07 86	95.87 78	96.73 122	99.05 40	98.67 23	98.84 41	98.45 76	97.58 28	99.88 18	96.45 71	99.86 25	99.54 36

3Dnovator+	96.13 3	97.73 73	97.59 77	98.15 79	98.11 189	95.60 90	98.04 45	98.70 134	98.13 38	96.93 189	98.45 76	95.30 129	99.62 144	95.64 104	98.96 213	99.24 121

VPNet		97.26 106	97.49 86	96.59 186	99.47 27	90.58 232	96.27 139	98.53 158	97.77 46	98.46 66	98.41 78	94.59 150	99.68 119	94.61 160	99.29 175	99.52 40

test_0402		97.84 64	97.97 41	97.47 132	99.19 67	94.07 151	96.71 123	98.73 124	98.66 24	98.56 57	98.41 78	96.84 65	99.69 113	94.82 152	99.81 30	98.64 214

v1240		96.74 136	97.02 117	95.91 223	98.18 177	88.52 262	95.39 192	98.88 81	93.15 228	98.46 66	98.40 80	92.80 192	99.71 96	98.45 13	99.49 112	99.49 51

SMA-MVS		97.48 91	97.11 108	98.60 46	98.83 102	96.67 52	96.74 118	98.73 124	91.61 255	98.48 63	98.36 81	96.53 79	99.68 119	95.17 133	99.54 93	99.45 66
Yufeng Yin; Xiaoyan Liu; Zichao Zhang: SMA-MVS: Segmentation-Guided Multi-Scale Anchor Deformation Patch Multi-View Stereo. IEEE Transactions on Circuits and Systems for Video Technology
ACMMP_NAP		97.89 59	97.63 72	98.67 41	99.35 41	96.84 47	96.36 135	98.79 111	95.07 164	97.88 132	98.35 82	97.24 42	99.72 82	96.05 83	99.58 78	99.45 66

v1192		96.83 130	97.06 114	96.15 212	98.28 163	89.29 249	95.36 194	98.77 116	93.73 207	98.11 105	98.34 83	93.02 189	99.67 124	98.35 14	99.58 78	99.50 43

pmmvs-eth3d		96.49 151	96.18 159	97.42 140	98.25 168	94.29 142	94.77 233	98.07 219	89.81 276	97.97 123	98.33 84	93.11 184	99.08 272	95.46 116	99.84 28	98.89 185

PM-MVS		97.36 101	97.10 109	98.14 80	98.91 98	96.77 49	96.20 145	98.63 150	93.82 205	98.54 58	98.33 84	93.98 167	99.05 275	95.99 89	99.45 125	98.61 219

test0726					99.24 52	95.51 94	96.89 111	98.89 76	95.92 125	98.64 50	98.31 86	97.06 50

MP-MVS-pluss		97.69 76	97.36 92	98.70 39	99.50 25	96.84 47	95.38 193	98.99 60	92.45 244	98.11 105	98.31 86	97.25 41	99.77 52	96.60 63	99.62 65	99.48 56
MP-MVS-pluss: MP-MVS-pluss. MP-MVS-pluss
v1144		96.84 127	97.08 111	96.13 213	98.42 153	89.28 250	95.41 190	98.67 142	94.21 194	97.97 123	98.31 86	93.06 185	99.65 131	98.06 19	99.62 65	99.45 66

LFMVS		95.32 198	94.88 206	96.62 184	98.03 192	91.47 219	97.65 67	90.72 347	99.11 9	97.89 131	98.31 86	79.20 311	99.48 183	93.91 194	99.12 197	98.93 176

V42		97.04 113	97.16 106	96.68 183	98.59 133	91.05 222	96.33 137	98.36 180	94.60 180	97.99 119	98.30 90	93.32 180	99.62 144	97.40 41	99.53 96	99.38 85

casdiffmvs		97.50 89	97.81 52	96.56 191	98.51 141	91.04 223	95.83 168	99.09 33	97.23 76	98.33 82	98.30 90	97.03 52	99.37 221	96.58 65	99.38 148	99.28 110

v144192		96.69 142	96.90 124	96.03 215	98.25 168	88.92 254	95.49 184	98.77 116	93.05 230	98.09 109	98.29 92	92.51 204	99.70 105	98.11 17	99.56 84	99.47 59

DVP-MVS		97.78 70	97.65 67	98.16 76	99.24 52	95.51 94	96.74 118	98.23 194	95.92 125	98.40 70	98.28 93	97.06 50	99.71 96	95.48 113	99.52 101	99.26 115
Zhenlong Yuan, Jinguo Luo, Fei Shen, Zhaoxin Li, Cong Liu, Tianlu Mao, Zhaoqi Wang: DVP-MVS: Synergize Depth-Edge and Visibility Prior for Multi-View Stereo. AAAI2025
test_0728_THIRD									96.62 89	98.40 70	98.28 93	97.10 45	99.71 96	95.70 98	99.62 65	99.58 28

MVS_Test		96.27 159	96.79 130	94.73 270	96.94 287	86.63 298	96.18 146	98.33 185	94.94 169	96.07 231	98.28 93	95.25 130	99.26 247	97.21 47	97.90 278	98.30 245

FMVSNet5		93.39 267	92.35 278	96.50 193	95.83 317	90.81 229	97.31 89	98.27 189	92.74 240	96.27 222	98.28 93	62.23 361	99.67 124	90.86 250	99.36 152	99.03 161

abl_6		98.42 23	98.19 32	99.09 3	99.16 69	98.10 5	97.73 65	99.11 26	97.76 49	98.62 51	98.27 97	97.88 19	99.80 37	95.67 100	99.50 108	99.38 85

v1921920		96.72 139	96.96 120	95.99 216	98.21 172	88.79 259	95.42 188	98.79 111	93.22 222	98.19 97	98.26 98	92.68 195	99.70 105	98.34 15	99.55 90	99.49 51

SED-MVS		97.94 51	97.90 44	98.07 83	99.22 57	95.35 103	96.79 115	98.83 101	96.11 111	99.08 31	98.24 99	97.87 20	99.72 82	95.44 117	99.51 106	99.14 136

test_241102_TWO								98.83 101	96.11 111	98.62 51	98.24 99	96.92 58	99.72 82	95.44 117	99.49 112	99.49 51

v2v482		96.78 134	97.06 114	95.95 220	98.57 135	88.77 260	95.36 194	98.26 191	95.18 159	97.85 137	98.23 101	92.58 199	99.63 136	97.80 27	99.69 55	99.45 66

LPG-MVS_test		97.94 51	97.67 64	98.74 35	99.15 72	97.02 42	97.09 102	99.02 49	95.15 160	98.34 78	98.23 101	97.91 17	99.70 105	94.41 169	99.73 45	99.50 43

LGP-MVS_train				98.74 35	99.15 72	97.02 42		99.02 49	95.15 160	98.34 78	98.23 101	97.91 17	99.70 105	94.41 169	99.73 45	99.50 43

HPM-MVS_fast		98.32 27	98.13 33	98.88 24	99.54 19	97.48 30	98.35 26	99.03 47	95.88 128	97.88 132	98.22 104	98.15 12	99.74 72	96.50 69	99.62 65	99.42 78

MIMVSNet		93.42 266	92.86 265	95.10 253	98.17 179	88.19 268	98.13 41	93.69 317	92.07 247	95.04 261	98.21 105	80.95 305	99.03 279	81.42 344	98.06 272	98.07 262

hse-mvs3		96.29 158	95.63 181	98.26 69	98.50 144	96.11 71	96.90 110	97.09 267	96.58 92	97.21 166	98.19 106	84.14 291	99.78 42	95.89 94	96.17 325	98.89 185

EI-MVSNet		96.63 146	96.93 121	95.74 229	97.26 274	88.13 272	95.29 201	97.65 245	96.99 79	97.94 126	98.19 106	92.55 200	99.58 154	96.91 59	99.56 84	99.50 43

CVMVSNet		92.33 285	92.79 268	90.95 329	97.26 274	75.84 357	95.29 201	92.33 333	81.86 336	96.27 222	98.19 106	81.44 301	98.46 329	94.23 179	98.29 264	98.55 224

PVSNet_Blended_VisFu		95.95 173	95.80 175	96.42 198	99.28 47	90.62 231	95.31 199	99.08 34	88.40 290	96.97 187	98.17 109	92.11 211	99.78 42	93.64 202	99.21 182	98.86 192

EI-MVSNet-UG-set		97.32 103	97.40 89	97.09 159	97.34 269	92.01 209	95.33 197	97.65 245	97.74 50	98.30 87	98.14 110	95.04 135	99.69 113	97.55 36	99.52 101	99.58 28

test_241102_ONE					99.22 57	95.35 103		98.83 101	96.04 116	99.08 31	98.13 111	97.87 20	99.33 231

APD-MVS_3200maxsize		98.13 37	97.90 44	98.79 31	98.79 106	97.31 36	97.55 74	98.92 73	97.72 53	98.25 90	98.13 111	97.10 45	99.75 65	95.44 117	99.24 181	99.32 96

QAPM		95.88 176	95.57 184	96.80 175	97.90 207	91.84 213	98.18 39	98.73 124	88.41 289	96.42 213	98.13 111	94.73 142	99.75 65	88.72 293	98.94 217	98.81 196

ACMM	93.33 11	98.05 41	97.79 53	98.85 25	99.15 72	97.55 26	96.68 124	98.83 101	95.21 156	98.36 75	98.13 111	98.13 14	99.62 144	96.04 84	99.54 93	99.39 83
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
EI-MVSNet-Vis-set		97.32 103	97.39 90	97.11 157	97.36 264	92.08 207	95.34 196	97.65 245	97.74 50	98.29 88	98.11 115	95.05 133	99.68 119	97.50 38	99.50 108	99.56 33

wuyk23d		93.25 271	95.20 190	87.40 343	96.07 312	95.38 100	97.04 105	94.97 308	95.33 152	99.70 5	98.11 115	98.14 13	91.94 360	77.76 353	99.68 57	74.89 360

DPE-MVS		97.64 78	97.35 93	98.50 51	98.85 101	96.18 67	95.21 208	98.99 60	95.84 132	98.78 44	98.08 117	96.84 65	99.81 31	93.98 191	99.57 81	99.52 40
Kehua Chen, Zhenlong Yuan, Tianlu Mao, Zhaoqi Wang: Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization. AAAI2025
SD-MVS		97.37 99	97.70 60	96.35 201	98.14 184	95.13 114	96.54 127	98.92 73	95.94 124	99.19 28	98.08 117	97.74 22	95.06 358	95.24 129	99.54 93	98.87 191
Zhenlong Yuan, Jiakai Cao, Zhaoxin Li, Hao Jiang and Zhaoqi Wang: SD-MVS: Segmentation-driven Deformation Multi-View Stereo with Spherical Refinement and EM optimization. AAAI2024
SR-MVS-dyc-post		98.14 34	97.84 49	99.02 9	98.81 103	98.05 8	97.55 74	98.86 85	97.77 46	98.20 94	98.07 119	96.60 76	99.76 57	95.49 110	99.20 183	99.26 115

RE-MVS-def			97.88 47		98.81 103	98.05 8	97.55 74	98.86 85	97.77 46	98.20 94	98.07 119	96.94 55		95.49 110	99.20 183	99.26 115

OPM-MVS		97.54 86	97.25 99	98.41 57	99.11 82	96.61 55	95.24 206	98.46 164	94.58 183	98.10 108	98.07 119	97.09 47	99.39 215	95.16 135	99.44 126	99.21 124

AllTest		97.20 110	96.92 122	98.06 85	99.08 84	96.16 68	97.14 99	99.16 17	94.35 189	97.78 141	98.07 119	95.84 102	99.12 265	91.41 237	99.42 137	98.91 181

TestCases				98.06 85	99.08 84	96.16 68		99.16 17	94.35 189	97.78 141	98.07 119	95.84 102	99.12 265	91.41 237	99.42 137	98.91 181

TSAR-MVS + MP.		97.42 95	97.23 102	98.00 90	99.38 38	95.00 117	97.63 69	98.20 198	93.00 232	98.16 99	98.06 124	95.89 100	99.72 82	95.67 100	99.10 200	99.28 110
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
EPP-MVSNet		96.84 127	96.58 138	97.65 114	99.18 68	93.78 164	98.68 10	96.34 285	97.91 44	97.30 162	98.06 124	88.46 260	99.85 22	93.85 195	99.40 145	99.32 96

ACMMP		98.05 41	97.75 59	98.93 21	99.23 54	97.60 22	98.09 43	98.96 68	95.75 137	97.91 128	98.06 124	96.89 60	99.76 57	95.32 124	99.57 81	99.43 77
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
Anonymous202405211		96.34 157	95.98 169	97.43 139	98.25 168	93.85 160	96.74 118	94.41 314	97.72 53	98.37 73	98.03 127	87.15 274	99.53 170	94.06 185	99.07 204	98.92 180

XVG-ACMP-BASELINE		97.58 84	97.28 98	98.49 52	99.16 69	96.90 46	96.39 132	98.98 63	95.05 165	98.06 113	98.02 128	95.86 101	99.56 161	94.37 172	99.64 63	99.00 164

baseline		97.44 94	97.78 56	96.43 197	98.52 140	90.75 230	96.84 112	99.03 47	96.51 95	97.86 136	98.02 128	96.67 70	99.36 223	97.09 53	99.47 118	99.19 126

PVSNet_BlendedMVS		95.02 212	94.93 203	95.27 247	97.79 228	87.40 287	94.14 258	98.68 139	88.94 284	94.51 274	98.01 130	93.04 186	99.30 238	89.77 279	99.49 112	99.11 148

OpenMVS	94.22 8	95.48 190	95.20 190	96.32 203	97.16 279	91.96 210	97.74 63	98.84 94	87.26 299	94.36 278	98.01 130	93.95 168	99.67 124	90.70 260	98.75 238	97.35 299

MVSTER		94.21 245	93.93 248	95.05 255	95.83 317	86.46 299	95.18 209	97.65 245	92.41 245	97.94 126	98.00 132	72.39 346	99.58 154	96.36 73	99.56 84	99.12 144

IS-MVSNet		96.93 121	96.68 134	97.70 110	99.25 51	94.00 154	98.57 15	96.74 281	98.36 30	98.14 103	97.98 133	88.23 263	99.71 96	93.10 213	99.72 48	99.38 85

test1172		98.08 39	97.76 57	99.05 6	98.78 108	98.07 6	97.41 86	98.85 89	97.57 60	98.15 101	97.96 134	96.60 76	99.76 57	95.30 125	99.18 187	99.33 95

zzz-MVS		98.01 44	97.66 65	99.06 4	99.44 30	97.90 11	95.66 176	98.73 124	97.69 56	97.90 129	97.96 134	95.81 109	99.82 29	96.13 79	99.61 71	99.45 66

MTAPA		98.14 34	97.84 49	99.06 4	99.44 30	97.90 11	97.25 92	98.73 124	97.69 56	97.90 129	97.96 134	95.81 109	99.82 29	96.13 79	99.61 71	99.45 66

v148		96.58 148	96.97 118	95.42 243	98.63 127	87.57 283	95.09 213	97.90 226	95.91 127	98.24 92	97.96 134	93.42 179	99.39 215	96.04 84	99.52 101	99.29 109

MDA-MVSNet-bldmvs		95.69 180	95.67 179	95.74 229	98.48 147	88.76 261	92.84 295	97.25 259	96.00 119	97.59 144	97.95 138	91.38 225	99.46 189	93.16 212	96.35 322	98.99 167

PGM-MVS		97.88 60	97.52 82	98.96 16	99.20 65	97.62 21	97.09 102	99.06 38	95.45 148	97.55 145	97.94 139	97.11 44	99.78 42	94.77 157	99.46 121	99.48 56

LS3D		97.77 71	97.50 85	98.57 48	96.24 302	97.58 24	98.45 23	98.85 89	98.58 26	97.51 148	97.94 139	95.74 113	99.63 136	95.19 131	98.97 212	98.51 225

USDC		94.56 234	94.57 226	94.55 278	97.78 232	86.43 301	92.75 298	98.65 149	85.96 310	96.91 191	97.93 141	90.82 231	98.74 305	90.71 259	99.59 76	98.47 228

test20.03		96.58 148	96.61 136	96.48 195	98.49 145	91.72 215	95.68 175	97.69 240	96.81 85	98.27 89	97.92 142	94.18 163	98.71 308	90.78 254	99.66 60	99.00 164

FMVSNet3		95.26 201	94.94 201	96.22 209	96.53 295	90.06 236	95.99 157	97.66 243	94.11 198	97.99 119	97.91 143	80.22 308	99.63 136	94.60 161	99.44 126	98.96 169

Regformer-3		97.25 107	97.29 96	97.11 157	97.35 265	92.32 198	95.26 203	97.62 250	97.67 58	98.17 98	97.89 144	95.05 133	99.56 161	97.16 51	99.42 137	99.46 61

Regformer-4		97.53 88	97.47 88	97.71 108	97.35 265	93.91 156	95.26 203	98.14 208	97.97 42	98.34 78	97.89 144	95.49 120	99.71 96	97.41 40	99.42 137	99.51 42

xxxxxxxxxxxxxcwj		97.24 108	97.03 116	97.89 96	98.48 147	94.71 126	94.53 241	99.07 37	95.02 167	97.83 138	97.88 146	96.44 86	99.72 82	94.59 164	99.39 146	99.25 119

SF-MVS		97.60 82	97.39 90	98.22 74	98.93 96	95.69 84	97.05 104	99.10 28	95.32 153	97.83 138	97.88 146	96.44 86	99.72 82	94.59 164	99.39 146	99.25 119

SteuartSystems-ACMMP		98.02 43	97.76 57	98.79 31	99.43 32	97.21 41	97.15 97	98.90 75	96.58 92	98.08 111	97.87 148	97.02 53	99.76 57	95.25 128	99.59 76	99.40 81
Skip Steuart: Steuart Systems R&D Blog.
SR-MVS		98.00 45	97.66 65	99.01 11	98.77 110	97.93 10	97.38 87	98.83 101	97.32 73	98.06 113	97.85 149	96.65 71	99.77 52	95.00 147	99.11 198	99.32 96

RRT_MVS		94.90 214	94.07 241	97.39 143	93.18 351	93.21 181	95.26 203	97.49 253	93.94 203	98.25 90	97.85 149	72.96 345	99.84 25	97.90 22	99.78 38	99.14 136

DU-MVS		97.79 69	97.60 76	98.36 61	98.73 112	95.78 80	95.65 179	98.87 83	97.57 60	98.31 85	97.83 151	94.69 144	99.85 22	97.02 56	99.71 51	99.46 61

NR-MVSNet		97.96 46	97.86 48	98.26 69	98.73 112	95.54 92	98.14 40	98.73 124	97.79 45	99.42 15	97.83 151	94.40 157	99.78 42	95.91 93	99.76 39	99.46 61

CHOSEN 1792x2688		94.10 249	93.41 256	96.18 211	99.16 69	90.04 237	92.15 310	98.68 139	79.90 346	96.22 225	97.83 151	87.92 269	99.42 199	89.18 287	99.65 61	99.08 153

TAMVS		95.49 188	94.94 201	97.16 154	98.31 159	93.41 176	95.07 216	96.82 277	91.09 264	97.51 148	97.82 154	89.96 244	99.42 199	88.42 298	99.44 126	98.64 214

UniMVSNet (Re)		97.83 65	97.65 67	98.35 63	98.80 105	95.86 79	95.92 164	99.04 46	97.51 64	98.22 93	97.81 155	94.68 146	99.78 42	97.14 52	99.75 43	99.41 80

VNet		96.84 127	96.83 126	96.88 170	98.06 190	92.02 208	96.35 136	97.57 252	97.70 55	97.88 132	97.80 156	92.40 206	99.54 168	94.73 159	98.96 213	99.08 153

YYNet1		94.73 221	94.84 208	94.41 282	97.47 259	85.09 318	90.29 338	95.85 296	92.52 241	97.53 146	97.76 157	91.97 215	99.18 256	93.31 207	96.86 311	98.95 170

MDA-MVSNet_test_wron		94.73 221	94.83 210	94.42 281	97.48 255	85.15 316	90.28 339	95.87 295	92.52 241	97.48 154	97.76 157	91.92 219	99.17 260	93.32 206	96.80 314	98.94 172

TinyColmap		96.00 172	96.34 153	94.96 258	97.90 207	87.91 275	94.13 259	98.49 162	94.41 186	98.16 99	97.76 157	96.29 93	98.68 313	90.52 266	99.42 137	98.30 245

Patchmatch-RL test		94.66 229	94.49 227	95.19 250	98.54 138	88.91 255	92.57 302	98.74 122	91.46 258	98.32 83	97.75 160	77.31 323	98.81 299	96.06 81	99.61 71	97.85 279

MP-MVS		97.64 78	97.18 105	99.00 12	99.32 45	97.77 17	97.49 80	98.73 124	96.27 104	95.59 249	97.75 160	96.30 92	99.78 42	93.70 201	99.48 116	99.45 66
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
ACMP	92.54 13	97.47 92	97.10 109	98.55 50	99.04 90	96.70 51	96.24 143	98.89 76	93.71 208	97.97 123	97.75 160	97.44 30	99.63 136	93.22 210	99.70 54	99.32 96
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
MVP-Stereo		95.69 180	95.28 189	96.92 167	98.15 183	93.03 184	95.64 181	98.20 198	90.39 270	96.63 204	97.73 163	91.63 223	99.10 270	91.84 230	97.31 304	98.63 216
Qingsong Yan: MVP-Stereo: A Parallel Multi-View Patchmatch Stereo Method with Dilation Matching for Photogrammetric Application.
mPP-MVS		97.91 58	97.53 81	99.04 7	99.22 57	97.87 14	97.74 63	98.78 115	96.04 116	97.10 174	97.73 163	96.53 79	99.78 42	95.16 135	99.50 108	99.46 61

RRT_test8_iter05		92.46 281	92.52 277	92.29 322	95.33 329	77.43 352	95.73 170	98.55 157	94.41 186	97.46 157	97.72 165	57.44 364	99.74 72	96.92 58	99.14 190	99.69 20

MVS_0304		95.50 187	95.05 199	96.84 173	96.28 301	93.12 182	97.00 107	96.16 287	95.03 166	89.22 348	97.70 166	90.16 243	99.48 183	94.51 166	99.34 160	97.93 276

XVG-OURS		97.12 111	96.74 131	98.26 69	98.99 93	97.45 32	93.82 271	99.05 40	95.19 158	98.32 83	97.70 166	95.22 131	98.41 331	94.27 177	98.13 269	98.93 176

UniMVSNet_NR-MVSNet		97.83 65	97.65 67	98.37 60	98.72 114	95.78 80	95.66 176	99.02 49	98.11 39	98.31 85	97.69 168	94.65 148	99.85 22	97.02 56	99.71 51	99.48 56

D2MVS		95.18 203	95.17 192	95.21 249	97.76 234	87.76 281	94.15 256	97.94 224	89.77 277	96.99 184	97.68 169	87.45 272	99.14 263	95.03 146	99.81 30	98.74 205

XVS		97.96 46	97.63 72	98.94 18	99.15 72	97.66 19	97.77 59	98.83 101	97.42 67	96.32 218	97.64 170	96.49 82	99.72 82	95.66 102	99.37 149	99.45 66

ACMMPR		97.95 49	97.62 74	98.94 18	99.20 65	97.56 25	97.59 71	98.83 101	96.05 114	97.46 157	97.63 171	96.77 67	99.76 57	95.61 106	99.46 121	99.49 51

Anonymous20231206		95.27 200	95.06 198	95.88 224	98.72 114	89.37 248	95.70 172	97.85 229	88.00 295	96.98 186	97.62 172	91.95 216	99.34 228	89.21 286	99.53 96	98.94 172

region2R		97.92 55	97.59 77	98.92 22	99.22 57	97.55 26	97.60 70	98.84 94	96.00 119	97.22 164	97.62 172	96.87 63	99.76 57	95.48 113	99.43 134	99.46 61

GeoE		97.75 72	97.70 60	97.89 96	98.88 100	94.53 133	97.10 101	98.98 63	95.75 137	97.62 143	97.59 174	97.61 27	99.77 52	96.34 74	99.44 126	99.36 91

ppachtmachnet_test		94.49 237	94.84 208	93.46 297	96.16 308	82.10 336	90.59 335	97.48 255	90.53 269	97.01 183	97.59 174	91.01 228	99.36 223	93.97 192	99.18 187	98.94 172

APD-MVS		97.00 114	96.53 144	98.41 57	98.55 137	96.31 64	96.32 138	98.77 116	92.96 237	97.44 159	97.58 176	95.84 102	99.74 72	91.96 224	99.35 157	99.19 126
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
HFP-MVS		97.94 51	97.64 70	98.83 26	99.15 72	97.50 28	97.59 71	98.84 94	96.05 114	97.49 151	97.54 177	97.07 48	99.70 105	95.61 106	99.46 121	99.30 102

#test#		97.62 80	97.22 103	98.83 26	99.15 72	97.50 28	96.81 114	98.84 94	94.25 193	97.49 151	97.54 177	97.07 48	99.70 105	94.37 172	99.46 121	99.30 102

UnsupCasMVSNet_eth		95.91 174	95.73 178	96.44 196	98.48 147	91.52 218	95.31 199	98.45 165	95.76 135	97.48 154	97.54 177	89.53 251	98.69 310	94.43 168	94.61 340	99.13 139

XVG-OURS-SEG-HR		97.38 98	97.07 112	98.30 67	99.01 92	97.41 34	94.66 236	99.02 49	95.20 157	98.15 101	97.52 180	98.83 4	98.43 330	94.87 150	96.41 321	99.07 155

MG-MVS		94.08 251	94.00 244	94.32 284	97.09 281	85.89 306	93.19 292	95.96 293	92.52 241	94.93 264	97.51 181	89.54 249	98.77 302	87.52 311	97.71 286	98.31 243

Regformer-1		97.27 105	97.16 106	97.61 117	97.21 276	93.86 159	94.85 229	98.04 222	97.62 59	98.03 117	97.50 182	95.34 126	99.63 136	96.52 67	99.31 171	99.35 93

Regformer-2		97.41 96	97.24 101	97.93 94	97.21 276	94.72 125	94.85 229	98.27 189	97.74 50	98.11 105	97.50 182	95.58 118	99.69 113	96.57 66	99.31 171	99.37 90

HPM-MVS		98.11 38	97.83 51	98.92 22	99.42 34	97.46 31	98.57 15	99.05 40	95.43 150	97.41 160	97.50 182	97.98 15	99.79 38	95.58 109	99.57 81	99.50 43
Chunlin Ren, Qingshan Xu, Shikun Zhang, Jiaqi Yang: Hierarchical Prior Mining for Non-local Multi-View Stereo. ICCV 2023
9.14			96.69 133		98.53 139		96.02 155	98.98 63	93.23 221	97.18 168	97.46 185	96.47 84	99.62 144	92.99 214	99.32 169

CP-MVS		97.92 55	97.56 80	98.99 13	98.99 93	97.82 15	97.93 50	98.96 68	96.11 111	96.89 192	97.45 186	96.85 64	99.78 42	95.19 131	99.63 64	99.38 85

ZNCC-MVS		97.92 55	97.62 74	98.83 26	99.32 45	97.24 39	97.45 81	98.84 94	95.76 135	96.93 189	97.43 187	97.26 40	99.79 38	96.06 81	99.53 96	99.45 66

N_pmnet		95.18 203	94.23 235	98.06 85	97.85 210	96.55 57	92.49 304	91.63 338	89.34 279	98.09 109	97.41 188	90.33 237	99.06 274	91.58 235	99.31 171	98.56 222

GST-MVS		97.82 67	97.49 86	98.81 29	99.23 54	97.25 38	97.16 96	98.79 111	95.96 122	97.53 146	97.40 189	96.93 57	99.77 52	95.04 144	99.35 157	99.42 78

tpm		91.08 301	90.85 299	91.75 324	95.33 329	78.09 348	95.03 221	91.27 342	88.75 286	93.53 304	97.40 189	71.24 348	99.30 238	91.25 242	93.87 343	97.87 278

MDTV_nov1_ep13			91.28 291		94.31 339	73.51 361	94.80 231	93.16 324	86.75 306	93.45 308	97.40 189	76.37 327	98.55 324	88.85 291	96.43 320

DeepPCF-MVS	94.58 5	96.90 124	96.43 150	98.31 66	97.48 255	97.23 40	92.56 303	98.60 152	92.84 239	98.54 58	97.40 189	96.64 73	98.78 301	94.40 171	99.41 144	98.93 176

MSLP-MVS++		96.42 156	96.71 132	95.57 235	97.82 217	90.56 234	95.71 171	98.84 94	94.72 176	96.71 200	97.39 193	94.91 141	98.10 345	95.28 126	99.02 209	98.05 269

EPNet		93.72 258	92.62 275	97.03 163	87.61 366	92.25 199	96.27 139	91.28 341	96.74 87	87.65 354	97.39 193	85.00 286	99.64 134	92.14 223	99.48 116	99.20 125
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
PMMVS2		93.66 261	94.07 241	92.45 319	97.57 248	80.67 343	86.46 354	96.00 291	93.99 201	97.10 174	97.38 195	89.90 245	97.82 347	88.76 292	99.47 118	98.86 192

DeepC-MVS_fast	94.34 7	96.74 136	96.51 147	97.44 138	97.69 239	94.15 149	96.02 155	98.43 168	93.17 227	97.30 162	97.38 195	95.48 121	99.28 244	93.74 198	99.34 160	98.88 189
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
miper_lstm_enhance		94.81 219	94.80 211	94.85 264	96.16 308	86.45 300	91.14 329	98.20 198	93.49 212	97.03 181	97.37 197	84.97 287	99.26 247	95.28 126	99.56 84	98.83 194

OPU-MVS				97.64 115	98.01 195	95.27 106	96.79 115				97.35 198	96.97 54	98.51 327	91.21 243	99.25 180	99.14 136

cl-mvsnet1		94.73 221	94.64 217	95.01 256	95.86 315	87.00 293	91.33 323	98.08 215	93.34 217	97.10 174	97.34 199	84.02 293	99.31 235	95.15 137	99.55 90	98.72 208

cl-mvsnet____		94.73 221	94.64 217	95.01 256	95.85 316	87.00 293	91.33 323	98.08 215	93.34 217	97.10 174	97.33 200	84.01 294	99.30 238	95.14 138	99.56 84	98.71 210

WR-MVS		96.90 124	96.81 127	97.16 154	98.56 136	92.20 203	94.33 245	98.12 211	97.34 72	98.20 94	97.33 200	92.81 191	99.75 65	94.79 154	99.81 30	99.54 36

ETH3D-3000-0.1		96.89 126	96.46 149	98.16 76	98.62 128	95.69 84	95.96 160	98.98 63	93.36 216	97.04 180	97.31 202	94.93 140	99.63 136	92.60 217	99.34 160	99.17 129

ITE_SJBPF				97.85 100	98.64 123	96.66 53		98.51 161	95.63 140	97.22 164	97.30 203	95.52 119	98.55 324	90.97 247	98.90 221	98.34 240

Vis-MVSNet (Re-imp)		95.11 206	94.85 207	95.87 225	99.12 81	89.17 251	97.54 79	94.92 309	96.50 96	96.58 205	97.27 204	83.64 295	99.48 183	88.42 298	99.67 58	98.97 168

cl_fuxian		95.20 202	95.32 188	94.83 266	96.19 306	86.43 301	91.83 316	98.35 184	93.47 213	97.36 161	97.26 205	88.69 258	99.28 244	95.41 123	99.36 152	98.78 200

eth_miper_zixun_eth		94.89 215	94.93 203	94.75 269	95.99 313	86.12 304	91.35 322	98.49 162	93.40 214	97.12 172	97.25 206	86.87 277	99.35 226	95.08 143	98.82 232	98.78 200

pmmvs4		94.82 218	94.19 238	96.70 181	97.42 262	92.75 191	92.09 313	96.76 279	86.80 305	95.73 246	97.22 207	89.28 255	98.89 292	93.28 208	99.14 190	98.46 230

OMC-MVS		96.48 152	96.00 167	97.91 95	98.30 160	96.01 76	94.86 228	98.60 152	91.88 252	97.18 168	97.21 208	96.11 95	99.04 276	90.49 269	99.34 160	98.69 211

pmmvs5		94.63 231	94.34 233	95.50 239	97.63 246	88.34 266	94.02 262	97.13 265	87.15 301	95.22 256	97.15 209	87.50 271	99.27 246	93.99 190	99.26 179	98.88 189

testtj		96.69 142	96.13 160	98.36 61	98.46 151	96.02 75	96.44 130	98.70 134	94.26 192	96.79 194	97.13 210	94.07 165	99.75 65	90.53 265	98.80 233	99.31 101

our_test_3		94.20 247	94.58 224	93.07 305	96.16 308	81.20 341	90.42 337	96.84 275	90.72 267	97.14 170	97.13 210	90.47 235	99.11 268	94.04 189	98.25 265	98.91 181

CPTT-MVS		96.69 142	96.08 164	98.49 52	98.89 99	96.64 54	97.25 92	98.77 116	92.89 238	96.01 234	97.13 210	92.23 208	99.67 124	92.24 222	99.34 160	99.17 129

MS-PatchMatch		94.83 217	94.91 205	94.57 277	96.81 291	87.10 292	94.23 251	97.34 258	88.74 287	97.14 170	97.11 213	91.94 217	98.23 341	92.99 214	97.92 276	98.37 234

FPMVS		89.92 312	88.63 319	93.82 290	98.37 156	96.94 45	91.58 318	93.34 323	88.00 295	90.32 341	97.10 214	70.87 351	91.13 361	71.91 359	96.16 326	93.39 351

ETH3D cwj APD-0.16		96.23 161	95.61 183	98.09 82	97.91 205	95.65 89	94.94 224	98.74 122	91.31 261	96.02 233	97.08 215	94.05 166	99.69 113	91.51 236	98.94 217	98.93 176

ZD-MVS					98.43 152	95.94 77		98.56 156	90.72 267	96.66 202	97.07 216	95.02 137	99.74 72	91.08 244	98.93 219

DELS-MVS		96.17 164	96.23 156	95.99 216	97.55 252	90.04 237	92.38 308	98.52 159	94.13 197	96.55 209	97.06 217	94.99 138	99.58 154	95.62 105	99.28 176	98.37 234
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
CNVR-MVS		96.92 122	96.55 141	98.03 89	98.00 199	95.54 92	94.87 227	98.17 204	94.60 180	96.38 215	97.05 218	95.67 115	99.36 223	95.12 141	99.08 202	99.19 126

旧先验1					97.80 222	93.87 158		97.75 236			97.04 219	93.57 177			98.68 243	98.72 208

testdata				95.70 232	98.16 181	90.58 232		97.72 238	80.38 344	95.62 248	97.02 220	92.06 214	98.98 284	89.06 290	98.52 255	97.54 292

PatchmatchNet		91.98 291	91.87 283	92.30 321	94.60 337	79.71 345	95.12 210	93.59 321	89.52 278	93.61 301	97.02 220	77.94 316	99.18 256	90.84 251	94.57 342	98.01 273
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
SCA		93.38 268	93.52 254	92.96 310	96.24 302	81.40 340	93.24 290	94.00 316	91.58 257	94.57 271	96.97 222	87.94 265	99.42 199	89.47 283	97.66 291	98.06 266

Patchmatch-test		93.60 263	93.25 259	94.63 272	96.14 311	87.47 285	96.04 153	94.50 313	93.57 210	96.47 211	96.97 222	76.50 326	98.61 318	90.67 261	98.41 260	97.81 283

CostFormer		89.75 313	89.25 312	91.26 328	94.69 336	78.00 350	95.32 198	91.98 335	81.50 339	90.55 339	96.96 224	71.06 350	98.89 292	88.59 296	92.63 347	96.87 311

diffmvs		96.04 169	96.23 156	95.46 242	97.35 265	88.03 274	93.42 283	99.08 34	94.09 199	96.66 202	96.93 225	93.85 170	99.29 242	96.01 88	98.67 244	99.06 157

114514_t		93.96 253	93.22 260	96.19 210	99.06 87	90.97 225	95.99 157	98.94 71	73.88 359	93.43 309	96.93 225	92.38 207	99.37 221	89.09 288	99.28 176	98.25 251

Test_1112_low_res		93.53 265	92.86 265	95.54 238	98.60 131	88.86 257	92.75 298	98.69 137	82.66 335	92.65 323	96.92 227	84.75 288	99.56 161	90.94 248	97.76 282	98.19 256

tpmrst		90.31 306	90.61 304	89.41 336	94.06 345	72.37 363	95.06 218	93.69 317	88.01 294	92.32 329	96.86 228	77.45 320	98.82 297	91.04 245	87.01 357	97.04 305

PHI-MVS		96.96 119	96.53 144	98.25 72	97.48 255	96.50 58	96.76 117	98.85 89	93.52 211	96.19 227	96.85 229	95.94 99	99.42 199	93.79 197	99.43 134	98.83 194

tttt0517		93.31 269	92.56 276	95.57 235	98.71 117	87.86 276	97.44 82	87.17 358	95.79 134	97.47 156	96.84 230	64.12 359	99.81 31	96.20 77	99.32 169	99.02 163

patchmatchnet-post											96.84 230	77.36 322	99.42 199

ADS-MVSNet2		91.47 297	90.51 305	94.36 283	95.51 324	85.63 307	95.05 219	95.70 297	83.46 332	92.69 321	96.84 230	79.15 312	99.41 208	85.66 324	90.52 350	98.04 270

ADS-MVSNet		90.95 303	90.26 307	93.04 306	95.51 324	82.37 335	95.05 219	93.41 322	83.46 332	92.69 321	96.84 230	79.15 312	98.70 309	85.66 324	90.52 350	98.04 270

HY-MVS	91.43 15	92.58 279	91.81 285	94.90 261	96.49 296	88.87 256	97.31 89	94.62 311	85.92 311	90.50 340	96.84 230	85.05 285	99.40 210	83.77 339	95.78 330	96.43 327

UnsupCasMVSNet_bld		94.72 225	94.26 234	96.08 214	98.62 128	90.54 235	93.38 286	98.05 221	90.30 271	97.02 182	96.80 235	89.54 249	99.16 261	88.44 297	96.18 324	98.56 222

HQP_MVS		96.66 145	96.33 154	97.68 113	98.70 119	94.29 142	96.50 128	98.75 120	96.36 101	96.16 228	96.77 236	91.91 220	99.46 189	92.59 219	99.20 183	99.28 110

plane_prior4											96.77 236

MVS_111021_HR		96.73 138	96.54 143	97.27 149	98.35 158	93.66 170	93.42 283	98.36 180	94.74 175	96.58 205	96.76 238	96.54 78	98.99 282	94.87 150	99.27 178	99.15 133

CANet		95.86 177	95.65 180	96.49 194	96.41 298	90.82 227	94.36 244	98.41 173	94.94 169	92.62 326	96.73 239	92.68 195	99.71 96	95.12 141	99.60 74	98.94 172

1121		94.26 241	93.26 258	97.27 149	98.26 167	94.73 124	95.86 165	97.71 239	77.96 353	94.53 273	96.71 240	91.93 218	99.40 210	87.71 304	98.64 249	97.69 287

TSAR-MVS + GP.		96.47 153	96.12 161	97.49 130	97.74 236	95.23 108	94.15 256	96.90 274	93.26 220	98.04 116	96.70 241	94.41 156	98.89 292	94.77 157	99.14 190	98.37 234

test222					98.17 179	93.24 180	92.74 300	97.61 251	75.17 357	94.65 270	96.69 242	90.96 230			98.66 246	97.66 288

新几何1				97.25 152	98.29 161	94.70 129		97.73 237	77.98 352	94.83 266	96.67 243	92.08 213	99.45 193	88.17 302	98.65 248	97.61 290

miper_ehance_all_eth		94.69 226	94.70 214	94.64 271	95.77 319	86.22 303	91.32 325	98.24 193	91.67 254	97.05 179	96.65 244	88.39 262	99.22 254	94.88 149	98.34 261	98.49 227

MVS_111021_LR		96.82 131	96.55 141	97.62 116	98.27 165	95.34 105	93.81 273	98.33 185	94.59 182	96.56 207	96.63 245	96.61 74	98.73 306	94.80 153	99.34 160	98.78 200

CDPH-MVS		95.45 193	94.65 216	97.84 101	98.28 163	94.96 118	93.73 275	98.33 185	85.03 324	95.44 251	96.60 246	95.31 128	99.44 196	90.01 275	99.13 194	99.11 148

CMPMVS	73.10 23	92.74 277	91.39 289	96.77 177	93.57 350	94.67 130	94.21 253	97.67 241	80.36 345	93.61 301	96.60 246	82.85 297	97.35 351	84.86 332	98.78 235	98.29 248
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
CDS-MVSNet		94.88 216	94.12 240	97.14 156	97.64 245	93.57 172	93.96 267	97.06 269	90.05 274	96.30 221	96.55 248	86.10 279	99.47 186	90.10 274	99.31 171	98.40 231
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
LF4IMVS		96.07 167	95.63 181	97.36 145	98.19 174	95.55 91	95.44 186	98.82 109	92.29 246	95.70 247	96.55 248	92.63 198	98.69 310	91.75 233	99.33 167	97.85 279

HPM-MVS++		96.99 115	96.38 151	98.81 29	98.64 123	97.59 23	95.97 159	98.20 198	95.51 146	95.06 258	96.53 250	94.10 164	99.70 105	94.29 176	99.15 189	99.13 139

EPMVS		89.26 316	88.55 320	91.39 326	92.36 359	79.11 346	95.65 179	79.86 363	88.60 288	93.12 314	96.53 250	70.73 352	98.10 345	90.75 255	89.32 354	96.98 306

HyFIR lowres test		93.72 258	92.65 273	96.91 169	98.93 96	91.81 214	91.23 327	98.52 159	82.69 334	96.46 212	96.52 252	80.38 307	99.90 13	90.36 271	98.79 234	99.03 161

BH-RMVSNet		94.56 234	94.44 231	94.91 259	97.57 248	87.44 286	93.78 274	96.26 286	93.69 209	96.41 214	96.50 253	92.10 212	99.00 280	85.96 320	97.71 286	98.31 243

MSP-MVS		97.45 93	96.92 122	99.03 8	99.26 48	97.70 18	97.66 66	98.89 76	95.65 139	98.51 60	96.46 254	92.15 209	99.81 31	95.14 138	98.58 254	99.58 28
Zhenlong Yuan, Cong Liu, Fei Shen, Zhaoxin Li, Jingguo luo, Tianlu Mao and Zhaoqi Wang: MSP-MVS: Multi-granularity Segmentation Prior Guided Multi-View Stereo. AAAI2025
原ACMM1				96.58 188	98.16 181	92.12 205		98.15 207	85.90 312	93.49 305	96.43 255	92.47 205	99.38 218	87.66 307	98.62 250	98.23 252

tpm2		88.47 321	87.69 325	90.79 330	94.98 333	77.34 353	95.09 213	91.83 336	77.51 355	89.40 346	96.41 256	67.83 356	98.73 306	83.58 341	92.60 348	96.29 329

OpenMVS_ROB	91.80 14	93.64 262	93.05 261	95.42 243	97.31 273	91.21 221	95.08 215	96.68 283	81.56 338	96.88 193	96.41 256	90.44 236	99.25 249	85.39 327	97.67 290	95.80 334

CL-MVSNet_2432*1600		95.04 209	94.79 212	95.82 226	97.51 254	89.79 241	91.14 329	96.82 277	93.05 230	96.72 199	96.40 258	90.82 231	99.16 261	91.95 225	98.66 246	98.50 226

F-COLMAP		95.30 199	94.38 232	98.05 88	98.64 123	96.04 73	95.61 182	98.66 144	89.00 283	93.22 313	96.40 258	92.90 190	99.35 226	87.45 312	97.53 296	98.77 203

bset_n11_16_dypcd		94.53 236	93.95 247	96.25 206	97.56 250	89.85 240	88.52 351	91.32 340	94.90 172	97.51 148	96.38 260	82.34 299	99.78 42	97.22 45	99.80 33	99.12 144

NCCC		96.52 150	95.99 168	98.10 81	97.81 218	95.68 86	95.00 222	98.20 198	95.39 151	95.40 253	96.36 261	93.81 171	99.45 193	93.55 204	98.42 259	99.17 129

new_pmnet		92.34 284	91.69 287	94.32 284	96.23 304	89.16 252	92.27 309	92.88 327	84.39 331	95.29 254	96.35 262	85.66 282	96.74 356	84.53 334	97.56 294	97.05 304

cl-mvsnet2		93.25 271	92.84 267	94.46 280	94.30 340	86.00 305	91.09 331	96.64 284	90.74 266	95.79 241	96.31 263	78.24 315	98.77 302	94.15 182	98.34 261	98.62 217

tpmvs		90.79 304	90.87 298	90.57 332	92.75 358	76.30 355	95.79 169	93.64 320	91.04 265	91.91 332	96.26 264	77.19 324	98.86 296	89.38 285	89.85 353	96.56 325

test_prior3		95.91 174	95.39 187	97.46 135	97.79 228	94.26 146	93.33 288	98.42 171	94.21 194	94.02 287	96.25 265	93.64 175	99.34 228	91.90 226	98.96 213	98.79 198

test_prior2							93.33 288		94.21 194	94.02 287	96.25 265	93.64 175		91.90 226	98.96 213

testgi		96.07 167	96.50 148	94.80 267	99.26 48	87.69 282	95.96 160	98.58 155	95.08 163	98.02 118	96.25 265	97.92 16	97.60 350	88.68 295	98.74 239	99.11 148

DP-MVS Recon		95.55 186	95.13 193	96.80 175	98.51 141	93.99 155	94.60 238	98.69 137	90.20 272	95.78 243	96.21 268	92.73 194	98.98 284	90.58 264	98.86 227	97.42 296

hse-mvs2		95.77 179	95.09 195	97.79 103	97.84 214	95.51 94	95.66 176	95.43 306	96.58 92	97.21 166	96.16 269	84.14 291	99.54 168	95.89 94	96.92 308	98.32 241

MVSFormer		96.14 165	96.36 152	95.49 240	97.68 240	87.81 279	98.67 11	99.02 49	96.50 96	94.48 276	96.15 270	86.90 275	99.92 4	98.73 7	99.13 194	98.74 205

jason		94.39 240	94.04 243	95.41 245	98.29 161	87.85 278	92.74 300	96.75 280	85.38 321	95.29 254	96.15 270	88.21 264	99.65 131	94.24 178	99.34 160	98.74 205
jason: jason.
test_yl		94.40 238	94.00 244	95.59 233	96.95 285	89.52 245	94.75 234	95.55 303	96.18 109	96.79 194	96.14 272	81.09 303	99.18 256	90.75 255	97.77 280	98.07 262

DCV-MVSNet		94.40 238	94.00 244	95.59 233	96.95 285	89.52 245	94.75 234	95.55 303	96.18 109	96.79 194	96.14 272	81.09 303	99.18 256	90.75 255	97.77 280	98.07 262

dp		88.08 324	88.05 322	88.16 342	92.85 356	68.81 365	94.17 254	92.88 327	85.47 317	91.38 335	96.14 272	68.87 355	98.81 299	86.88 315	83.80 360	96.87 311

AUN-MVS		93.95 255	92.69 272	97.74 106	97.80 222	95.38 100	95.57 183	95.46 305	91.26 262	92.64 324	96.10 275	74.67 334	99.55 165	93.72 200	96.97 307	98.30 245

MCST-MVS		96.24 160	95.80 175	97.56 119	98.75 111	94.13 150	94.66 236	98.17 204	90.17 273	96.21 226	96.10 275	95.14 132	99.43 198	94.13 183	98.85 229	99.13 139

TEST9					97.84 214	95.23 108	93.62 277	98.39 176	86.81 304	93.78 292	95.99 277	94.68 146	99.52 174

train_agg		95.46 192	94.66 215	97.88 98	97.84 214	95.23 108	93.62 277	98.39 176	87.04 302	93.78 292	95.99 277	94.58 151	99.52 174	91.76 232	98.90 221	98.89 185

MSDG		95.33 197	95.13 193	95.94 222	97.40 263	91.85 212	91.02 332	98.37 179	95.30 154	96.31 220	95.99 277	94.51 154	98.38 334	89.59 281	97.65 292	97.60 291

agg_prior1		95.39 195	94.60 221	97.75 105	97.80 222	94.96 118	93.39 285	98.36 180	87.20 300	93.49 305	95.97 280	94.65 148	99.53 170	91.69 234	98.86 227	98.77 203

test_8					97.81 218	95.07 116	93.54 280	98.38 178	87.04 302	93.71 296	95.96 281	94.58 151	99.52 174

CSCG		97.40 97	97.30 95	97.69 112	98.95 95	94.83 121	97.28 91	98.99 60	96.35 103	98.13 104	95.95 282	95.99 98	99.66 130	94.36 175	99.73 45	98.59 220

TAPA-MVS	93.32 12	94.93 213	94.23 235	97.04 162	98.18 177	94.51 134	95.22 207	98.73 124	81.22 341	96.25 224	95.95 282	93.80 172	98.98 284	89.89 277	98.87 225	97.62 289
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
ETH3 D test6400		94.77 220	93.87 249	97.47 132	98.12 188	93.73 165	94.56 240	98.70 134	85.45 319	94.70 269	95.93 284	91.77 222	99.63 136	86.45 318	99.14 190	99.05 159

CS-MVS		96.95 120	97.07 112	96.59 186	97.86 209	92.74 192	97.38 87	99.52 5	95.98 121	94.89 265	95.89 285	96.05 97	99.76 57	96.65 61	99.42 137	97.26 300

baseline1		93.14 273	92.64 274	94.62 273	97.34 269	87.20 291	96.67 125	93.02 325	94.71 177	96.51 210	95.83 286	81.64 300	98.60 320	90.00 276	88.06 355	98.07 262

sss		94.22 243	93.72 251	95.74 229	97.71 238	89.95 239	93.84 270	96.98 271	88.38 291	93.75 295	95.74 287	87.94 265	98.89 292	91.02 246	98.10 270	98.37 234

CNLPA		95.04 209	94.47 228	96.75 178	97.81 218	95.25 107	94.12 260	97.89 227	94.41 186	94.57 271	95.69 288	90.30 240	98.35 337	86.72 317	98.76 237	96.64 322

PCF-MVS	89.43 18	92.12 289	90.64 303	96.57 190	97.80 222	93.48 175	89.88 345	98.45 165	74.46 358	96.04 232	95.68 289	90.71 233	99.31 235	73.73 356	99.01 211	96.91 310
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
BH-untuned		94.69 226	94.75 213	94.52 279	97.95 204	87.53 284	94.07 261	97.01 270	93.99 201	97.10 174	95.65 290	92.65 197	98.95 289	87.60 308	96.74 315	97.09 302

CANet_DTU		94.65 230	94.21 237	95.96 218	95.90 314	89.68 242	93.92 268	97.83 233	93.19 223	90.12 343	95.64 291	88.52 259	99.57 160	93.27 209	99.47 118	98.62 217

PatchMatch-RL		94.61 232	93.81 250	97.02 164	98.19 174	95.72 82	93.66 276	97.23 260	88.17 293	94.94 263	95.62 292	91.43 224	98.57 321	87.36 313	97.68 289	96.76 319

tpm cat1		88.01 325	87.33 326	90.05 335	94.48 338	76.28 356	94.47 243	94.35 315	73.84 360	89.26 347	95.61 293	73.64 339	98.30 339	84.13 335	86.20 358	95.57 339

Effi-MVS+-dtu		96.81 132	96.09 163	98.99 13	96.90 289	98.69 2	96.42 131	98.09 213	95.86 130	95.15 257	95.54 294	94.26 160	99.81 31	94.06 185	98.51 257	98.47 228

AdaColmap		95.11 206	94.62 220	96.58 188	97.33 271	94.45 137	94.92 225	98.08 215	93.15 228	93.98 290	95.53 295	94.34 158	99.10 270	85.69 323	98.61 251	96.20 330

thisisatest0530		92.71 278	91.76 286	95.56 237	98.42 153	88.23 267	96.03 154	87.35 357	94.04 200	96.56 207	95.47 296	64.03 360	99.77 52	94.78 156	99.11 198	98.68 213

WTY-MVS		93.55 264	93.00 263	95.19 250	97.81 218	87.86 276	93.89 269	96.00 291	89.02 282	94.07 285	95.44 297	86.27 278	99.33 231	87.69 306	96.82 312	98.39 233

PLC	91.02 16	94.05 252	92.90 264	97.51 124	98.00 199	95.12 115	94.25 249	98.25 192	86.17 308	91.48 334	95.25 298	91.01 228	99.19 255	85.02 331	96.69 316	98.22 253
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
pmmvs3		90.00 309	88.90 318	93.32 298	94.20 344	85.34 311	91.25 326	92.56 332	78.59 350	93.82 291	95.17 299	67.36 357	98.69 310	89.08 289	98.03 273	95.92 331

NP-MVS					98.14 184	93.72 166					95.08 300

HQP-MVS		95.17 205	94.58 224	96.92 167	97.85 210	92.47 195	94.26 246	98.43 168	93.18 224	92.86 318	95.08 300	90.33 237	99.23 252	90.51 267	98.74 239	99.05 159

cdsmvs_eth3d_5k		24.22 334	32.30 337	0.00 350	0.00 371	0.00 372	0.00 362	98.10 212	0.00 367	0.00 368	95.06 302	97.54 29	0.00 368	0.00 366	0.00 366	0.00 364

lupinMVS		93.77 256	93.28 257	95.24 248	97.68 240	87.81 279	92.12 311	96.05 289	84.52 328	94.48 276	95.06 302	86.90 275	99.63 136	93.62 203	99.13 194	98.27 249

1112_ss		94.12 248	93.42 255	96.23 207	98.59 133	90.85 226	94.24 250	98.85 89	85.49 316	92.97 316	94.94 304	86.01 280	99.64 134	91.78 231	97.92 276	98.20 255

ab-mvs-re		7.91 338	10.55 341	0.00 350	0.00 371	0.00 372	0.00 362	0.00 372	0.00 367	0.00 368	94.94 304	0.00 373	0.00 368	0.00 366	0.00 366	0.00 364

Fast-Effi-MVS+-dtu		96.44 154	96.12 161	97.39 143	97.18 278	94.39 138	95.46 185	98.73 124	96.03 118	94.72 267	94.92 306	96.28 94	99.69 113	93.81 196	97.98 274	98.09 259

EPNet_dtu		91.39 298	90.75 301	93.31 299	90.48 363	82.61 333	94.80 231	92.88 327	93.39 215	81.74 362	94.90 307	81.36 302	99.11 268	88.28 300	98.87 225	98.21 254
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
DPM-MVS		93.68 260	92.77 271	96.42 198	97.91 205	92.54 193	91.17 328	97.47 256	84.99 325	93.08 315	94.74 308	89.90 245	99.00 280	87.54 310	98.09 271	97.72 285

Effi-MVS+		96.19 163	96.01 166	96.71 180	97.43 261	92.19 204	96.12 149	99.10 28	95.45 148	93.33 312	94.71 309	97.23 43	99.56 161	93.21 211	97.54 295	98.37 234

GA-MVS		92.83 276	92.15 281	94.87 263	96.97 284	87.27 290	90.03 340	96.12 288	91.83 253	94.05 286	94.57 310	76.01 330	98.97 288	92.46 221	97.34 303	98.36 239

miper_enhance_ethall		93.14 273	92.78 270	94.20 287	93.65 348	85.29 313	89.97 341	97.85 229	85.05 323	96.15 230	94.56 311	85.74 281	99.14 263	93.74 198	98.34 261	98.17 258

xiu_mvs_v1_base_debu		95.62 183	95.96 170	94.60 274	98.01 195	88.42 263	93.99 264	98.21 195	92.98 233	95.91 236	94.53 312	96.39 88	99.72 82	95.43 120	98.19 266	95.64 336

xiu_mvs_v1_base		95.62 183	95.96 170	94.60 274	98.01 195	88.42 263	93.99 264	98.21 195	92.98 233	95.91 236	94.53 312	96.39 88	99.72 82	95.43 120	98.19 266	95.64 336

xiu_mvs_v1_base_debi		95.62 183	95.96 170	94.60 274	98.01 195	88.42 263	93.99 264	98.21 195	92.98 233	95.91 236	94.53 312	96.39 88	99.72 82	95.43 120	98.19 266	95.64 336

PVSNet_Blended		93.96 253	93.65 252	94.91 259	97.79 228	87.40 287	91.43 320	98.68 139	84.50 329	94.51 274	94.48 315	93.04 186	99.30 238	89.77 279	98.61 251	98.02 272

PAPM_NR		94.61 232	94.17 239	95.96 218	98.36 157	91.23 220	95.93 163	97.95 223	92.98 233	93.42 310	94.43 316	90.53 234	98.38 334	87.60 308	96.29 323	98.27 249

API-MVS		95.09 208	95.01 200	95.31 246	96.61 293	94.02 153	96.83 113	97.18 263	95.60 142	95.79 241	94.33 317	94.54 153	98.37 336	85.70 322	98.52 255	93.52 349

mvs-test1		96.20 162	95.50 186	98.32 64	96.90 289	98.16 4	95.07 216	98.09 213	95.86 130	93.63 299	94.32 318	94.26 160	99.71 96	94.06 185	97.27 306	97.07 303

alignmvs		96.01 171	95.52 185	97.50 127	97.77 233	94.71 126	96.07 151	96.84 275	97.48 65	96.78 198	94.28 319	85.50 283	99.40 210	96.22 76	98.73 242	98.40 231

CLD-MVS		95.47 191	95.07 196	96.69 182	98.27 165	92.53 194	91.36 321	98.67 142	91.22 263	95.78 243	94.12 320	95.65 116	98.98 284	90.81 252	99.72 48	98.57 221
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
TR-MVS		92.54 280	92.20 280	93.57 295	96.49 296	86.66 297	93.51 281	94.73 310	89.96 275	94.95 262	93.87 321	90.24 242	98.61 318	81.18 345	94.88 337	95.45 340

canonicalmvs		97.23 109	97.21 104	97.30 148	97.65 244	94.39 138	97.84 56	99.05 40	97.42 67	96.68 201	93.85 322	97.63 26	99.33 231	96.29 75	98.47 258	98.18 257

xiu_mvs_v2_base		94.22 243	94.63 219	92.99 309	97.32 272	84.84 321	92.12 311	97.84 231	91.96 250	94.17 281	93.43 323	96.07 96	99.71 96	91.27 240	97.48 298	94.42 345

CHOSEN 280x420		89.98 310	89.19 316	92.37 320	95.60 323	81.13 342	86.22 355	97.09 267	81.44 340	87.44 355	93.15 324	73.99 335	99.47 186	88.69 294	99.07 204	96.52 326

KD-MVS_2432*1600		88.93 318	87.74 323	92.49 316	88.04 364	81.99 337	89.63 347	95.62 299	91.35 259	95.06 258	93.11 325	56.58 366	98.63 316	85.19 328	95.07 335	96.85 313

miper_refine_blended		88.93 318	87.74 323	92.49 316	88.04 364	81.99 337	89.63 347	95.62 299	91.35 259	95.06 258	93.11 325	56.58 366	98.63 316	85.19 328	95.07 335	96.85 313

thres600view7		92.03 290	91.43 288	93.82 290	98.19 174	84.61 323	96.27 139	90.39 348	96.81 85	96.37 216	93.11 325	73.44 343	99.49 180	80.32 346	97.95 275	97.36 297

E-PMN		89.52 315	89.78 310	88.73 338	93.14 353	77.61 351	83.26 358	92.02 334	94.82 174	93.71 296	93.11 325	75.31 332	96.81 354	85.81 321	96.81 313	91.77 355

thres100view900		91.76 294	91.26 293	93.26 300	98.21 172	84.50 324	96.39 132	90.39 348	96.87 83	96.33 217	93.08 329	73.44 343	99.42 199	78.85 350	97.74 283	95.85 332

1314		92.38 283	92.30 279	92.64 315	95.42 328	85.15 316	95.86 165	96.97 272	85.40 320	90.62 337	93.06 330	91.12 227	97.80 348	86.74 316	95.49 334	94.97 343

PAPM		87.64 328	85.84 333	93.04 306	96.54 294	84.99 319	88.42 352	95.57 302	79.52 347	83.82 359	93.05 331	80.57 306	98.41 331	62.29 362	92.79 346	95.71 335

Fast-Effi-MVS+		95.49 188	95.07 196	96.75 178	97.67 243	92.82 188	94.22 252	98.60 152	91.61 255	93.42 310	92.90 332	96.73 69	99.70 105	92.60 217	97.89 279	97.74 284

ET-MVSNet_ETH3D		91.12 299	89.67 311	95.47 241	96.41 298	89.15 253	91.54 319	90.23 351	89.07 281	86.78 358	92.84 333	69.39 354	99.44 196	94.16 181	96.61 318	97.82 281

MVS		90.02 308	89.20 315	92.47 318	94.71 335	86.90 295	95.86 165	96.74 281	64.72 361	90.62 337	92.77 334	92.54 202	98.39 333	79.30 348	95.56 333	92.12 353

BH-w/o		92.14 288	91.94 282	92.73 314	97.13 280	85.30 312	92.46 305	95.64 298	89.33 280	94.21 280	92.74 335	89.60 247	98.24 340	81.68 343	94.66 339	94.66 344

PAPR		92.22 286	91.27 292	95.07 254	95.73 321	88.81 258	91.97 314	97.87 228	85.80 313	90.91 336	92.73 336	91.16 226	98.33 338	79.48 347	95.76 331	98.08 260

MAR-MVS		94.21 245	93.03 262	97.76 104	96.94 287	97.44 33	96.97 109	97.15 264	87.89 297	92.00 331	92.73 336	92.14 210	99.12 265	83.92 336	97.51 297	96.73 320
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
baseline2		89.65 314	88.44 321	93.25 301	95.62 322	82.71 332	93.82 271	85.94 360	88.89 285	87.35 356	92.54 338	71.23 349	99.33 231	86.01 319	94.60 341	97.72 285

PS-MVSNAJ		94.10 249	94.47 228	93.00 308	97.35 265	84.88 320	91.86 315	97.84 231	91.96 250	94.17 281	92.50 339	95.82 105	99.71 96	91.27 240	97.48 298	94.40 346

PMMVS		92.39 282	91.08 294	96.30 205	93.12 354	92.81 189	90.58 336	95.96 293	79.17 349	91.85 333	92.27 340	90.29 241	98.66 315	89.85 278	96.68 317	97.43 295

PVSNet	86.72 19	91.10 300	90.97 297	91.49 325	97.56 250	78.04 349	87.17 353	94.60 312	84.65 327	92.34 328	92.20 341	87.37 273	98.47 328	85.17 330	97.69 288	97.96 274

tfpn200view9		91.55 296	91.00 295	93.21 303	98.02 193	84.35 326	95.70 172	90.79 345	96.26 105	95.90 239	92.13 342	73.62 340	99.42 199	78.85 350	97.74 283	95.85 332

thres400		91.68 295	91.00 295	93.71 292	98.02 193	84.35 326	95.70 172	90.79 345	96.26 105	95.90 239	92.13 342	73.62 340	99.42 199	78.85 350	97.74 283	97.36 297

MVE	73.61 22	86.48 330	85.92 332	88.18 341	96.23 304	85.28 314	81.78 360	75.79 364	86.01 309	82.53 361	91.88 344	92.74 193	87.47 363	71.42 360	94.86 338	91.78 354
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
EMVS		89.06 317	89.22 313	88.61 339	93.00 355	77.34 353	82.91 359	90.92 344	94.64 179	92.63 325	91.81 345	76.30 328	97.02 352	83.83 338	96.90 310	91.48 356

thisisatest0515		90.43 305	89.18 317	94.17 289	97.07 282	85.44 310	89.75 346	87.58 356	88.28 292	93.69 298	91.72 346	65.27 358	99.58 154	90.59 263	98.67 244	97.50 294

test_method		66.88 332	66.13 335	69.11 346	62.68 367	25.73 369	49.76 361	96.04 290	14.32 364	64.27 365	91.69 347	73.45 342	88.05 362	76.06 355	66.94 362	93.54 348

EIA-MVS		96.04 169	95.77 177	96.85 172	97.80 222	92.98 185	96.12 149	99.16 17	94.65 178	93.77 294	91.69 347	95.68 114	99.67 124	94.18 180	98.85 229	97.91 277

cascas		91.89 292	91.35 290	93.51 296	94.27 341	85.60 308	88.86 350	98.61 151	79.32 348	92.16 330	91.44 349	89.22 256	98.12 344	90.80 253	97.47 300	96.82 316

IB-MVS	85.98 20	88.63 320	86.95 329	93.68 293	95.12 331	84.82 322	90.85 333	90.17 352	87.55 298	88.48 351	91.34 350	58.01 363	99.59 152	87.24 314	93.80 344	96.63 324
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
thres200		91.00 302	90.42 306	92.77 313	97.47 259	83.98 329	94.01 263	91.18 343	95.12 162	95.44 251	91.21 351	73.93 336	99.31 235	77.76 353	97.63 293	95.01 342

test0.0.03 1		90.11 307	89.21 314	92.83 312	93.89 346	86.87 296	91.74 317	88.74 355	92.02 248	94.71 268	91.14 352	73.92 337	94.48 359	83.75 340	92.94 345	97.16 301

ETV-MVS		96.13 166	95.90 173	96.82 174	97.76 234	93.89 157	95.40 191	98.95 70	95.87 129	95.58 250	91.00 353	96.36 91	99.72 82	93.36 205	98.83 231	96.85 313

test-LLR		89.97 311	89.90 309	90.16 333	94.24 342	74.98 358	89.89 342	89.06 353	92.02 248	89.97 344	90.77 354	73.92 337	98.57 321	91.88 228	97.36 301	96.92 308

test-mter		87.92 326	87.17 327	90.16 333	94.24 342	74.98 358	89.89 342	89.06 353	86.44 307	89.97 344	90.77 354	54.96 370	98.57 321	91.88 228	97.36 301	96.92 308

TESTMET0.1,1		87.20 329	86.57 331	89.07 337	93.62 349	72.84 362	89.89 342	87.01 359	85.46 318	89.12 349	90.20 356	56.00 369	97.72 349	90.91 249	96.92 308	96.64 322

gm-plane-assit					91.79 360	71.40 364			81.67 337		90.11 357		98.99 282	84.86 332

DWT-MVSNet_test		87.92 326	86.77 330	91.39 326	93.18 351	78.62 347	95.10 211	91.42 339	85.58 315	88.00 352	88.73 358	60.60 362	98.90 290	90.60 262	87.70 356	96.65 321

DeepMVS_CX				77.17 345	90.94 362	85.28 314		74.08 367	52.51 362	80.87 363	88.03 359	75.25 333	70.63 364	59.23 363	84.94 359	75.62 359

PVSNet_0	81.89 21	84.49 331	83.21 334	88.34 340	95.76 320	74.97 360	83.49 357	92.70 331	78.47 351	87.94 353	86.90 360	83.38 296	96.63 357	73.44 357	66.86 363	93.40 350

GG-mvs-BLEND				90.60 331	91.00 361	84.21 328	98.23 32	72.63 368		82.76 360	84.11 361	56.14 368	96.79 355	72.20 358	92.09 349	90.78 357

tmp_tt		57.23 333	62.50 336	41.44 347	34.77 368	49.21 368	83.93 356	60.22 369	15.31 363	71.11 364	79.37 362	70.09 353	44.86 365	64.76 361	82.93 361	30.25 361

X-MVStestdata		92.86 275	90.83 300	98.94 18	99.15 72	97.66 19	97.77 59	98.83 101	97.42 67	96.32 218	36.50 363	96.49 82	99.72 82	95.66 102	99.37 149	99.45 66

testmvs		12.33 336	15.23 339	3.64 349	5.77 370	2.23 371	88.99 349	3.62 370	2.30 366	5.29 366	13.09 364	4.52 372	1.95 366	5.16 365	8.32 365	6.75 363

test123		12.59 335	15.49 338	3.87 348	6.07 369	2.55 370	90.75 334	2.59 371	2.52 365	5.20 367	13.02 365	4.96 371	1.85 367	5.20 364	9.09 364	7.23 362

test_post											10.87 366	76.83 325	99.07 273

test_post1							94.98 223				10.37 367	76.21 329	99.04 276	89.47 283

uanet_test		0.00 339	0.00 342	0.00 350	0.00 371	0.00 372	0.00 362	0.00 372	0.00 367	0.00 368	0.00 368	0.00 373	0.00 368	0.00 366	0.00 366	0.00 364

pcd_1.5k_mvsjas		7.98 337	10.65 340	0.00 350	0.00 371	0.00 372	0.00 362	0.00 372	0.00 367	0.00 368	0.00 368	95.82 105	0.00 368	0.00 366	0.00 366	0.00 364

sosnet-low-res		0.00 339	0.00 342	0.00 350	0.00 371	0.00 372	0.00 362	0.00 372	0.00 367	0.00 368	0.00 368	0.00 373	0.00 368	0.00 366	0.00 366	0.00 364

sosnet		0.00 339	0.00 342	0.00 350	0.00 371	0.00 372	0.00 362	0.00 372	0.00 367	0.00 368	0.00 368	0.00 373	0.00 368	0.00 366	0.00 366	0.00 364

uncertanet		0.00 339	0.00 342	0.00 350	0.00 371	0.00 372	0.00 362	0.00 372	0.00 367	0.00 368	0.00 368	0.00 373	0.00 368	0.00 366	0.00 366	0.00 364

Regformer		0.00 339	0.00 342	0.00 350	0.00 371	0.00 372	0.00 362	0.00 372	0.00 367	0.00 368	0.00 368	0.00 373	0.00 368	0.00 366	0.00 366	0.00 364

uanet		0.00 339	0.00 342	0.00 350	0.00 371	0.00 372	0.00 362	0.00 372	0.00 367	0.00 368	0.00 368	0.00 373	0.00 368	0.00 366	0.00 366	0.00 364

IU-MVS					99.22 57	95.40 99		98.14 208	85.77 314	98.36 75				95.23 130	99.51 106	99.49 51

save fliter					98.48 147	94.71 126	94.53 241	98.41 173	95.02 167

test_0728_SECOND				98.25 72	99.23 54	95.49 97	96.74 118	98.89 76					99.75 65	95.48 113	99.52 101	99.53 39

GSMVS																98.06 266

test_part2					99.03 91	96.07 72				98.08 111

sam_mvs1												77.80 317				98.06 266

sam_mvs												77.38 321

MTGPA								98.73 124

MTMP							96.55 126	74.60 365

test9_res														91.29 239	98.89 224	99.00 164

agg_prior2														90.34 272	98.90 221	99.10 152

agg_prior					97.80 222	94.96 118		98.36 180		93.49 305			99.53 170

test_prior4						95.38 100	93.61 279

test_prior				97.46 135	97.79 228	94.26 146		98.42 171					99.34 228			98.79 198

旧先验2							93.35 287		77.95 354	95.77 245			98.67 314	90.74 258

新几何2							93.43 282

无先验							93.20 291	97.91 225	80.78 342				99.40 210	87.71 304		97.94 275

原ACMM2							92.82 296

testdata2													99.46 189	87.84 303

segment_acmp												95.34 126

testdata1							92.77 297		93.78 206

test12				97.46 135	97.61 247	94.07 151		97.78 235		93.57 303		93.31 181	99.42 199		98.78 235	98.89 185

plane_prior7					98.70 119	94.67 130

plane_prior6					98.38 155	94.37 140						91.91 220

plane_prior5								98.75 120					99.46 189	92.59 219	99.20 183	99.28 110

plane_prior3						94.51 134			95.29 155	96.16 228

plane_prior2							96.50 128		96.36 101

plane_prior1					98.49 145

plane_prior						94.29 142	95.42 188		94.31 191						98.93 219

n2								0.00 372

nn								0.00 372

door-mid								98.17 204

test11								98.08 215

door								97.81 234

HQP5-MVS						92.47 195

HQP-NCC					97.85 210		94.26 246		93.18 224	92.86 318

ACMP_Plane					97.85 210		94.26 246		93.18 224	92.86 318

BP-MVS														90.51 267

HQP4-MVS										92.87 317			99.23 252			99.06 157

HQP3-MVS								98.43 168							98.74 239

HQP2-MVS												90.33 237

MDTV_nov1_ep13_2view						57.28 367	94.89 226		80.59 343	94.02 287		78.66 314		85.50 326		97.82 281

ACMMP++_ref															99.52 101

ACMMP++															99.55 90

Test By Simon												94.51 154

High-resolution multi-view benchmark