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.

CHOSEN 1792x2688		97.12 100	96.80 94	98.08 114	99.30 63	94.56 199	98.05 205	99.71 1	93.57 187	97.09 122	98.91 88	88.17 186	99.89 33	96.87 92	99.56 68	99.81 6

HyFIR lowres test		96.90 108	96.49 112	98.14 108	99.33 53	95.56 152	97.38 254	99.65 2	92.34 229	97.61 109	98.20 158	89.29 159	99.10 174	96.97 79	97.60 158	99.77 18

MVS_111021_LR		98.34 41	98.23 36	98.67 70	99.27 71	96.90 91	97.95 214	99.58 3	97.14 37	98.44 61	99.01 73	95.03 62	99.62 118	97.91 36	99.75 35	99.50 79

MVS_111021_HR		98.47 32	98.34 24	98.88 62	99.22 81	97.32 73	97.91 217	99.58 3	97.20 33	98.33 66	99.00 74	95.99 30	99.64 113	98.05 32	99.76 28	99.69 43

PGM-MVS		98.49 31	98.23 36	99.27 30	99.72 11	98.08 48	98.99 56	99.49 5	95.43 99	99.03 23	99.32 25	95.56 41	99.94 3	96.80 96	99.77 22	99.78 11

ACMMP		98.23 47	97.95 47	99.09 49	99.74 7	97.62 64	99.03 50	99.41 6	95.98 77	97.60 111	99.36 21	94.45 77	99.93 13	97.14 73	98.85 109	99.70 42
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
CSCG		97.85 58	97.74 54	98.20 105	99.67 22	95.16 167	99.22 25	99.32 7	93.04 204	97.02 128	98.92 87	95.36 50	99.91 28	97.43 65	99.64 52	99.52 74

PVSNet_BlendedMVS		96.73 113	96.60 107	97.12 167	99.25 74	95.35 162	98.26 179	99.26 8	94.28 148	97.94 87	97.46 215	92.74 96	99.81 60	96.88 89	93.32 235	96.20 289

PVSNet_Blended		97.38 87	97.12 81	98.14 108	99.25 74	95.35 162	97.28 265	99.26 8	93.13 202	97.94 87	98.21 157	92.74 96	99.81 60	96.88 89	99.40 86	99.27 112

UniMVSNet_NR-MVSNet		95.71 150	95.15 159	97.40 156	96.84 243	96.97 87	98.74 104	99.24 10	95.16 114	93.88 223	97.72 196	91.68 116	98.31 261	95.81 124	87.25 299	96.92 222

WR-MVS_H		95.05 187	94.46 189	96.81 186	96.86 242	95.82 144	99.24 20	99.24 10	93.87 167	92.53 262	96.84 263	90.37 143	98.24 267	93.24 199	87.93 291	96.38 283

FC-MVSNet-test		96.42 124	96.05 124	97.53 150	96.95 235	97.27 75	99.36 8	99.23 12	95.83 82	93.93 221	98.37 139	92.00 111	98.32 259	96.02 118	92.72 242	97.00 217

VPA-MVSNet		95.75 148	95.11 162	97.69 139	97.24 216	97.27 75	98.94 66	99.23 12	95.13 115	95.51 172	97.32 224	85.73 232	98.91 196	97.33 70	89.55 272	96.89 230

FIs		96.51 121	96.12 123	97.67 141	97.13 227	97.54 67	99.36 8	99.22 14	95.89 79	94.03 219	98.35 141	91.98 112	98.44 240	96.40 108	92.76 241	97.01 216

tfpnnormal		93.66 248	92.70 256	96.55 211	96.94 236	95.94 135	98.97 60	99.19 15	91.04 266	91.38 280	97.34 222	84.94 244	98.61 224	85.45 304	89.02 280	95.11 306

UniMVSNet (Re)		95.78 147	95.19 158	97.58 147	96.99 234	97.47 69	98.79 99	99.18 16	95.60 91	93.92 222	97.04 246	91.68 116	98.48 234	95.80 126	87.66 294	96.79 240

PVSNet_Blended_VisFu		97.70 65	97.46 68	98.44 88	99.27 71	95.91 141	98.63 127	99.16 17	94.48 144	97.67 103	98.88 90	92.80 95	99.91 28	97.11 74	99.12 97	99.50 79

CHOSEN 280x420		97.18 97	97.18 80	97.20 161	98.81 110	93.27 235	95.78 309	99.15 18	95.25 110	96.79 141	98.11 164	92.29 101	99.07 177	98.56 8	99.85 3	99.25 114

D2MVS		95.18 181	95.08 163	95.48 253	97.10 229	92.07 250	98.30 173	99.13 19	94.02 157	92.90 252	96.73 266	89.48 154	98.73 216	94.48 169	93.60 229	95.65 301

PHI-MVS		98.34 41	98.06 42	99.18 39	99.15 88	98.12 47	99.04 49	99.09 20	93.32 196	98.83 38	99.10 59	96.54 12	99.83 51	97.70 50	99.76 28	99.59 69

UA-Net		97.96 51	97.62 56	98.98 55	98.86 106	97.47 69	98.89 73	99.08 21	96.67 54	98.72 47	99.54 1	93.15 92	99.81 60	94.87 154	98.83 110	99.65 58

PatchMatch-RL		96.59 118	96.03 126	98.27 99	99.31 58	96.51 108	97.91 217	99.06 22	93.72 175	96.92 133	98.06 167	88.50 181	99.65 111	91.77 238	99.00 101	98.66 167

3Dnovator	94.51 5	97.46 78	96.93 90	99.07 50	97.78 182	97.64 62	99.35 10	99.06 22	97.02 43	93.75 230	99.16 52	89.25 160	99.92 18	97.22 71	99.75 35	99.64 61

MSLP-MVS++		98.56 25	98.57 7	98.55 77	99.26 73	96.80 94	98.71 113	99.05 24	97.28 25	98.84 36	99.28 32	96.47 14	99.40 142	98.52 13	99.70 44	99.47 86

PS-CasMVS		94.67 206	93.99 211	96.71 190	96.68 252	95.26 165	99.13 38	99.03 25	93.68 181	92.33 268	97.95 175	85.35 238	98.10 272	93.59 192	88.16 290	96.79 240

TranMVSNet+NR-MVSNet		95.14 183	94.48 187	97.11 168	96.45 263	96.36 115	99.03 50	99.03 25	95.04 120	93.58 232	97.93 177	88.27 184	98.03 279	94.13 178	86.90 304	96.95 221

PEN-MVS		94.42 219	93.73 229	96.49 215	96.28 269	94.84 183	99.17 32	99.00 27	93.51 188	92.23 270	97.83 188	86.10 227	97.90 287	92.55 219	86.92 303	96.74 245

Vis-MVSNet		97.42 84	97.11 82	98.34 96	98.66 123	96.23 120	99.22 25	99.00 27	96.63 56	98.04 75	99.21 39	88.05 191	99.35 146	96.01 119	99.21 93	99.45 92
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
DU-MVS		95.42 164	94.76 176	97.40 156	96.53 258	96.97 87	98.66 125	98.99 29	95.43 99	93.88 223	97.69 197	88.57 177	98.31 261	95.81 124	87.25 299	96.92 222

VPNet		94.99 189	94.19 200	97.40 156	97.16 225	96.57 105	98.71 113	98.97 30	95.67 88	94.84 182	98.24 156	80.36 286	98.67 221	96.46 105	87.32 298	96.96 219

OpenMVS	93.04 13	95.83 145	95.00 166	98.32 97	97.18 224	97.32 73	99.21 28	98.97 30	89.96 281	91.14 282	99.05 68	86.64 218	99.92 18	93.38 195	99.47 77	97.73 198

HFP-MVS		98.63 14	98.40 16	99.32 22	99.72 11	98.29 35	99.23 21	98.96 32	96.10 75	98.94 29	99.17 47	96.06 25	99.92 18	97.62 54	99.78 19	99.75 26

#test#		98.54 28	98.27 30	99.32 22	99.72 11	98.29 35	98.98 59	98.96 32	95.65 90	98.94 29	99.17 47	96.06 25	99.92 18	97.21 72	99.78 19	99.75 26

ACMMPR		98.59 18	98.36 20	99.29 24	99.74 7	98.15 45	99.23 21	98.95 34	96.10 75	98.93 33	99.19 46	95.70 39	99.94 3	97.62 54	99.79 15	99.78 11

CP-MVSNet		94.94 195	94.30 196	96.83 185	96.72 250	95.56 152	99.11 41	98.95 34	93.89 165	92.42 267	97.90 179	87.19 209	98.12 271	94.32 173	88.21 288	96.82 239

NR-MVSNet		94.98 191	94.16 201	97.44 152	96.53 258	97.22 80	98.74 104	98.95 34	94.96 124	89.25 296	97.69 197	89.32 158	98.18 269	94.59 165	87.40 297	96.92 222

region2R		98.61 15	98.38 18	99.29 24	99.74 7	98.16 44	99.23 21	98.93 37	96.15 70	98.94 29	99.17 47	95.91 34	99.94 3	97.55 61	99.79 15	99.78 11

APDe-MVS		99.02 2	98.84 1	99.55 5	99.57 29	98.96 8	99.39 5	98.93 37	97.38 20	99.41 7	99.54 1	96.66 9	99.84 50	98.86 1	99.85 3	99.87 1

VNet		97.79 61	97.40 72	98.96 57	98.88 104	97.55 66	98.63 127	98.93 37	96.74 51	99.02 24	98.84 93	90.33 145	99.83 51	98.53 9	96.66 174	99.50 79

UGNet		96.78 112	96.30 117	98.19 107	98.24 151	95.89 143	98.88 76	98.93 37	97.39 19	96.81 139	97.84 185	82.60 271	99.90 31	96.53 103	99.49 75	98.79 156
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
sss		97.39 86	96.98 89	98.61 73	98.60 129	96.61 102	98.22 181	98.93 37	93.97 162	98.01 81	98.48 128	91.98 112	99.85 47	96.45 106	98.15 139	99.39 96

QAPM		96.29 128	95.40 143	98.96 57	97.85 179	97.60 65	99.23 21	98.93 37	89.76 283	93.11 248	99.02 69	89.11 164	99.93 13	91.99 233	99.62 55	99.34 99

DPE-MVS		98.92 3	98.67 5	99.65 2	99.58 28	99.20 4	98.42 158	98.91 43	97.58 10	99.54 4	99.46 7	97.10 5	99.94 3	97.64 53	99.84 8	99.83 5

114514_t		96.93 106	96.27 118	98.92 59	99.50 35	97.63 63	98.85 81	98.90 44	84.80 312	97.77 95	99.11 56	92.84 94	99.66 110	94.85 155	99.77 22	99.47 86

LS3D		97.16 98	96.66 106	98.68 69	98.53 133	97.19 81	98.93 67	98.90 44	92.83 215	95.99 169	99.37 17	92.12 108	99.87 42	93.67 190	99.57 63	98.97 144

DELS-MVS		98.40 36	98.20 38	98.99 53	99.00 95	97.66 61	97.75 233	98.89 46	97.71 7	98.33 66	98.97 76	94.97 63	99.88 41	98.42 18	99.76 28	99.42 95
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
DP-MVS Recon		97.86 57	97.46 68	99.06 51	99.53 31	98.35 32	98.33 166	98.89 46	92.62 219	98.05 73	98.94 84	95.34 51	99.65 111	96.04 117	99.42 83	99.19 120

AdaColmap		97.15 99	96.70 102	98.48 85	99.16 86	96.69 99	98.01 209	98.89 46	94.44 146	96.83 136	98.68 108	90.69 140	99.76 91	94.36 171	99.29 92	98.98 143

test_0728_SECOND				99.71 1	99.72 11	99.35 1	98.97 60	98.88 49					99.94 3	98.47 15	99.81 10	99.84 4

test0726					99.72 11	99.25 2	99.06 47	98.88 49	97.62 8	99.56 2	99.50 4	97.42 3

MSP-MVS		98.74 7	98.55 9	99.29 24	99.75 3	98.23 38	99.26 18	98.88 49	97.52 11	99.41 7	98.78 99	96.00 29	99.79 80	97.79 45	99.59 60	99.85 2

Anonymous20231211		94.10 239	93.26 247	96.61 200	99.11 91	94.28 207	99.01 52	98.88 49	86.43 303	92.81 254	97.57 209	81.66 276	98.68 220	94.83 156	89.02 280	96.88 231

XVS		98.70 8	98.49 14	99.34 18	99.70 19	98.35 32	99.29 14	98.88 49	97.40 17	98.46 57	99.20 43	95.90 35	99.89 33	97.85 41	99.74 38	99.78 11

X-MVStestdata		94.06 243	92.30 262	99.34 18	99.70 19	98.35 32	99.29 14	98.88 49	97.40 17	98.46 57	43.50 332	95.90 35	99.89 33	97.85 41	99.74 38	99.78 11

CP-MVS		98.57 23	98.36 20	99.19 35	99.66 23	97.86 56	99.34 11	98.87 55	95.96 78	98.60 53	99.13 54	96.05 27	99.94 3	97.77 46	99.86 1	99.77 18

SteuartSystems-ACMMP		98.90 4	98.75 3	99.36 17	99.22 81	98.43 24	99.10 42	98.87 55	97.38 20	99.35 10	99.40 11	97.78 2	99.87 42	97.77 46	99.85 3	99.78 11
Skip Steuart: Steuart Systems R&D Blog.
DeepPCF-MVS	96.37 2	97.93 55	98.48 15	96.30 230	99.00 95	89.54 287	97.43 251	98.87 55	98.16 2	99.26 12	99.38 16	96.12 23	99.64 113	98.30 23	99.77 22	99.72 36

testtj		98.33 43	97.95 47	99.47 10	99.49 39	98.70 14	98.83 85	98.86 58	95.48 96	98.91 35	99.17 47	95.48 44	99.93 13	95.80 126	99.53 72	99.76 24

DTE-MVSNet		93.98 245	93.26 247	96.14 236	96.06 278	94.39 204	99.20 29	98.86 58	93.06 203	91.78 276	97.81 190	85.87 231	97.58 299	90.53 256	86.17 308	96.46 281

SD-MVS		98.64 12	98.68 4	98.53 81	99.33 53	98.36 31	98.90 69	98.85 60	97.28 25	99.72 1	99.39 12	96.63 11	97.60 298	98.17 25	99.85 3	99.64 61
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
test_part1				0.00 323		0.00 341	0.00 334	98.84 61				0.00 343	0.00 339	0.00 336	0.00 336	0.00 335

test_prior3		98.22 48	97.90 50	99.19 35	99.31 58	98.22 39	97.80 229	98.84 61	96.12 73	97.89 92	98.69 106	95.96 31	99.70 102	96.89 86	99.60 57	99.65 58

test_prior				99.19 35	99.31 58	98.22 39		98.84 61					99.70 102			99.65 58

Anonymous20240529		95.10 185	94.22 198	97.75 133	99.01 94	94.26 209	98.87 78	98.83 64	85.79 309	96.64 144	98.97 76	78.73 293	99.85 47	96.27 110	94.89 205	99.12 130

9.14			98.06 42		99.47 40		98.71 113	98.82 65	94.36 147	99.16 18	99.29 31	96.05 27	99.81 60	97.00 77	99.71 43

SR-MVS		98.57 23	98.35 22	99.24 32	99.53 31	98.18 42	99.09 43	98.82 65	96.58 57	99.10 20	99.32 25	95.39 47	99.82 57	97.70 50	99.63 53	99.72 36

GST-MVS		98.43 34	98.12 40	99.34 18	99.72 11	98.38 25	99.09 43	98.82 65	95.71 86	98.73 46	99.06 67	95.27 53	99.93 13	97.07 76	99.63 53	99.72 36

abl_6		98.30 46	98.03 44	99.13 45	99.56 30	97.76 60	99.13 38	98.82 65	96.14 71	99.26 12	99.37 17	93.33 89	99.93 13	96.96 81	99.67 46	99.69 43

HPM-MVS_fast		98.38 37	98.13 39	99.12 47	99.75 3	97.86 56	99.44 4	98.82 65	94.46 145	98.94 29	99.20 43	95.16 58	99.74 95	97.58 58	99.85 3	99.77 18

APD-MVS		98.35 40	98.00 46	99.42 14	99.51 33	98.72 13	98.80 95	98.82 65	94.52 142	99.23 14	99.25 35	95.54 43	99.80 68	96.52 104	99.77 22	99.74 31
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
MVS_0304		92.81 264	92.01 265	95.23 259	97.46 201	91.33 262	98.17 192	98.81 71	91.13 265	93.80 228	95.68 301	66.08 324	98.06 276	90.79 252	96.13 196	96.32 286

ACMMP_NAP		98.61 15	98.30 28	99.55 5	99.62 26	98.95 9	98.82 88	98.81 71	95.80 83	99.16 18	99.47 6	95.37 49	99.92 18	97.89 39	99.75 35	99.79 8

Regformer-2		98.69 9	98.52 11	99.19 35	99.35 48	98.01 51	98.37 162	98.81 71	97.48 14	99.21 15	99.21 39	96.13 22	99.80 68	98.40 20	99.73 40	99.75 26

APD-MVS_3200maxsize		98.53 29	98.33 27	99.15 44	99.50 35	97.92 55	99.15 34	98.81 71	96.24 66	99.20 16	99.37 17	95.30 52	99.80 68	97.73 48	99.67 46	99.72 36

WR-MVS		95.15 182	94.46 189	97.22 160	96.67 253	96.45 110	98.21 182	98.81 71	94.15 151	93.16 244	97.69 197	87.51 203	98.30 263	95.29 145	88.62 284	96.90 229

mPP-MVS		98.51 30	98.26 31	99.25 31	99.75 3	98.04 49	99.28 16	98.81 71	96.24 66	98.35 65	99.23 36	95.46 45	99.94 3	97.42 66	99.81 10	99.77 18

CNVR-MVS		98.78 5	98.56 8	99.45 13	99.32 56	98.87 11	98.47 150	98.81 71	97.72 5	98.76 43	99.16 52	97.05 6	99.78 84	98.06 30	99.66 49	99.69 43

CPTT-MVS		97.72 64	97.32 75	98.92 59	99.64 24	97.10 84	99.12 40	98.81 71	92.34 229	98.09 71	99.08 65	93.01 93	99.92 18	96.06 116	99.77 22	99.75 26

SMA-MVS		98.58 20	98.25 32	99.56 4	99.51 33	99.04 7	98.95 64	98.80 79	93.67 183	99.37 9	99.52 3	96.52 13	99.89 33	98.06 30	99.81 10	99.76 24

HPM-MVS		98.36 39	98.10 41	99.13 45	99.74 7	97.82 58	99.53 1	98.80 79	94.63 138	98.61 52	98.97 76	95.13 59	99.77 89	97.65 52	99.83 9	99.79 8
Chunlin Ren, Qingshan Xu, Shikun Zhang, Jiaqi Yang: Hierarchical Prior Mining for Non-local Multi-View Stereo. ICCV 2023
Regformer-4		98.64 12	98.53 10	98.99 53	99.43 46	97.37 72	98.40 160	98.79 81	97.46 15	99.09 21	99.31 27	95.86 37	99.80 68	98.64 3	99.76 28	99.79 8

MP-MVS		98.33 43	98.01 45	99.28 27	99.75 3	98.18 42	99.22 25	98.79 81	96.13 72	97.92 90	99.23 36	94.54 72	99.94 3	96.74 98	99.78 19	99.73 33
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
CANet		98.05 49	97.76 53	98.90 61	98.73 114	97.27 75	98.35 164	98.78 83	97.37 22	97.72 100	98.96 81	91.53 123	99.92 18	98.79 2	99.65 50	99.51 77

MP-MVS-pluss		98.31 45	97.92 49	99.49 8	99.72 11	98.88 10	98.43 156	98.78 83	94.10 153	97.69 102	99.42 10	95.25 55	99.92 18	98.09 29	99.80 14	99.67 53
MP-MVS-pluss: MP-MVS-pluss. MP-MVS-pluss
DeepC-MVS_fast	96.70 1	98.55 26	98.34 24	99.18 39	99.25 74	98.04 49	98.50 147	98.78 83	97.72 5	98.92 34	99.28 32	95.27 53	99.82 57	97.55 61	99.77 22	99.69 43
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
MG-MVS		97.81 59	97.60 58	98.44 88	99.12 90	95.97 132	97.75 233	98.78 83	96.89 46	98.46 57	99.22 38	93.90 86	99.68 108	94.81 158	99.52 74	99.67 53

Regformer-1		98.66 10	98.51 12	99.12 47	99.35 48	97.81 59	98.37 162	98.76 87	97.49 13	99.20 16	99.21 39	96.08 24	99.79 80	98.42 18	99.73 40	99.75 26

NCCC		98.61 15	98.35 22	99.38 15	99.28 70	98.61 17	98.45 151	98.76 87	97.82 4	98.45 60	98.93 85	96.65 10	99.83 51	97.38 68	99.41 84	99.71 40

PLC	95.07 4	97.20 96	96.78 97	98.44 88	99.29 66	96.31 119	98.14 194	98.76 87	92.41 227	96.39 159	98.31 148	94.92 65	99.78 84	94.06 180	98.77 113	99.23 115
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
DeepC-MVS	95.98 3	97.88 56	97.58 59	98.77 65	99.25 74	96.93 89	98.83 85	98.75 90	96.96 45	96.89 135	99.50 4	90.46 142	99.87 42	97.84 43	99.76 28	99.52 74
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
zzz-MVS		98.55 26	98.25 32	99.46 11	99.76 1	98.64 15	98.55 140	98.74 91	97.27 29	98.02 77	99.39 12	94.81 66	99.96 1	97.91 36	99.79 15	99.77 18

MTGPA								98.74 91

MTAPA		98.58 20	98.29 29	99.46 11	99.76 1	98.64 15	98.90 69	98.74 91	97.27 29	98.02 77	99.39 12	94.81 66	99.96 1	97.91 36	99.79 15	99.77 18

ab-mvs		96.42 124	95.71 136	98.55 77	98.63 126	96.75 97	97.88 222	98.74 91	93.84 168	96.54 152	98.18 160	85.34 239	99.75 93	95.93 120	96.35 184	99.15 126

TEST9					99.31 58	98.50 20	97.92 215	98.73 95	92.63 218	97.74 98	98.68 108	96.20 18	99.80 68

train_agg		97.97 50	97.52 64	99.33 21	99.31 58	98.50 20	97.92 215	98.73 95	92.98 207	97.74 98	98.68 108	96.20 18	99.80 68	96.59 101	99.57 63	99.68 49

test_8					99.29 66	98.44 22	97.89 221	98.72 97	92.98 207	97.70 101	98.66 111	96.20 18	99.80 68

agg_prior1		97.95 53	97.51 65	99.28 27	99.30 63	98.38 25	97.81 228	98.72 97	93.16 201	97.57 112	98.66 111	96.14 21	99.81 60	96.63 100	99.56 68	99.66 56

agg_prior					99.30 63	98.38 25		98.72 97		97.57 112			99.81 60

无先验							97.58 245	98.72 97	91.38 251				99.87 42	93.36 196		99.60 67

save fliter					99.46 42	98.38 25	98.21 182	98.71 101	97.95 3

WTY-MVS		97.37 88	96.92 91	98.72 67	98.86 106	96.89 93	98.31 171	98.71 101	95.26 109	97.67 103	98.56 122	92.21 105	99.78 84	95.89 121	96.85 169	99.48 84

3Dnovator+	94.38 6	97.43 83	96.78 97	99.38 15	97.83 180	98.52 19	99.37 7	98.71 101	97.09 41	92.99 251	99.13 54	89.36 157	99.89 33	96.97 79	99.57 63	99.71 40

旧先验1					99.29 66	97.48 68		98.70 104			99.09 63	95.56 41			99.47 77	99.61 64

EI-MVSNet-Vis-set		98.47 32	98.39 17	98.69 68	99.46 42	96.49 109	98.30 173	98.69 105	97.21 32	98.84 36	99.36 21	95.41 46	99.78 84	98.62 5	99.65 50	99.80 7

新几何1				99.16 42	99.34 50	98.01 51		98.69 105	90.06 280	98.13 69	98.95 83	94.60 70	99.89 33	91.97 234	99.47 77	99.59 69

API-MVS		97.41 85	97.25 77	97.91 123	98.70 119	96.80 94	98.82 88	98.69 105	94.53 140	98.11 70	98.28 150	94.50 76	99.57 122	94.12 179	99.49 75	97.37 208

EI-MVSNet-UG-set		98.41 35	98.34 24	98.61 73	99.45 44	96.32 117	98.28 176	98.68 108	97.17 35	98.74 44	99.37 17	95.25 55	99.79 80	98.57 7	99.54 71	99.73 33

Regformer-3		98.59 18	98.50 13	98.86 63	99.43 46	97.05 85	98.40 160	98.68 108	97.43 16	99.06 22	99.31 27	95.80 38	99.77 89	98.62 5	99.76 28	99.78 11

testdata				98.26 101	99.20 84	95.36 160		98.68 108	91.89 240	98.60 53	99.10 59	94.44 78	99.82 57	94.27 175	99.44 82	99.58 71

1121		97.37 88	96.77 101	99.16 42	99.34 50	97.99 54	98.19 187	98.68 108	90.14 279	98.01 81	98.97 76	94.80 68	99.87 42	93.36 196	99.46 80	99.61 64

MCST-MVS		98.65 11	98.37 19	99.48 9	99.60 27	98.87 11	98.41 159	98.68 108	97.04 42	98.52 56	98.80 97	96.78 8	99.83 51	97.93 35	99.61 56	99.74 31

PVSNet	91.96 18	96.35 126	96.15 122	96.96 177	99.17 85	92.05 251	96.08 302	98.68 108	93.69 179	97.75 97	97.80 191	88.86 172	99.69 107	94.26 176	99.01 100	99.15 126

MAR-MVS		96.91 107	96.40 114	98.45 87	98.69 121	96.90 91	98.66 125	98.68 108	92.40 228	97.07 125	97.96 174	91.54 122	99.75 93	93.68 189	98.92 103	98.69 163
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
原ACMM1				98.65 71	99.32 56	96.62 100		98.67 115	93.27 199	97.81 94	98.97 76	95.18 57	99.83 51	93.84 185	99.46 80	99.50 79

CDPH-MVS		97.94 54	97.49 66	99.28 27	99.47 40	98.44 22	97.91 217	98.67 115	92.57 222	98.77 42	98.85 92	95.93 33	99.72 96	95.56 136	99.69 45	99.68 49

UnsupCasMVSNet_eth		90.99 279	89.92 281	94.19 288	94.08 312	89.83 282	97.13 275	98.67 115	93.69 179	85.83 311	96.19 287	75.15 310	96.74 310	89.14 280	79.41 320	96.00 294

TSAR-MVS + MP.		98.78 5	98.62 6	99.24 32	99.69 21	98.28 37	99.14 35	98.66 118	96.84 47	99.56 2	99.31 27	96.34 15	99.70 102	98.32 22	99.73 40	99.73 33
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
HPM-MVS++		98.58 20	98.25 32	99.55 5	99.50 35	99.08 6	98.72 112	98.66 118	97.51 12	98.15 68	98.83 94	95.70 39	99.92 18	97.53 63	99.67 46	99.66 56

test222					99.23 80	97.17 82	97.40 252	98.66 118	88.68 294	98.05 73	98.96 81	94.14 82			99.53 72	99.61 64

test11								98.66 118

XXY-MVS		95.20 180	94.45 191	97.46 151	96.75 248	96.56 106	98.86 80	98.65 122	93.30 198	93.27 241	98.27 153	84.85 246	98.87 203	94.82 157	91.26 258	96.96 219

TAPA-MVS	93.98 7	95.35 171	94.56 183	97.74 134	99.13 89	94.83 185	98.33 166	98.64 123	86.62 301	96.29 161	98.61 114	94.00 85	99.29 150	80.00 316	99.41 84	99.09 132
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
F-COLMAP		97.09 102	96.80 94	97.97 120	99.45 44	94.95 180	98.55 140	98.62 124	93.02 205	96.17 164	98.58 119	94.01 84	99.81 60	93.95 182	98.90 104	99.14 128

ETV-MVS		97.75 62	97.58 59	98.27 99	98.38 139	96.44 111	99.01 52	98.60 125	95.88 80	97.26 117	97.53 212	94.97 63	99.33 148	97.38 68	99.20 94	99.05 137

PAPM_NR		97.46 78	97.11 82	98.50 83	99.50 35	96.41 113	98.63 127	98.60 125	95.18 113	97.06 126	98.06 167	94.26 81	99.57 122	93.80 187	98.87 108	99.52 74

cdsmvs_eth3d_5k		23.98 308	31.98 309	0.00 323	0.00 340	0.00 341	0.00 334	98.59 127	0.00 336	0.00 338	98.61 114	90.60 141	0.00 339	0.00 336	0.00 336	0.00 335

1314		96.25 132	95.73 132	97.79 129	97.13 227	95.55 154	98.19 187	98.59 127	93.47 190	92.03 274	97.82 189	91.33 126	99.49 133	94.62 162	98.44 128	98.32 184

CVMVSNet		95.43 163	96.04 125	93.57 292	97.93 174	83.62 317	98.12 197	98.59 127	95.68 87	96.56 148	99.02 69	87.51 203	97.51 302	93.56 193	97.44 160	99.60 67

OMC-MVS		97.55 76	97.34 74	98.20 105	99.33 53	95.92 139	98.28 176	98.59 127	95.52 95	97.97 84	99.10 59	93.28 91	99.49 133	95.09 151	98.88 106	99.19 120

LTVRE_ROB	92.95 15	94.60 209	93.90 216	96.68 194	97.41 209	94.42 202	98.52 142	98.59 127	91.69 244	91.21 281	98.35 141	84.87 245	99.04 181	91.06 248	93.44 233	96.60 263
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
DVP-MVS		99.03 1	98.83 2	99.63 3	99.72 11	99.25 2	98.97 60	98.58 132	97.62 8	99.45 5	99.46 7	97.42 3	99.94 3	98.47 15	99.81 10	99.69 43

UniMVSNet_ETH3D		94.24 229	93.33 244	96.97 176	97.19 223	93.38 234	98.74 104	98.57 133	91.21 263	93.81 227	98.58 119	72.85 318	98.77 214	95.05 152	93.93 222	98.77 158

PAPR		96.84 110	96.24 120	98.65 71	98.72 118	96.92 90	97.36 258	98.57 133	93.33 195	96.67 143	97.57 209	94.30 80	99.56 124	91.05 250	98.59 120	99.47 86

HQP_MVS		96.14 133	95.90 129	96.85 184	97.42 206	94.60 197	98.80 95	98.56 135	97.28 25	95.34 173	98.28 150	87.09 210	99.03 182	96.07 114	94.27 208	96.92 222

plane_prior5								98.56 135					99.03 182	96.07 114	94.27 208	96.92 222

EIA-MVS		97.96 51	97.81 51	98.40 93	98.42 137	97.27 75	98.73 108	98.55 137	96.84 47	98.38 63	97.44 218	95.39 47	99.35 146	97.62 54	98.89 105	98.58 173

CS-MVS		97.81 59	97.61 57	98.41 92	98.52 134	97.15 83	99.09 43	98.55 137	96.18 69	97.61 109	97.20 232	94.59 71	99.39 143	97.62 54	99.10 98	98.70 161

mvs_tets		95.41 166	95.00 166	96.65 195	95.58 292	94.42 202	99.00 54	98.55 137	95.73 85	93.21 243	98.38 138	83.45 269	98.63 223	97.09 75	94.00 219	96.91 227

LPG-MVS_test		95.62 155	95.34 149	96.47 217	97.46 201	93.54 228	98.99 56	98.54 140	94.67 135	94.36 201	98.77 101	85.39 236	99.11 171	95.71 131	94.15 214	96.76 243

LGP-MVS_train				96.47 217	97.46 201	93.54 228		98.54 140	94.67 135	94.36 201	98.77 101	85.39 236	99.11 171	95.71 131	94.15 214	96.76 243

test12				99.18 39	99.16 86	98.19 41		98.53 142		98.07 72		95.13 59	99.72 96		99.56 68	99.63 63

CNLPA		97.45 81	97.03 86	98.73 66	99.05 92	97.44 71	98.07 203	98.53 142	95.32 107	96.80 140	98.53 123	93.32 90	99.72 96	94.31 174	99.31 91	99.02 139

jajsoiax		95.45 162	95.03 165	96.73 189	95.42 299	94.63 192	99.14 35	98.52 144	95.74 84	93.22 242	98.36 140	83.87 265	98.65 222	96.95 82	94.04 217	96.91 227

XVG-OURS		96.55 120	96.41 113	96.99 173	98.75 113	93.76 221	97.50 248	98.52 144	95.67 88	96.83 136	99.30 30	88.95 171	99.53 130	95.88 122	96.26 191	97.69 200

xiu_mvs_v1_base_debu		97.60 70	97.56 61	97.72 135	98.35 141	95.98 127	97.86 224	98.51 146	97.13 38	99.01 25	98.40 135	91.56 119	99.80 68	98.53 9	98.68 114	97.37 208

xiu_mvs_v1_base		97.60 70	97.56 61	97.72 135	98.35 141	95.98 127	97.86 224	98.51 146	97.13 38	99.01 25	98.40 135	91.56 119	99.80 68	98.53 9	98.68 114	97.37 208

xiu_mvs_v1_base_debi		97.60 70	97.56 61	97.72 135	98.35 141	95.98 127	97.86 224	98.51 146	97.13 38	99.01 25	98.40 135	91.56 119	99.80 68	98.53 9	98.68 114	97.37 208

PS-MVSNAJ		97.73 63	97.77 52	97.62 145	98.68 122	95.58 150	97.34 260	98.51 146	97.29 24	98.66 49	97.88 181	94.51 73	99.90 31	97.87 40	99.17 96	97.39 206

cascas		94.63 208	93.86 219	96.93 180	96.91 239	94.27 208	96.00 306	98.51 146	85.55 310	94.54 190	96.23 284	84.20 258	98.87 203	95.80 126	96.98 168	97.66 201

PS-MVSNAJss		96.43 123	96.26 119	96.92 182	95.84 286	95.08 172	99.16 33	98.50 151	95.87 81	93.84 226	98.34 145	94.51 73	98.61 224	96.88 89	93.45 232	97.06 214

MVS		94.67 206	93.54 238	98.08 114	96.88 241	96.56 106	98.19 187	98.50 151	78.05 321	92.69 257	98.02 169	91.07 133	99.63 116	90.09 261	98.36 133	98.04 189

XVG-OURS-SEG-HR		96.51 121	96.34 115	97.02 172	98.77 112	93.76 221	97.79 231	98.50 151	95.45 98	96.94 130	99.09 63	87.87 196	99.55 129	96.76 97	95.83 200	97.74 197

PVSNet_0	88.72 19	91.28 276	90.03 280	95.00 267	97.99 171	87.29 311	94.84 316	98.50 151	92.06 236	89.86 291	95.19 303	79.81 288	99.39 143	92.27 225	69.79 326	98.33 183

ACMH	92.88 16	94.55 213	93.95 213	96.34 228	97.63 189	93.26 236	98.81 94	98.49 155	93.43 192	89.74 292	98.53 123	81.91 274	99.08 176	93.69 188	93.30 236	96.70 252
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
xiu_mvs_v2_base		97.66 67	97.70 55	97.56 149	98.61 128	95.46 157	97.44 249	98.46 156	97.15 36	98.65 50	98.15 161	94.33 79	99.80 68	97.84 43	98.66 118	97.41 204

HQP3-MVS								98.46 156							94.18 212

HQP-MVS		95.72 149	95.40 143	96.69 193	97.20 220	94.25 210	98.05 205	98.46 156	96.43 60	94.45 194	97.73 194	86.75 216	98.96 189	95.30 143	94.18 212	96.86 235

CLD-MVS		95.62 155	95.34 149	96.46 220	97.52 198	93.75 223	97.27 266	98.46 156	95.53 94	94.42 199	98.00 172	86.21 225	98.97 186	96.25 112	94.37 206	96.66 258
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
XVG-ACMP-BASELINE		94.54 214	94.14 203	95.75 248	96.55 257	91.65 259	98.11 199	98.44 160	94.96 124	94.22 209	97.90 179	79.18 292	99.11 171	94.05 181	93.85 223	96.48 279

ACMP	93.49 10	95.34 172	94.98 168	96.43 222	97.67 187	93.48 230	98.73 108	98.44 160	94.94 127	92.53 262	98.53 123	84.50 252	99.14 165	95.48 139	94.00 219	96.66 258
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
ACMM	93.85 9	95.69 152	95.38 147	96.61 200	97.61 190	93.84 219	98.91 68	98.44 160	95.25 110	94.28 205	98.47 129	86.04 230	99.12 167	95.50 138	93.95 221	96.87 233
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
Effi-MVS+		97.12 100	96.69 103	98.39 94	98.19 157	96.72 98	97.37 256	98.43 163	93.71 176	97.65 106	98.02 169	92.20 106	99.25 152	96.87 92	97.79 150	99.19 120

anonymousdsp		95.42 164	94.91 171	96.94 179	95.10 302	95.90 142	99.14 35	98.41 164	93.75 171	93.16 244	97.46 215	87.50 205	98.41 250	95.63 135	94.03 218	96.50 277

PMMVS		96.60 116	96.33 116	97.41 154	97.90 176	93.93 216	97.35 259	98.41 164	92.84 214	97.76 96	97.45 217	91.10 132	99.20 158	96.26 111	97.91 145	99.11 131

MVSFormer		97.57 74	97.49 66	97.84 126	98.07 165	95.76 146	99.47 2	98.40 166	94.98 122	98.79 40	98.83 94	92.34 99	98.41 250	96.91 83	99.59 60	99.34 99

test_djsdf		96.00 137	95.69 138	96.93 180	95.72 288	95.49 156	99.47 2	98.40 166	94.98 122	94.58 189	97.86 182	89.16 163	98.41 250	96.91 83	94.12 216	96.88 231

OPM-MVS		95.69 152	95.33 151	96.76 188	96.16 275	94.63 192	98.43 156	98.39 168	96.64 55	95.02 179	98.78 99	85.15 241	99.05 178	95.21 150	94.20 211	96.60 263

canonicalmvs		97.67 66	97.23 78	98.98 55	98.70 119	98.38 25	99.34 11	98.39 168	96.76 50	97.67 103	97.40 221	92.26 102	99.49 133	98.28 24	96.28 190	99.08 135

DP-MVS		96.59 118	95.93 128	98.57 75	99.34 50	96.19 123	98.70 117	98.39 168	89.45 288	94.52 191	99.35 23	91.85 114	99.85 47	92.89 212	98.88 106	99.68 49

diffmvs		97.58 73	97.40 72	98.13 110	98.32 148	95.81 145	98.06 204	98.37 171	96.20 68	98.74 44	98.89 89	91.31 127	99.25 152	98.16 26	98.52 123	99.34 99

ACMH+	92.99 14	94.30 225	93.77 225	95.88 243	97.81 181	92.04 252	98.71 113	98.37 171	93.99 160	90.60 288	98.47 129	80.86 282	99.05 178	92.75 214	92.40 244	96.55 271

MSDG		95.93 140	95.30 154	97.83 127	98.90 102	95.36 160	96.83 292	98.37 171	91.32 256	94.43 198	98.73 105	90.27 146	99.60 119	90.05 264	98.82 111	98.52 174

DPM-MVS		97.55 76	96.99 88	99.23 34	99.04 93	98.55 18	97.17 273	98.35 174	94.85 129	97.93 89	98.58 119	95.07 61	99.71 101	92.60 215	99.34 89	99.43 94

CMPMVS	66.06 21	89.70 287	89.67 283	89.78 306	93.19 315	76.56 325	97.00 277	98.35 174	80.97 318	81.57 318	97.75 193	74.75 312	98.61 224	89.85 267	93.63 227	94.17 314
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
v7n		94.19 232	93.43 242	96.47 217	95.90 283	94.38 205	99.26 18	98.34 176	91.99 237	92.76 255	97.13 235	88.31 183	98.52 232	89.48 276	87.70 293	96.52 274

CDS-MVSNet		96.99 104	96.69 103	97.90 124	98.05 168	95.98 127	98.20 184	98.33 177	93.67 183	96.95 129	98.49 127	93.54 87	98.42 243	95.24 149	97.74 153	99.31 105
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
casdiffmvs		97.63 69	97.41 71	98.28 98	98.33 146	96.14 124	98.82 88	98.32 178	96.38 63	97.95 85	99.21 39	91.23 129	99.23 155	98.12 27	98.37 131	99.48 84

baseline		97.64 68	97.44 70	98.25 102	98.35 141	96.20 121	99.00 54	98.32 178	96.33 65	98.03 76	99.17 47	91.35 125	99.16 161	98.10 28	98.29 136	99.39 96

test_yl		97.22 93	96.78 97	98.54 79	98.73 114	96.60 103	98.45 151	98.31 180	94.70 131	98.02 77	98.42 133	90.80 137	99.70 102	96.81 94	96.79 171	99.34 99

DCV-MVSNet		97.22 93	96.78 97	98.54 79	98.73 114	96.60 103	98.45 151	98.31 180	94.70 131	98.02 77	98.42 133	90.80 137	99.70 102	96.81 94	96.79 171	99.34 99

nrg030		96.28 130	95.72 133	97.96 122	96.90 240	98.15 45	99.39 5	98.31 180	95.47 97	94.42 199	98.35 141	92.09 109	98.69 217	97.50 64	89.05 278	97.04 215

TAMVS		97.02 103	96.79 96	97.70 138	98.06 167	95.31 164	98.52 142	98.31 180	93.95 163	97.05 127	98.61 114	93.49 88	98.52 232	95.33 142	97.81 149	99.29 110

EPP-MVSNet		97.46 78	97.28 76	97.99 119	98.64 125	95.38 159	99.33 13	98.31 180	93.61 186	97.19 119	99.07 66	94.05 83	99.23 155	96.89 86	98.43 130	99.37 98

UnsupCasMVSNet_bld		87.17 293	85.12 296	93.31 295	91.94 318	88.77 297	94.92 315	98.30 185	84.30 314	82.30 317	90.04 319	63.96 326	97.25 305	85.85 301	74.47 325	93.93 319

Vis-MVSNet (Re-imp)		96.87 109	96.55 109	97.83 127	98.73 114	95.46 157	99.20 29	98.30 185	94.96 124	96.60 147	98.87 91	90.05 148	98.59 227	93.67 190	98.60 119	99.46 90

TSAR-MVS + GP.		98.38 37	98.24 35	98.81 64	99.22 81	97.25 79	98.11 199	98.29 187	97.19 34	98.99 28	99.02 69	96.22 17	99.67 109	98.52 13	98.56 122	99.51 77

MS-PatchMatch		93.84 247	93.63 233	94.46 284	96.18 272	89.45 288	97.76 232	98.27 188	92.23 234	92.13 272	97.49 213	79.50 289	98.69 217	89.75 269	99.38 87	95.25 304

EI-MVSNet		95.96 138	95.83 131	96.36 226	97.93 174	93.70 226	98.12 197	98.27 188	93.70 178	95.07 177	99.02 69	92.23 104	98.54 230	94.68 159	93.46 230	96.84 236

MVSTER		96.06 135	95.72 133	97.08 170	98.23 152	95.93 138	98.73 108	98.27 188	94.86 128	95.07 177	98.09 165	88.21 185	98.54 230	96.59 101	93.46 230	96.79 240

FMVSNet2		94.47 217	93.61 235	97.04 171	98.21 154	96.43 112	98.79 99	98.27 188	92.46 223	93.50 237	97.09 239	81.16 277	98.00 281	91.09 246	91.93 249	96.70 252

FMVSNet3		94.97 192	94.26 197	97.11 168	98.18 159	96.62 100	98.56 138	98.26 192	93.67 183	94.09 215	97.10 236	84.25 255	98.01 280	92.08 228	92.14 245	96.70 252

Fast-Effi-MVS+		96.28 130	95.70 137	98.03 117	98.29 150	95.97 132	98.58 133	98.25 193	91.74 243	95.29 176	97.23 230	91.03 134	99.15 164	92.90 210	97.96 144	98.97 144

PAPM		94.95 193	94.00 209	97.78 130	97.04 231	95.65 148	96.03 305	98.25 193	91.23 261	94.19 211	97.80 191	91.27 128	98.86 205	82.61 311	97.61 157	98.84 154

CANet_DTU		96.96 105	96.55 109	98.21 104	98.17 161	96.07 126	97.98 212	98.21 195	97.24 31	97.13 121	98.93 85	86.88 215	99.91 28	95.00 153	99.37 88	98.66 167

HY-MVS	93.96 8	96.82 111	96.23 121	98.57 75	98.46 136	97.00 86	98.14 194	98.21 195	93.95 163	96.72 142	97.99 173	91.58 118	99.76 91	94.51 168	96.54 179	98.95 148

PCF-MVS	93.45 11	94.68 205	93.43 242	98.42 91	98.62 127	96.77 96	95.48 311	98.20 197	84.63 313	93.34 240	98.32 147	88.55 179	99.81 60	84.80 307	98.96 102	98.68 164
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
v8		94.47 217	93.77 225	96.57 207	96.36 266	94.83 185	99.05 48	98.19 198	91.92 239	93.16 244	96.97 253	88.82 174	98.48 234	91.69 240	87.79 292	96.39 282

v10		94.29 226	93.55 237	96.51 214	96.39 265	94.80 187	98.99 56	98.19 198	91.35 254	93.02 250	96.99 251	88.09 189	98.41 250	90.50 257	88.41 286	96.33 285

mvs_anonymous		96.70 114	96.53 111	97.18 163	98.19 157	93.78 220	98.31 171	98.19 198	94.01 158	94.47 193	98.27 153	92.08 110	98.46 237	97.39 67	97.91 145	99.31 105

AllTest		95.24 177	94.65 179	96.99 173	99.25 74	93.21 238	98.59 131	98.18 201	91.36 252	93.52 235	98.77 101	84.67 248	99.72 96	89.70 271	97.87 147	98.02 190

TestCases				96.99 173	99.25 74	93.21 238		98.18 201	91.36 252	93.52 235	98.77 101	84.67 248	99.72 96	89.70 271	97.87 147	98.02 190

GBi-Net		94.49 215	93.80 222	96.56 208	98.21 154	95.00 174	98.82 88	98.18 201	92.46 223	94.09 215	97.07 241	81.16 277	97.95 283	92.08 228	92.14 245	96.72 248

test1		94.49 215	93.80 222	96.56 208	98.21 154	95.00 174	98.82 88	98.18 201	92.46 223	94.09 215	97.07 241	81.16 277	97.95 283	92.08 228	92.14 245	96.72 248

FMVSNet1		93.19 260	92.07 264	96.56 208	97.54 196	95.00 174	98.82 88	98.18 201	90.38 275	92.27 269	97.07 241	73.68 316	97.95 283	89.36 278	91.30 256	96.72 248

v1192		94.32 224	93.58 236	96.53 212	96.10 276	94.45 201	98.50 147	98.17 206	91.54 247	94.19 211	97.06 244	86.95 214	98.43 242	90.14 260	89.57 270	96.70 252

v1240		94.06 243	93.29 246	96.34 228	96.03 280	93.90 217	98.44 154	98.17 206	91.18 264	94.13 214	97.01 250	86.05 228	98.42 243	89.13 281	89.50 273	96.70 252

v144192		94.39 221	93.70 230	96.48 216	96.06 278	94.35 206	98.58 133	98.16 208	91.45 249	94.33 203	97.02 248	87.50 205	98.45 238	91.08 247	89.11 277	96.63 260

Fast-Effi-MVS+-dtu		95.87 142	95.85 130	95.91 241	97.74 185	91.74 258	98.69 119	98.15 209	95.56 93	94.92 180	97.68 200	88.98 169	98.79 212	93.19 201	97.78 151	97.20 212

v1921920		94.20 231	93.47 241	96.40 224	95.98 281	94.08 213	98.52 142	98.15 209	91.33 255	94.25 207	97.20 232	86.41 222	98.42 243	90.04 265	89.39 275	96.69 257

v1144		94.59 211	93.92 214	96.60 202	96.21 270	94.78 189	98.59 131	98.14 211	91.86 242	94.21 210	97.02 248	87.97 192	98.41 250	91.72 239	89.57 270	96.61 262

IterMVS-LS		95.46 160	95.21 157	96.22 233	98.12 163	93.72 225	98.32 170	98.13 212	93.71 176	94.26 206	97.31 225	92.24 103	98.10 272	94.63 160	90.12 264	96.84 236
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
EU-MVSNet		93.66 248	94.14 203	92.25 301	95.96 282	83.38 318	98.52 142	98.12 213	94.69 133	92.61 259	98.13 163	87.36 208	96.39 318	91.82 236	90.00 266	96.98 218

IterMVS		94.09 240	93.85 220	94.80 275	97.99 171	90.35 279	97.18 271	98.12 213	93.68 181	92.46 266	97.34 222	84.05 260	97.41 303	92.51 221	91.33 255	96.62 261
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
IterMVS-SCA-FT		94.11 238	93.87 218	94.85 272	97.98 173	90.56 277	97.18 271	98.11 215	93.75 171	92.58 260	97.48 214	83.97 262	97.41 303	92.48 223	91.30 256	96.58 265

COLMAP_ROB	93.27 12	95.33 173	94.87 173	96.71 190	99.29 66	93.24 237	98.58 133	98.11 215	89.92 282	93.57 233	99.10 59	86.37 223	99.79 80	90.78 253	98.10 141	97.09 213
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
Effi-MVS+-dtu		96.29 128	96.56 108	95.51 252	97.89 177	90.22 280	98.80 95	98.10 217	96.57 58	96.45 158	96.66 269	90.81 135	98.91 196	95.72 129	97.99 143	97.40 205

mvs-test1		96.60 116	96.68 105	96.37 225	97.89 177	91.81 254	98.56 138	98.10 217	96.57 58	96.52 154	97.94 176	90.81 135	99.45 140	95.72 129	98.01 142	97.86 194

1112_ss		96.63 115	96.00 127	98.50 83	98.56 130	96.37 114	98.18 191	98.10 217	92.92 210	94.84 182	98.43 131	92.14 107	99.58 121	94.35 172	96.51 180	99.56 73

V42		94.78 200	94.14 203	96.70 192	96.33 268	95.22 166	98.97 60	98.09 220	92.32 231	94.31 204	97.06 244	88.39 182	98.55 229	92.90 210	88.87 282	96.34 284

v2v482		94.69 203	94.03 207	96.65 195	96.17 273	94.79 188	98.67 123	98.08 221	92.72 216	94.00 220	97.16 234	87.69 201	98.45 238	92.91 209	88.87 282	96.72 248

MVS_Test		97.28 91	97.00 87	98.13 110	98.33 146	95.97 132	98.74 104	98.07 222	94.27 149	98.44 61	98.07 166	92.48 98	99.26 151	96.43 107	98.19 138	99.16 125

Test_1112_low_res		96.34 127	95.66 140	98.36 95	98.56 130	95.94 135	97.71 235	98.07 222	92.10 235	94.79 186	97.29 226	91.75 115	99.56 124	94.17 177	96.50 181	99.58 71

alignmvs		97.56 75	97.07 85	99.01 52	98.66 123	98.37 30	98.83 85	98.06 224	96.74 51	98.00 83	97.65 201	90.80 137	99.48 137	98.37 21	96.56 178	99.19 120

testing_2		90.61 283	88.50 288	96.95 178	90.08 323	95.57 151	97.69 237	98.06 224	93.02 205	76.55 320	92.48 316	61.18 327	98.44 240	95.45 140	91.98 248	96.84 236

RPSCF		94.87 197	95.40 143	93.26 296	98.89 103	82.06 323	98.33 166	98.06 224	90.30 276	96.56 148	99.26 34	87.09 210	99.49 133	93.82 186	96.32 186	98.24 185

test_normal		83.22 296	80.23 298	92.18 302	88.06 325	82.87 320	69.03 332	98.05 227	92.70 217	63.67 328	80.19 327	50.72 329	98.05 277	91.41 244	88.24 287	95.62 302

miper_lstm_enhance		94.33 223	94.07 206	95.11 264	97.75 183	90.97 268	97.22 268	98.03 228	91.67 245	92.76 255	96.97 253	90.03 149	97.78 294	92.51 221	89.64 269	96.56 269

pm-mvs1		93.94 246	93.06 249	96.59 203	96.49 261	95.16 167	98.95 64	98.03 228	92.32 231	91.08 283	97.84 185	84.54 251	98.41 250	92.16 226	86.13 310	96.19 290

v148		94.29 226	93.76 227	95.91 241	96.10 276	92.93 242	98.58 133	97.97 230	92.59 221	93.47 238	96.95 256	88.53 180	98.32 259	92.56 218	87.06 301	96.49 278

IS-MVSNet		97.22 93	96.88 92	98.25 102	98.85 108	96.36 115	99.19 31	97.97 230	95.39 101	97.23 118	98.99 75	91.11 131	98.93 194	94.60 163	98.59 120	99.47 86

pmmvs6		91.77 272	90.63 275	95.17 262	94.69 309	91.24 265	98.67 123	97.92 232	86.14 305	89.62 293	97.56 211	75.79 308	98.34 257	90.75 254	84.56 312	95.94 296

jason		97.32 90	97.08 84	98.06 116	97.45 205	95.59 149	97.87 223	97.91 233	94.79 130	98.55 55	98.83 94	91.12 130	99.23 155	97.58 58	99.60 57	99.34 99
jason: jason.
ppachtmachnet_test		93.22 258	92.63 257	94.97 268	95.45 297	90.84 269	96.88 288	97.88 234	90.60 270	92.08 273	97.26 227	88.08 190	97.86 293	85.12 306	90.33 263	96.22 288

tpm cat1		93.36 252	92.80 253	95.07 266	97.58 193	87.97 306	96.76 294	97.86 235	82.17 317	93.53 234	96.04 291	86.13 226	99.13 166	89.24 279	95.87 199	98.10 188

EG-PatchMatch MVS		91.13 277	90.12 279	94.17 289	94.73 308	89.00 296	98.13 196	97.81 236	89.22 291	85.32 313	96.46 276	67.71 321	98.42 243	87.89 290	93.82 224	95.08 307

BH-untuned		95.95 139	95.72 133	96.65 195	98.55 132	92.26 247	98.23 180	97.79 237	93.73 174	94.62 188	98.01 171	88.97 170	99.00 185	93.04 206	98.51 124	98.68 164

lupinMVS		97.44 82	97.22 79	98.12 112	98.07 165	95.76 146	97.68 238	97.76 238	94.50 143	98.79 40	98.61 114	92.34 99	99.30 149	97.58 58	99.59 60	99.31 105

VDDNet		95.36 170	94.53 184	97.86 125	98.10 164	95.13 170	98.85 81	97.75 239	90.46 272	98.36 64	99.39 12	73.27 317	99.64 113	97.98 33	96.58 177	98.81 155

ADS-MVSNet		95.00 188	94.45 191	96.63 198	98.00 169	91.91 253	96.04 303	97.74 240	90.15 277	96.47 156	96.64 271	87.89 194	98.96 189	90.08 262	97.06 165	99.02 139

tpmvs		94.60 209	94.36 194	95.33 258	97.46 201	88.60 300	96.88 288	97.68 241	91.29 258	93.80 228	96.42 279	88.58 176	99.24 154	91.06 248	96.04 198	98.17 186

pmmvs4		94.69 203	93.99 211	96.81 186	95.74 287	95.94 135	97.40 252	97.67 242	90.42 274	93.37 239	97.59 207	89.08 165	98.20 268	92.97 208	91.67 252	96.30 287

our_test_3		93.65 250	93.30 245	94.69 277	95.45 297	89.68 286	96.91 282	97.65 243	91.97 238	91.66 278	96.88 260	89.67 152	97.93 286	88.02 288	91.49 254	96.48 279

MVP-Stereo		94.28 228	93.92 214	95.35 257	94.95 304	92.60 245	97.97 213	97.65 243	91.61 246	90.68 287	97.09 239	86.32 224	98.42 243	89.70 271	99.34 89	95.02 309
Qingsong Yan: MVP-Stereo: A Parallel Multi-View Patchmatch Stereo Method with Dilation Matching for Photogrammetric Application.
SCA		95.46 160	95.13 160	96.46 220	97.67 187	91.29 264	97.33 261	97.60 245	94.68 134	96.92 133	97.10 236	83.97 262	98.89 200	92.59 216	98.32 135	99.20 117

GA-MVS		94.81 199	94.03 207	97.14 165	97.15 226	93.86 218	96.76 294	97.58 246	94.00 159	94.76 187	97.04 246	80.91 280	98.48 234	91.79 237	96.25 192	99.09 132

test20.03		90.89 280	90.38 277	92.43 299	93.48 314	88.14 305	98.33 166	97.56 247	93.40 193	87.96 301	96.71 268	80.69 284	94.13 325	79.15 319	86.17 308	95.01 310

CR-MVSNet		94.76 201	94.15 202	96.59 203	97.00 232	93.43 231	94.96 313	97.56 247	92.46 223	96.93 131	96.24 282	88.15 187	97.88 291	87.38 291	96.65 175	98.46 176

Patchmtry		93.22 258	92.35 261	95.84 244	96.77 245	93.09 241	94.66 318	97.56 247	87.37 299	92.90 252	96.24 282	88.15 187	97.90 287	87.37 292	90.10 265	96.53 273

tpmrst		95.63 154	95.69 138	95.44 255	97.54 196	88.54 301	96.97 278	97.56 247	93.50 189	97.52 114	96.93 258	89.49 153	99.16 161	95.25 148	96.42 183	98.64 169

FMVSNet5		91.81 271	90.92 273	94.49 281	97.21 219	92.09 249	98.00 211	97.55 251	89.31 290	90.86 285	95.61 302	74.48 313	95.32 321	85.57 302	89.70 268	96.07 293

testgi		93.06 262	92.45 260	94.88 271	96.43 264	89.90 281	98.75 101	97.54 252	95.60 91	91.63 279	97.91 178	74.46 314	97.02 308	86.10 298	93.67 225	97.72 199

PatchmatchNet		95.71 150	95.52 141	96.29 231	97.58 193	90.72 273	96.84 291	97.52 253	94.06 154	97.08 123	96.96 255	89.24 161	98.90 199	92.03 232	98.37 131	99.26 113
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
MDA-MVSNet-bldmvs		89.97 286	88.35 290	94.83 274	95.21 301	91.34 261	97.64 241	97.51 254	88.36 295	71.17 326	96.13 289	79.22 291	96.63 315	83.65 308	86.27 307	96.52 274

USDC		93.33 255	92.71 255	95.21 260	96.83 244	90.83 270	96.91 282	97.50 255	93.84 168	90.72 286	98.14 162	77.69 299	98.82 209	89.51 275	93.21 238	95.97 295

ITE_SJBPF				95.44 255	97.42 206	91.32 263		97.50 255	95.09 119	93.59 231	98.35 141	81.70 275	98.88 202	89.71 270	93.39 234	96.12 291

Patchmatch-test		94.42 219	93.68 232	96.63 198	97.60 191	91.76 256	94.83 317	97.49 257	89.45 288	94.14 213	97.10 236	88.99 166	98.83 208	85.37 305	98.13 140	99.29 110

YYNet1		90.70 282	89.39 284	94.62 280	94.79 307	90.65 275	97.20 269	97.46 258	87.54 298	72.54 324	95.74 295	86.51 219	96.66 314	86.00 299	86.76 306	96.54 272

MDA-MVSNet_test_wron		90.71 281	89.38 285	94.68 278	94.83 306	90.78 272	97.19 270	97.46 258	87.60 297	72.41 325	95.72 298	86.51 219	96.71 313	85.92 300	86.80 305	96.56 269

BH-RMVSNet		95.92 141	95.32 152	97.69 139	98.32 148	94.64 191	98.19 187	97.45 260	94.56 139	96.03 167	98.61 114	85.02 242	99.12 167	90.68 255	99.06 99	99.30 108

MIMVSNet1		89.67 288	88.28 291	93.82 290	92.81 317	91.08 267	98.01 209	97.45 260	87.95 296	87.90 302	95.87 294	67.63 322	94.56 324	78.73 321	88.18 289	95.83 298

OurMVSNet-221017-0		94.21 230	94.00 209	94.85 272	95.60 291	89.22 292	98.89 73	97.43 262	95.29 108	92.18 271	98.52 126	82.86 270	98.59 227	93.46 194	91.76 251	96.74 245

BH-w/o		95.38 167	95.08 163	96.26 232	98.34 145	91.79 255	97.70 236	97.43 262	92.87 213	94.24 208	97.22 231	88.66 175	98.84 206	91.55 242	97.70 155	98.16 187

VDD-MVS		95.82 146	95.23 156	97.61 146	98.84 109	93.98 215	98.68 120	97.40 264	95.02 121	97.95 85	99.34 24	74.37 315	99.78 84	98.64 3	96.80 170	99.08 135

Gipuma		78.40 298	76.75 300	83.38 312	95.54 293	80.43 324	79.42 331	97.40 264	64.67 326	73.46 323	80.82 326	45.65 331	93.14 326	66.32 327	87.43 296	76.56 329
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
DI_MVS_plusplus_test		94.74 202	93.62 234	98.09 113	95.34 300	95.92 139	98.09 202	97.34 266	94.66 137	85.89 309	95.91 293	80.49 285	99.38 145	96.66 99	98.22 137	98.97 144

PatchFormer-LS_test		95.47 159	95.27 155	96.08 237	97.59 192	90.66 274	98.10 201	97.34 266	93.98 161	96.08 165	96.15 288	87.65 202	99.12 167	95.27 147	95.24 204	98.44 178

new-patchmatchnet		88.50 291	87.45 293	91.67 304	90.31 322	85.89 314	97.16 274	97.33 268	89.47 287	83.63 316	92.77 313	76.38 305	95.06 323	82.70 310	77.29 322	94.06 317

ADS-MVSNet2		94.58 212	94.40 193	95.11 264	98.00 169	88.74 298	96.04 303	97.30 269	90.15 277	96.47 156	96.64 271	87.89 194	97.56 300	90.08 262	97.06 165	99.02 139

MDTV_nov1_ep13			95.40 143		97.48 199	88.34 303	96.85 290	97.29 270	93.74 173	97.48 115	97.26 227	89.18 162	99.05 178	91.92 235	97.43 161

pmmvs5		93.65 250	92.97 251	95.68 249	95.49 295	92.37 246	98.20 184	97.28 271	89.66 285	92.58 260	97.26 227	82.14 272	98.09 274	93.18 202	90.95 259	96.58 265

EPNet_dtu		95.21 179	94.95 170	95.99 238	96.17 273	90.45 278	98.16 193	97.27 272	96.77 49	93.14 247	98.33 146	90.34 144	98.42 243	85.57 302	98.81 112	99.09 132
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
Anonymous20231206		91.66 273	91.10 272	93.33 294	94.02 313	87.35 310	98.58 133	97.26 273	90.48 271	90.16 289	96.31 280	83.83 266	96.53 316	79.36 318	89.90 267	96.12 291

RPMNet		92.52 267	91.17 271	96.59 203	97.00 232	93.43 231	94.96 313	97.26 273	82.27 316	96.93 131	92.12 318	86.98 213	97.88 291	76.32 324	96.65 175	98.46 176

test_0402		91.32 275	90.27 278	94.48 282	96.60 255	91.12 266	98.50 147	97.22 275	86.10 306	88.30 300	96.98 252	77.65 301	97.99 282	78.13 322	92.94 240	94.34 312

dp		94.15 235	93.90 216	94.90 270	97.31 213	86.82 313	96.97 278	97.19 276	91.22 262	96.02 168	96.61 273	85.51 235	99.02 184	90.00 266	94.30 207	98.85 152

thres200		95.25 176	94.57 182	97.28 159	98.81 110	94.92 181	98.20 184	97.11 277	95.24 112	96.54 152	96.22 286	84.58 250	99.53 130	87.93 289	96.50 181	97.39 206

PatchT		93.06 262	91.97 266	96.35 227	96.69 251	92.67 244	94.48 319	97.08 278	86.62 301	97.08 123	92.23 317	87.94 193	97.90 287	78.89 320	96.69 173	98.49 175

TDRefinement		91.06 278	89.68 282	95.21 260	85.35 327	91.49 260	98.51 146	97.07 279	91.47 248	88.83 298	97.84 185	77.31 303	99.09 175	92.79 213	77.98 321	95.04 308

LF4IMVS		93.14 261	92.79 254	94.20 287	95.88 284	88.67 299	97.66 240	97.07 279	93.81 170	91.71 277	97.65 201	77.96 298	98.81 210	91.47 243	91.92 250	95.12 305

Anonymous202405211		95.28 175	94.49 186	97.67 141	99.00 95	93.75 223	98.70 117	97.04 281	90.66 269	96.49 155	98.80 97	78.13 296	99.83 51	96.21 113	95.36 203	99.44 93

baseline1		95.84 144	95.12 161	98.01 118	98.49 135	95.98 127	98.73 108	97.03 282	95.37 104	96.22 162	98.19 159	89.96 150	99.16 161	94.60 163	87.48 295	98.90 151

MIMVSNet		93.26 257	92.21 263	96.41 223	97.73 186	93.13 240	95.65 310	97.03 282	91.27 260	94.04 218	96.06 290	75.33 309	97.19 306	86.56 295	96.23 193	98.92 150

EPNet		97.28 91	96.87 93	98.51 82	94.98 303	96.14 124	98.90 69	97.02 284	98.28 1	95.99 169	99.11 56	91.36 124	99.89 33	96.98 78	99.19 95	99.50 79
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
TR-MVS		94.94 195	94.20 199	97.17 164	97.75 183	94.14 212	97.59 244	97.02 284	92.28 233	95.75 171	97.64 203	83.88 264	98.96 189	89.77 268	96.15 195	98.40 179

JIA-IIPM		93.35 253	92.49 259	95.92 240	96.48 262	90.65 275	95.01 312	96.96 286	85.93 307	96.08 165	87.33 322	87.70 200	98.78 213	91.35 245	95.58 202	98.34 182

pmmvs-eth3d		90.36 284	89.05 286	94.32 286	91.10 320	92.12 248	97.63 243	96.95 287	88.86 293	84.91 314	93.13 312	78.32 295	96.74 310	88.70 283	81.81 317	94.09 316

tfpn200view9		95.32 174	94.62 180	97.43 153	98.94 100	94.98 177	98.68 120	96.93 288	95.33 105	96.55 150	96.53 274	84.23 256	99.56 124	88.11 285	96.29 187	97.76 195

thres400		95.38 167	94.62 180	97.65 144	98.94 100	94.98 177	98.68 120	96.93 288	95.33 105	96.55 150	96.53 274	84.23 256	99.56 124	88.11 285	96.29 187	98.40 179

thres100view900		95.38 167	94.70 178	97.41 154	98.98 98	94.92 181	98.87 78	96.90 290	95.38 102	96.61 146	96.88 260	84.29 253	99.56 124	88.11 285	96.29 187	97.76 195

thres600view7		95.49 158	94.77 175	97.67 141	98.98 98	95.02 173	98.85 81	96.90 290	95.38 102	96.63 145	96.90 259	84.29 253	99.59 120	88.65 284	96.33 185	98.40 179

CostFormer		94.95 193	94.73 177	95.60 251	97.28 214	89.06 294	97.53 247	96.89 292	89.66 285	96.82 138	96.72 267	86.05 228	98.95 193	95.53 137	96.13 196	98.79 156

new_pmnet		90.06 285	89.00 287	93.22 297	94.18 310	88.32 304	96.42 301	96.89 292	86.19 304	85.67 312	93.62 310	77.18 304	97.10 307	81.61 313	89.29 276	94.23 313

OpenMVS_ROB	86.42 20	89.00 290	87.43 294	93.69 291	93.08 316	89.42 289	97.91 217	96.89 292	78.58 320	85.86 310	94.69 307	69.48 320	98.29 265	77.13 323	93.29 237	93.36 321

tpm2		94.19 232	93.76 227	95.46 254	97.23 217	89.04 295	97.31 263	96.85 295	87.08 300	96.21 163	96.79 265	83.75 268	98.74 215	92.43 224	96.23 193	98.59 171

TransMVSNet (Re)		92.67 265	91.51 270	96.15 235	96.58 256	94.65 190	98.90 69	96.73 296	90.86 268	89.46 295	97.86 182	85.62 234	98.09 274	86.45 296	81.12 318	95.71 299

ambc				89.49 307	86.66 326	75.78 326	92.66 324	96.72 297		86.55 307	92.50 315	46.01 330	97.90 287	90.32 258	82.09 314	94.80 311

LCM-MVSNet		78.70 297	76.24 301	86.08 309	77.26 333	71.99 329	94.34 320	96.72 297	61.62 327	76.53 321	89.33 320	33.91 336	92.78 327	81.85 312	74.60 324	93.46 320

TinyColmap		92.31 269	91.53 269	94.65 279	96.92 237	89.75 283	96.92 280	96.68 299	90.45 273	89.62 293	97.85 184	76.06 307	98.81 210	86.74 294	92.51 243	95.41 303

Baseline_NR-MVSNet		94.35 222	93.81 221	95.96 239	96.20 271	94.05 214	98.61 130	96.67 300	91.44 250	93.85 225	97.60 206	88.57 177	98.14 270	94.39 170	86.93 302	95.68 300

SixPastTwentyTwo		93.34 254	92.86 252	94.75 276	95.67 289	89.41 290	98.75 101	96.67 300	93.89 165	90.15 290	98.25 155	80.87 281	98.27 266	90.90 251	90.64 261	96.57 267

DWT-MVSNet_test		94.82 198	94.36 194	96.20 234	97.35 211	90.79 271	98.34 165	96.57 302	92.91 211	95.33 175	96.44 278	82.00 273	99.12 167	94.52 167	95.78 201	98.70 161

LFMVS		95.86 143	94.98 168	98.47 86	98.87 105	96.32 117	98.84 84	96.02 303	93.40 193	98.62 51	99.20 43	74.99 311	99.63 116	97.72 49	97.20 164	99.46 90

IB-MVS	91.98 17	93.27 256	91.97 266	97.19 162	97.47 200	93.41 233	97.09 276	95.99 304	93.32 196	92.47 265	95.73 296	78.06 297	99.53 130	94.59 165	82.98 313	98.62 170
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
test0.0.03 1		94.08 241	93.51 239	95.80 245	95.53 294	92.89 243	97.38 254	95.97 305	95.11 116	92.51 264	96.66 269	87.71 198	96.94 309	87.03 293	93.67 225	97.57 202

FPMVS		77.62 300	77.14 299	79.05 314	79.25 331	60.97 333	95.79 308	95.94 306	65.96 325	67.93 327	94.40 308	37.73 334	88.88 330	68.83 326	88.46 285	87.29 324

Patchmatch-RL test		91.49 274	90.85 274	93.41 293	91.37 319	84.40 315	92.81 323	95.93 307	91.87 241	87.25 303	94.87 306	88.99 166	96.53 316	92.54 220	82.00 315	99.30 108

tpm		94.13 236	93.80 222	95.12 263	96.50 260	87.91 307	97.44 249	95.89 308	92.62 219	96.37 160	96.30 281	84.13 259	98.30 263	93.24 199	91.66 253	99.14 128

LCM-MVSNet-Re		95.22 178	95.32 152	94.91 269	98.18 159	87.85 308	98.75 101	95.66 309	95.11 116	88.96 297	96.85 262	90.26 147	97.65 296	95.65 134	98.44 128	99.22 116

ET-MVSNet_ETH3D		94.13 236	92.98 250	97.58 147	98.22 153	96.20 121	97.31 263	95.37 310	94.53 140	79.56 319	97.63 205	86.51 219	97.53 301	96.91 83	90.74 260	99.02 139

test-LLR		95.10 185	94.87 173	95.80 245	96.77 245	89.70 284	96.91 282	95.21 311	95.11 116	94.83 184	95.72 298	87.71 198	98.97 186	93.06 204	98.50 125	98.72 159

test-mter		94.08 241	93.51 239	95.80 245	96.77 245	89.70 284	96.91 282	95.21 311	92.89 212	94.83 184	95.72 298	77.69 299	98.97 186	93.06 204	98.50 125	98.72 159

PM-MVS		87.77 292	86.55 295	91.40 305	91.03 321	83.36 319	96.92 280	95.18 313	91.28 259	86.48 308	93.42 311	53.27 328	96.74 310	89.43 277	81.97 316	94.11 315

DeepMVS_CX				86.78 308	97.09 230	72.30 328		95.17 314	75.92 322	84.34 315	95.19 303	70.58 319	95.35 320	79.98 317	89.04 279	92.68 322

K. test v3		92.55 266	91.91 268	94.48 282	95.64 290	89.24 291	99.07 46	94.88 315	94.04 155	86.78 305	97.59 207	77.64 302	97.64 297	92.08 228	89.43 274	96.57 267

TESTMET0.1,1		94.18 234	93.69 231	95.63 250	96.92 237	89.12 293	96.91 282	94.78 316	93.17 200	94.88 181	96.45 277	78.52 294	98.92 195	93.09 203	98.50 125	98.85 152

pmmvs3		86.67 295	84.86 297	92.11 303	88.16 324	87.19 312	96.63 297	94.75 317	79.88 319	87.22 304	92.75 314	66.56 323	95.20 322	81.24 314	76.56 323	93.96 318

door								94.64 318

thisisatest0515		95.61 157	94.89 172	97.76 132	98.15 162	95.15 169	96.77 293	94.41 319	92.95 209	97.18 120	97.43 219	84.78 247	99.45 140	94.63 160	97.73 154	98.68 164

door-mid								94.37 320

tttt0517		96.07 134	95.51 142	97.78 130	98.41 138	94.84 183	99.28 16	94.33 321	94.26 150	97.64 107	98.64 113	84.05 260	99.47 138	95.34 141	97.60 158	99.03 138

DSMNet-mixed		92.52 267	92.58 258	92.33 300	94.15 311	82.65 321	98.30 173	94.26 322	89.08 292	92.65 258	95.73 296	85.01 243	95.76 319	86.24 297	97.76 152	98.59 171

thisisatest0530		96.01 136	95.36 148	97.97 120	98.38 139	95.52 155	98.88 76	94.19 323	94.04 155	97.64 107	98.31 148	83.82 267	99.46 139	95.29 145	97.70 155	98.93 149

MTMP							98.89 73	94.14 324

baseline2		95.11 184	94.52 185	96.87 183	96.65 254	93.56 227	98.27 178	94.10 325	93.45 191	92.02 275	97.43 219	87.45 207	99.19 159	93.88 184	97.41 162	97.87 193

PMVS	61.03 23	65.95 303	63.57 306	73.09 317	57.90 336	51.22 337	85.05 330	93.93 326	54.45 328	44.32 334	83.57 323	13.22 338	89.15 329	58.68 329	81.00 319	78.91 328
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
PMMVS2		77.95 299	75.44 302	85.46 310	82.54 328	74.95 327	94.23 321	93.08 327	72.80 324	74.68 322	87.38 321	36.36 335	91.56 328	73.95 325	63.94 327	89.87 323

MVS-HIRNet		89.46 289	88.40 289	92.64 298	97.58 193	82.15 322	94.16 322	93.05 328	75.73 323	90.90 284	82.52 324	79.42 290	98.33 258	83.53 309	98.68 114	97.43 203

EPMVS		94.99 189	94.48 187	96.52 213	97.22 218	91.75 257	97.23 267	91.66 329	94.11 152	97.28 116	96.81 264	85.70 233	98.84 206	93.04 206	97.28 163	98.97 144

lessismore_v0				94.45 285	94.93 305	88.44 302		91.03 330		86.77 306	97.64 203	76.23 306	98.42 243	90.31 259	85.64 311	96.51 276

ANet_high		69.08 301	65.37 304	80.22 313	65.99 335	71.96 330	90.91 327	90.09 331	82.62 315	49.93 333	78.39 328	29.36 337	81.75 331	62.49 328	38.52 331	86.95 326

gg-mvs-nofinetune		92.21 270	90.58 276	97.13 166	96.75 248	95.09 171	95.85 307	89.40 332	85.43 311	94.50 192	81.98 325	80.80 283	98.40 256	92.16 226	98.33 134	97.88 192

GG-mvs-BLEND				96.59 203	96.34 267	94.98 177	96.51 300	88.58 333		93.10 249	94.34 309	80.34 287	98.05 277	89.53 274	96.99 167	96.74 245

E-PMN		64.94 304	64.25 305	67.02 318	82.28 329	59.36 335	91.83 326	85.63 334	52.69 329	60.22 330	77.28 329	41.06 333	80.12 333	46.15 331	41.14 329	61.57 331

EMVS		64.07 305	63.26 307	66.53 319	81.73 330	58.81 336	91.85 325	84.75 335	51.93 331	59.09 331	75.13 330	43.32 332	79.09 334	42.03 332	39.47 330	61.69 330

tmp_tt		68.90 302	66.97 303	74.68 316	50.78 337	59.95 334	87.13 328	83.47 336	38.80 332	62.21 329	96.23 284	64.70 325	76.91 335	88.91 282	30.49 332	87.19 325

MVE	62.14 22	63.28 306	59.38 308	74.99 315	74.33 334	65.47 331	85.55 329	80.50 337	52.02 330	51.10 332	75.00 331	10.91 341	80.50 332	51.60 330	53.40 328	78.99 327
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
N_pmnet		87.12 294	87.77 292	85.17 311	95.46 296	61.92 332	97.37 256	70.66 338	85.83 308	88.73 299	96.04 291	85.33 240	97.76 295	80.02 315	90.48 262	95.84 297

wuyk23d		30.17 307	30.18 310	30.16 320	78.61 332	43.29 338	66.79 333	14.21 339	17.31 333	14.82 337	11.93 337	11.55 340	41.43 336	37.08 333	19.30 333	5.76 334

testmvs		21.48 309	24.95 311	11.09 322	14.89 338	6.47 340	96.56 298	9.87 340	7.55 334	17.93 335	39.02 333	9.43 342	5.90 338	16.56 335	12.72 334	20.91 333

test123		20.95 310	23.72 312	12.64 321	13.54 339	8.19 339	96.55 299	6.13 341	7.48 335	16.74 336	37.98 334	12.97 339	6.05 337	16.69 334	5.43 335	23.68 332

pcd_1.5k_mvsjas		7.88 312	10.50 314	0.00 323	0.00 340	0.00 341	0.00 334	0.00 342	0.00 336	0.00 338	0.00 338	94.51 73	0.00 339	0.00 336	0.00 336	0.00 335

sosnet-low-res		0.00 313	0.00 315	0.00 323	0.00 340	0.00 341	0.00 334	0.00 342	0.00 336	0.00 338	0.00 338	0.00 343	0.00 339	0.00 336	0.00 336	0.00 335

sosnet		0.00 313	0.00 315	0.00 323	0.00 340	0.00 341	0.00 334	0.00 342	0.00 336	0.00 338	0.00 338	0.00 343	0.00 339	0.00 336	0.00 336	0.00 335

uncertanet		0.00 313	0.00 315	0.00 323	0.00 340	0.00 341	0.00 334	0.00 342	0.00 336	0.00 338	0.00 338	0.00 343	0.00 339	0.00 336	0.00 336	0.00 335

Regformer		0.00 313	0.00 315	0.00 323	0.00 340	0.00 341	0.00 334	0.00 342	0.00 336	0.00 338	0.00 338	0.00 343	0.00 339	0.00 336	0.00 336	0.00 335

n2								0.00 342

nn								0.00 342

ab-mvs-re		8.20 311	10.94 313	0.00 323	0.00 340	0.00 341	0.00 334	0.00 342	0.00 336	0.00 338	98.43 131	0.00 343	0.00 339	0.00 336	0.00 336	0.00 335

uanet		0.00 313	0.00 315	0.00 323	0.00 340	0.00 341	0.00 334	0.00 342	0.00 336	0.00 338	0.00 338	0.00 343	0.00 339	0.00 336	0.00 336	0.00 335

save filter2										98.81 39	99.11 56	96.33 16	99.92 18	97.95 34	99.76 28	99.67 53

test_0728_THIRD									97.32 23	99.45 5	99.46 7	97.88 1	99.94 3	98.47 15	99.86 1	99.85 2

GSMVS																99.20 117

test_part2					99.63 25	99.18 5				99.27 11

sam_mvs1												89.45 155				99.20 117

sam_mvs												88.99 166

test_post1							96.68 296				30.43 336	87.85 197	98.69 217	92.59 216

test_post											31.83 335	88.83 173	98.91 196

patchmatchnet-post											95.10 305	89.42 156	98.89 200

gm-plane-assit					95.88 284	87.47 309			89.74 284		96.94 257		99.19 159	93.32 198

test9_res														96.39 109	99.57 63	99.69 43

agg_prior2														95.87 123	99.57 63	99.68 49

test_prior4						98.01 51	97.86 224

test_prior2							97.80 229		96.12 73	97.89 92	98.69 106	95.96 31		96.89 86	99.60 57

旧先验2							97.57 246		91.30 257	98.67 48			99.80 68	95.70 133

新几何2							97.64 241

原ACMM2							97.67 239

testdata2													99.89 33	91.65 241

segment_acmp												96.85 7

testdata1							97.32 262		96.34 64

plane_prior7					97.42 206	94.63 192

plane_prior6					97.35 211	94.61 195						87.09 210

plane_prior4											98.28 150

plane_prior3						94.61 195			97.02 43	95.34 173

plane_prior2							98.80 95		97.28 25

plane_prior1					97.37 210

plane_prior						94.60 197	98.44 154		96.74 51						94.22 210

HQP5-MVS						94.25 210

HQP-NCC					97.20 220		98.05 205		96.43 60	94.45 194

ACMP_Plane					97.20 220		98.05 205		96.43 60	94.45 194

BP-MVS														95.30 143

HQP4-MVS										94.45 194			98.96 189			96.87 233

HQP2-MVS												86.75 216

NP-MVS					97.28 214	94.51 200					97.73 194

MDTV_nov1_ep13_2view						84.26 316	96.89 287		90.97 267	97.90 91		89.89 151		93.91 183		99.18 124

ACMMP++_ref															92.97 239

ACMMP++															93.61 228

Test By Simon												94.64 69

High-resolution multi-view benchmark