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.

APDe-MVS		99.02 3	98.84 2	99.55 6	99.57 33	98.96 12	99.39 5	98.93 37	97.38 24	99.41 11	99.54 1	96.66 13	99.84 52	98.86 1	99.85 3	99.87 1

CANet		98.05 56	97.76 61	98.90 71	98.73 128	97.27 90	98.35 175	98.78 94	97.37 26	97.72 113	98.96 93	91.53 138	99.92 21	98.79 2	99.65 58	99.51 89

Regformer-4		98.64 14	98.53 11	98.99 63	99.43 57	97.37 86	98.40 171	98.79 91	97.46 19	99.09 30	99.31 35	95.86 42	99.80 79	98.64 3	99.76 32	99.79 10

VDD-MVS		95.82 156	95.23 166	97.61 155	98.84 123	93.98 229	98.68 129	97.40 287	95.02 132	97.95 98	99.34 31	74.37 336	99.78 95	98.64 3	96.80 182	99.08 147

EI-MVSNet-Vis-set		98.47 38	98.39 19	98.69 78	99.46 51	96.49 124	98.30 185	98.69 117	97.21 36	98.84 46	99.36 26	95.41 54	99.78 95	98.62 5	99.65 58	99.80 9

Regformer-3		98.59 20	98.50 14	98.86 73	99.43 57	97.05 100	98.40 171	98.68 120	97.43 20	99.06 31	99.31 35	95.80 43	99.77 100	98.62 5	99.76 32	99.78 13

EI-MVSNet-UG-set		98.41 41	98.34 28	98.61 83	99.45 55	96.32 132	98.28 188	98.68 120	97.17 39	98.74 53	99.37 22	95.25 66	99.79 91	98.57 7	99.54 84	99.73 36

CHOSEN 280x420		97.18 106	97.18 89	97.20 172	98.81 124	93.27 255	95.78 328	99.15 18	95.25 119	96.79 154	98.11 178	92.29 115	99.07 188	98.56 8	99.85 3	99.25 126

xiu_mvs_v1_base_debu		97.60 78	97.56 69	97.72 144	98.35 155	95.98 142	97.86 236	98.51 161	97.13 42	99.01 35	98.40 149	91.56 134	99.80 79	98.53 9	98.68 127	97.37 218

xiu_mvs_v1_base		97.60 78	97.56 69	97.72 144	98.35 155	95.98 142	97.86 236	98.51 161	97.13 42	99.01 35	98.40 149	91.56 134	99.80 79	98.53 9	98.68 127	97.37 218

xiu_mvs_v1_base_debi		97.60 78	97.56 69	97.72 144	98.35 155	95.98 142	97.86 236	98.51 161	97.13 42	99.01 35	98.40 149	91.56 134	99.80 79	98.53 9	98.68 127	97.37 218

VNet		97.79 69	97.40 81	98.96 67	98.88 118	97.55 80	98.63 137	98.93 37	96.74 55	99.02 34	98.84 106	90.33 162	99.83 55	98.53 9	96.66 186	99.50 91

MSLP-MVS++		98.56 28	98.57 8	98.55 87	99.26 85	96.80 109	98.71 122	99.05 24	97.28 29	98.84 46	99.28 40	96.47 18	99.40 154	98.52 13	99.70 51	99.47 98

TSAR-MVS + GP.		98.38 43	98.24 41	98.81 74	99.22 94	97.25 94	98.11 213	98.29 204	97.19 38	98.99 38	99.02 80	96.22 20	99.67 121	98.52 13	98.56 135	99.51 89

DVP-MVS		99.03 2	98.83 3	99.63 3	99.72 12	99.25 2	98.97 68	98.58 146	97.62 11	99.45 9	99.46 9	97.42 6	99.94 3	98.47 15	99.81 10	99.69 51

test_0728_THIRD									97.32 27	99.45 9	99.46 9	97.88 1	99.94 3	98.47 15	99.86 1	99.85 2

test_0728_SECOND				99.71 1	99.72 12	99.35 1	98.97 68	98.88 49					99.94 3	98.47 15	99.81 10	99.84 4

SED-MVS		99.09 1	98.91 1	99.63 3	99.71 20	99.24 4	99.02 58	98.87 55	97.65 9	99.73 1	99.48 6	97.53 4	99.94 3	98.43 18	99.81 10	99.70 48

test_241102_TWO								98.87 55	97.65 9	99.53 8	99.48 6	97.34 8	99.94 3	98.43 18	99.80 17	99.83 5

Regformer-1		98.66 12	98.51 13	99.12 57	99.35 60	97.81 73	98.37 173	98.76 98	97.49 17	99.20 22	99.21 48	96.08 29	99.79 91	98.42 20	99.73 43	99.75 28

DELS-MVS		98.40 42	98.20 44	98.99 63	99.00 109	97.66 75	97.75 245	98.89 46	97.71 8	98.33 78	98.97 87	94.97 74	99.88 43	98.42 20	99.76 32	99.42 107
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
Regformer-2		98.69 11	98.52 12	99.19 43	99.35 60	98.01 62	98.37 173	98.81 76	97.48 18	99.21 21	99.21 48	96.13 27	99.80 79	98.40 22	99.73 43	99.75 28

alignmvs		97.56 84	97.07 94	99.01 62	98.66 137	98.37 41	98.83 94	98.06 245	96.74 55	98.00 96	97.65 217	90.80 153	99.48 149	98.37 23	96.56 190	99.19 132

IU-MVS					99.71 20	99.23 6		98.64 136	95.28 117	99.63 4				98.35 24	99.81 10	99.83 5

TSAR-MVS + MP.		98.78 6	98.62 7	99.24 40	99.69 25	98.28 48	99.14 36	98.66 131	96.84 51	99.56 5	99.31 35	96.34 19	99.70 114	98.32 25	99.73 43	99.73 36
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
DeepPCF-MVS	96.37 2	97.93 63	98.48 17	96.30 241	99.00 109	89.54 309	97.43 263	98.87 55	98.16 2	99.26 18	99.38 21	96.12 28	99.64 125	98.30 26	99.77 26	99.72 40

canonicalmvs		97.67 74	97.23 87	98.98 65	98.70 133	98.38 35	99.34 11	98.39 184	96.76 54	97.67 116	97.40 237	92.26 116	99.49 145	98.28 27	96.28 202	99.08 147

SD-MVS		98.64 14	98.68 5	98.53 91	99.33 65	98.36 42	98.90 78	98.85 64	97.28 29	99.72 3	99.39 14	96.63 15	97.60 317	98.17 28	99.85 3	99.64 70
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
diffmvs		97.58 82	97.40 81	98.13 120	98.32 162	95.81 159	98.06 216	98.37 187	96.20 75	98.74 53	98.89 101	91.31 143	99.25 164	98.16 29	98.52 136	99.34 111

casdiffmvs		97.63 77	97.41 80	98.28 108	98.33 160	96.14 139	98.82 97	98.32 194	96.38 70	97.95 98	99.21 48	91.23 145	99.23 167	98.12 30	98.37 144	99.48 96

baseline		97.64 76	97.44 79	98.25 112	98.35 155	96.20 136	99.00 62	98.32 194	96.33 72	98.03 89	99.17 56	91.35 141	99.16 173	98.10 31	98.29 149	99.39 108

MP-MVS-pluss		98.31 52	97.92 57	99.49 9	99.72 12	98.88 14	98.43 167	98.78 94	94.10 165	97.69 115	99.42 12	95.25 66	99.92 21	98.09 32	99.80 17	99.67 61
MP-MVS-pluss: MP-MVS-pluss. MP-MVS-pluss
SMA-MVS		98.58 23	98.25 38	99.56 5	99.51 39	99.04 11	98.95 72	98.80 86	93.67 194	99.37 13	99.52 3	96.52 17	99.89 35	98.06 33	99.81 10	99.76 26
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
CNVR-MVS		98.78 6	98.56 9	99.45 14	99.32 68	98.87 15	98.47 161	98.81 76	97.72 6	98.76 52	99.16 61	97.05 10	99.78 95	98.06 33	99.66 57	99.69 51

MVS_111021_HR		98.47 38	98.34 28	98.88 72	99.22 94	97.32 87	97.91 229	99.58 3	97.20 37	98.33 78	99.00 85	95.99 35	99.64 125	98.05 35	99.76 32	99.69 51

VDDNet		95.36 179	94.53 196	97.86 134	98.10 179	95.13 184	98.85 90	97.75 262	90.46 292	98.36 76	99.39 14	73.27 338	99.64 125	97.98 36	96.58 189	98.81 166

MCST-MVS		98.65 13	98.37 21	99.48 10	99.60 31	98.87 15	98.41 170	98.68 120	97.04 46	98.52 67	98.80 110	96.78 12	99.83 55	97.93 37	99.61 67	99.74 33

zzz-MVS		98.55 30	98.25 38	99.46 12	99.76 1	98.64 22	98.55 151	98.74 102	97.27 33	98.02 90	99.39 14	94.81 77	99.96 1	97.91 38	99.79 19	99.77 20

MTAPA		98.58 23	98.29 35	99.46 12	99.76 1	98.64 22	98.90 78	98.74 102	97.27 33	98.02 90	99.39 14	94.81 77	99.96 1	97.91 38	99.79 19	99.77 20

MVS_111021_LR		98.34 48	98.23 42	98.67 80	99.27 83	96.90 106	97.95 226	99.58 3	97.14 41	98.44 72	99.01 84	95.03 73	99.62 130	97.91 38	99.75 38	99.50 91

ACMMP_NAP		98.61 17	98.30 34	99.55 6	99.62 30	98.95 13	98.82 97	98.81 76	95.80 90	99.16 26	99.47 8	95.37 57	99.92 21	97.89 41	99.75 38	99.79 10

PS-MVSNAJ		97.73 71	97.77 60	97.62 154	98.68 136	95.58 164	97.34 272	98.51 161	97.29 28	98.66 60	97.88 195	94.51 85	99.90 33	97.87 42	99.17 109	97.39 216

test1172		98.56 28	98.35 24	99.16 50	99.53 36	97.94 66	99.09 46	98.83 68	96.52 64	99.05 32	99.34 31	95.34 59	99.82 63	97.86 43	99.64 62	99.73 36

XVS		98.70 9	98.49 16	99.34 23	99.70 23	98.35 43	99.29 14	98.88 49	97.40 21	98.46 68	99.20 52	95.90 40	99.89 35	97.85 44	99.74 41	99.78 13

X-MVStestdata		94.06 259	92.30 279	99.34 23	99.70 23	98.35 43	99.29 14	98.88 49	97.40 21	98.46 68	43.50 352	95.90 40	99.89 35	97.85 44	99.74 41	99.78 13

xiu_mvs_v2_base		97.66 75	97.70 63	97.56 158	98.61 142	95.46 171	97.44 261	98.46 171	97.15 40	98.65 61	98.15 175	94.33 91	99.80 79	97.84 46	98.66 131	97.41 214

DeepC-MVS	95.98 3	97.88 64	97.58 67	98.77 75	99.25 86	96.93 104	98.83 94	98.75 101	96.96 49	96.89 148	99.50 4	90.46 159	99.87 44	97.84 46	99.76 32	99.52 85
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
MSP-MVS		98.74 8	98.55 10	99.29 31	99.75 3	98.23 49	99.26 18	98.88 49	97.52 15	99.41 11	98.78 112	96.00 34	99.79 91	97.79 48	99.59 71	99.85 2

CP-MVS		98.57 26	98.36 22	99.19 43	99.66 27	97.86 68	99.34 11	98.87 55	95.96 85	98.60 64	99.13 64	96.05 32	99.94 3	97.77 49	99.86 1	99.77 20

SteuartSystems-ACMMP		98.90 5	98.75 4	99.36 21	99.22 94	98.43 33	99.10 45	98.87 55	97.38 24	99.35 14	99.40 13	97.78 3	99.87 44	97.77 49	99.85 3	99.78 13
Skip Steuart: Steuart Systems R&D Blog.
APD-MVS_3200maxsize		98.53 34	98.33 32	99.15 53	99.50 41	97.92 67	99.15 35	98.81 76	96.24 73	99.20 22	99.37 22	95.30 62	99.80 79	97.73 51	99.67 54	99.72 40

SR-MVS-dyc-post		98.54 32	98.35 24	99.13 54	99.49 45	97.86 68	99.11 42	98.80 86	96.49 65	99.17 24	99.35 28	95.34 59	99.82 63	97.72 52	99.65 58	99.71 44

RE-MVS-def			98.34 28		99.49 45	97.86 68	99.11 42	98.80 86	96.49 65	99.17 24	99.35 28	95.29 63		97.72 52	99.65 58	99.71 44

xxxxxxxxxxxxxcwj		98.70 9	98.50 14	99.30 30	99.46 51	98.38 35	98.21 194	98.52 158	97.95 3	99.32 15	99.39 14	96.22 20	99.84 52	97.72 52	99.73 43	99.67 61

SF-MVS		98.59 20	98.32 33	99.41 16	99.54 35	98.71 18	99.04 53	98.81 76	95.12 126	99.32 15	99.39 14	96.22 20	99.84 52	97.72 52	99.73 43	99.67 61

LFMVS		95.86 153	94.98 178	98.47 96	98.87 119	96.32 132	98.84 93	96.02 323	93.40 204	98.62 62	99.20 52	74.99 332	99.63 128	97.72 52	97.20 176	99.46 102

SR-MVS		98.57 26	98.35 24	99.24 40	99.53 36	98.18 53	99.09 46	98.82 70	96.58 61	99.10 29	99.32 33	95.39 55	99.82 63	97.70 57	99.63 64	99.72 40

PHI-MVS		98.34 48	98.06 49	99.18 47	99.15 101	98.12 58	99.04 53	99.09 20	93.32 207	98.83 48	99.10 69	96.54 16	99.83 55	97.70 57	99.76 32	99.59 80

HPM-MVS		98.36 45	98.10 48	99.13 54	99.74 7	97.82 72	99.53 1	98.80 86	94.63 150	98.61 63	98.97 87	95.13 70	99.77 100	97.65 59	99.83 9	99.79 10
Chunlin Ren, Qingshan Xu, Shikun Zhang, Jiaqi Yang: Hierarchical Prior Mining for Non-local Multi-View Stereo. ICCV 2023
DPE-MVS		98.92 4	98.67 6	99.65 2	99.58 32	99.20 7	98.42 169	98.91 43	97.58 14	99.54 7	99.46 9	97.10 9	99.94 3	97.64 60	99.84 8	99.83 5

ETV-MVS		97.96 58	97.81 59	98.40 103	98.42 151	97.27 90	98.73 117	98.55 151	96.84 51	98.38 75	97.44 234	95.39 55	99.35 158	97.62 61	98.89 118	98.58 184

CS-MVS		97.81 67	97.61 65	98.41 102	98.52 148	97.15 98	99.09 46	98.55 151	96.18 76	97.61 122	97.20 249	94.59 83	99.39 155	97.62 61	99.10 111	98.70 172

HFP-MVS		98.63 16	98.40 18	99.32 28	99.72 12	98.29 46	99.23 21	98.96 32	96.10 82	98.94 39	99.17 56	96.06 30	99.92 21	97.62 61	99.78 23	99.75 28

ACMMPR		98.59 20	98.36 22	99.29 31	99.74 7	98.15 56	99.23 21	98.95 34	96.10 82	98.93 43	99.19 55	95.70 44	99.94 3	97.62 61	99.79 19	99.78 13

jason		97.32 99	97.08 93	98.06 125	97.45 227	95.59 163	97.87 235	97.91 256	94.79 141	98.55 66	98.83 107	91.12 146	99.23 167	97.58 65	99.60 68	99.34 111
jason: jason.
lupinMVS		97.44 91	97.22 88	98.12 122	98.07 180	95.76 160	97.68 250	97.76 261	94.50 155	98.79 49	98.61 127	92.34 113	99.30 161	97.58 65	99.59 71	99.31 117

HPM-MVS_fast		98.38 43	98.13 46	99.12 57	99.75 3	97.86 68	99.44 4	98.82 70	94.46 157	98.94 39	99.20 52	95.16 69	99.74 106	97.58 65	99.85 3	99.77 20

ZNCC-MVS		98.49 36	98.20 44	99.35 22	99.73 11	98.39 34	99.19 31	98.86 61	95.77 91	98.31 80	99.10 69	95.46 51	99.93 15	97.57 68	99.81 10	99.74 33

region2R		98.61 17	98.38 20	99.29 31	99.74 7	98.16 55	99.23 21	98.93 37	96.15 77	98.94 39	99.17 56	95.91 39	99.94 3	97.55 69	99.79 19	99.78 13

DeepC-MVS_fast	96.70 1	98.55 30	98.34 28	99.18 47	99.25 86	98.04 60	98.50 158	98.78 94	97.72 6	98.92 44	99.28 40	95.27 64	99.82 63	97.55 69	99.77 26	99.69 51
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
HPM-MVS++		98.58 23	98.25 38	99.55 6	99.50 41	99.08 9	98.72 121	98.66 131	97.51 16	98.15 81	98.83 107	95.70 44	99.92 21	97.53 71	99.67 54	99.66 65

nrg030		96.28 139	95.72 143	97.96 131	96.90 263	98.15 56	99.39 5	98.31 196	95.47 105	94.42 212	98.35 155	92.09 123	98.69 228	97.50 72	89.05 298	97.04 226

CSCG		97.85 66	97.74 62	98.20 115	99.67 26	95.16 181	99.22 25	99.32 7	93.04 217	97.02 141	98.92 99	95.36 58	99.91 30	97.43 73	99.64 62	99.52 85

mPP-MVS		98.51 35	98.26 37	99.25 39	99.75 3	98.04 60	99.28 16	98.81 76	96.24 73	98.35 77	99.23 45	95.46 51	99.94 3	97.42 74	99.81 10	99.77 20

mvs_anonymous		96.70 123	96.53 121	97.18 174	98.19 171	93.78 234	98.31 183	98.19 215	94.01 170	94.47 206	98.27 167	92.08 124	98.46 251	97.39 75	97.91 157	99.31 117

EIA-MVS		97.75 70	97.58 67	98.27 109	98.38 153	96.44 126	99.01 60	98.60 139	95.88 87	97.26 130	97.53 228	94.97 74	99.33 160	97.38 76	99.20 107	99.05 149

NCCC		98.61 17	98.35 24	99.38 17	99.28 82	98.61 24	98.45 162	98.76 98	97.82 5	98.45 71	98.93 97	96.65 14	99.83 55	97.38 76	99.41 97	99.71 44

VPA-MVSNet		95.75 158	95.11 172	97.69 148	97.24 238	97.27 90	98.94 74	99.23 12	95.13 125	95.51 184	97.32 240	85.73 252	98.91 207	97.33 78	89.55 291	96.89 241

OPU-MVS				99.37 20	99.24 92	99.05 10	99.02 58				99.16 61	97.81 2	99.37 157	97.24 79	99.73 43	99.70 48

3Dnovator	94.51 5	97.46 87	96.93 100	99.07 60	97.78 197	97.64 76	99.35 10	99.06 22	97.02 47	93.75 244	99.16 61	89.25 177	99.92 21	97.22 80	99.75 38	99.64 70

#test#		98.54 32	98.27 36	99.32 28	99.72 12	98.29 46	98.98 67	98.96 32	95.65 98	98.94 39	99.17 56	96.06 30	99.92 21	97.21 81	99.78 23	99.75 28

ACMMP		98.23 54	97.95 55	99.09 59	99.74 7	97.62 78	99.03 56	99.41 6	95.98 84	97.60 124	99.36 26	94.45 89	99.93 15	97.14 82	98.85 122	99.70 48
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
PVSNet_Blended_VisFu		97.70 73	97.46 77	98.44 98	99.27 83	95.91 155	98.63 137	99.16 17	94.48 156	97.67 116	98.88 102	92.80 108	99.91 30	97.11 83	99.12 110	99.50 91

mvs_tets		95.41 175	95.00 176	96.65 206	95.58 315	94.42 216	99.00 62	98.55 151	95.73 93	93.21 262	98.38 152	83.45 289	98.63 234	97.09 84	94.00 230	96.91 238

GST-MVS		98.43 40	98.12 47	99.34 23	99.72 12	98.38 35	99.09 46	98.82 70	95.71 94	98.73 55	99.06 78	95.27 64	99.93 15	97.07 85	99.63 64	99.72 40

9.14			98.06 49		99.47 48		98.71 122	98.82 70	94.36 159	99.16 26	99.29 39	96.05 32	99.81 70	97.00 86	99.71 50

EPNet		97.28 100	96.87 103	98.51 92	94.98 325	96.14 139	98.90 78	97.02 304	98.28 1	95.99 181	99.11 67	91.36 140	99.89 35	96.98 87	99.19 108	99.50 91
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
HyFIR lowres test		96.90 117	96.49 122	98.14 118	99.33 65	95.56 166	97.38 266	99.65 2	92.34 243	97.61 122	98.20 172	89.29 176	99.10 185	96.97 88	97.60 170	99.77 20

3Dnovator+	94.38 6	97.43 92	96.78 107	99.38 17	97.83 195	98.52 27	99.37 7	98.71 113	97.09 45	92.99 270	99.13 64	89.36 174	99.89 35	96.97 88	99.57 75	99.71 44

abl_6		98.30 53	98.03 51	99.13 54	99.56 34	97.76 74	99.13 39	98.82 70	96.14 78	99.26 18	99.37 22	93.33 102	99.93 15	96.96 90	99.67 54	99.69 51

jajsoiax		95.45 171	95.03 175	96.73 200	95.42 322	94.63 206	99.14 36	98.52 158	95.74 92	93.22 261	98.36 154	83.87 285	98.65 233	96.95 91	94.04 228	96.91 238

ET-MVSNet_ETH3D		94.13 252	92.98 267	97.58 156	98.22 167	96.20 136	97.31 275	95.37 330	94.53 152	79.56 340	97.63 221	86.51 238	97.53 320	96.91 92	90.74 276	99.02 151

MVSFormer		97.57 83	97.49 75	97.84 135	98.07 180	95.76 160	99.47 2	98.40 182	94.98 133	98.79 49	98.83 107	92.34 113	98.41 264	96.91 92	99.59 71	99.34 111

test_djsdf		96.00 147	95.69 148	96.93 191	95.72 311	95.49 170	99.47 2	98.40 182	94.98 133	94.58 202	97.86 197	89.16 180	98.41 264	96.91 92	94.12 227	96.88 242

test_prior3		98.22 55	97.90 58	99.19 43	99.31 70	98.22 50	97.80 241	98.84 65	96.12 80	97.89 105	98.69 119	95.96 36	99.70 114	96.89 95	99.60 68	99.65 67

test_prior2							97.80 241		96.12 80	97.89 105	98.69 119	95.96 36		96.89 95	99.60 68

EPP-MVSNet		97.46 87	97.28 85	97.99 128	98.64 139	95.38 173	99.33 13	98.31 196	93.61 197	97.19 132	99.07 77	94.05 95	99.23 167	96.89 95	98.43 143	99.37 110

PS-MVSNAJss		96.43 132	96.26 129	96.92 193	95.84 309	95.08 186	99.16 34	98.50 166	95.87 88	93.84 240	98.34 159	94.51 85	98.61 235	96.88 98	93.45 243	97.06 225

PVSNet_BlendedMVS		96.73 122	96.60 117	97.12 178	99.25 86	95.35 176	98.26 191	99.26 8	94.28 160	97.94 100	97.46 231	92.74 109	99.81 70	96.88 98	93.32 246	96.20 309

PVSNet_Blended		97.38 96	97.12 90	98.14 118	99.25 86	95.35 176	97.28 277	99.26 8	93.13 215	97.94 100	98.21 171	92.74 109	99.81 70	96.88 98	99.40 99	99.27 124

Effi-MVS+		97.12 109	96.69 113	98.39 104	98.19 171	96.72 113	97.37 268	98.43 178	93.71 187	97.65 119	98.02 183	92.20 120	99.25 164	96.87 101	97.79 162	99.19 132

CHOSEN 1792x2688		97.12 109	96.80 104	98.08 123	99.30 75	94.56 213	98.05 217	99.71 1	93.57 198	97.09 135	98.91 100	88.17 206	99.89 35	96.87 101	99.56 80	99.81 8

test_yl		97.22 102	96.78 107	98.54 89	98.73 128	96.60 118	98.45 162	98.31 196	94.70 143	98.02 90	98.42 147	90.80 153	99.70 114	96.81 103	96.79 183	99.34 111

DCV-MVSNet		97.22 102	96.78 107	98.54 89	98.73 128	96.60 118	98.45 162	98.31 196	94.70 143	98.02 90	98.42 147	90.80 153	99.70 114	96.81 103	96.79 183	99.34 111

ETH3D-3000-0.1		98.35 46	98.00 53	99.38 17	99.47 48	98.68 21	98.67 132	98.84 65	94.66 149	99.11 28	99.25 43	95.46 51	99.81 70	96.80 105	99.73 43	99.63 73

PGM-MVS		98.49 36	98.23 42	99.27 38	99.72 12	98.08 59	98.99 64	99.49 5	95.43 107	99.03 33	99.32 33	95.56 47	99.94 3	96.80 105	99.77 26	99.78 13

RRT_test8_iter05		94.56 226	94.19 214	95.67 266	97.60 209	91.34 283	98.93 75	98.42 179	94.75 142	93.39 256	97.87 196	79.00 312	98.61 235	96.78 107	90.99 274	97.07 224

XVG-OURS-SEG-HR		96.51 130	96.34 125	97.02 183	98.77 126	93.76 235	97.79 243	98.50 166	95.45 106	96.94 143	99.09 74	87.87 216	99.55 141	96.76 108	95.83 212	97.74 207

MP-MVS		98.33 50	98.01 52	99.28 35	99.75 3	98.18 53	99.22 25	98.79 91	96.13 79	97.92 103	99.23 45	94.54 84	99.94 3	96.74 109	99.78 23	99.73 36
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
agg_prior1		97.95 61	97.51 74	99.28 35	99.30 75	98.38 35	97.81 240	98.72 109	93.16 214	97.57 125	98.66 124	96.14 26	99.81 70	96.63 110	99.56 80	99.66 65

train_agg		97.97 57	97.52 72	99.33 27	99.31 70	98.50 29	97.92 227	98.73 107	92.98 220	97.74 111	98.68 121	96.20 23	99.80 79	96.59 111	99.57 75	99.68 57

MVSTER		96.06 144	95.72 143	97.08 181	98.23 166	95.93 153	98.73 117	98.27 205	94.86 139	95.07 189	98.09 179	88.21 204	98.54 244	96.59 111	93.46 241	96.79 252

UGNet		96.78 121	96.30 127	98.19 117	98.24 165	95.89 157	98.88 85	98.93 37	97.39 23	96.81 152	97.84 200	82.60 291	99.90 33	96.53 113	99.49 88	98.79 167
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
APD-MVS		98.35 46	98.00 53	99.42 15	99.51 39	98.72 17	98.80 104	98.82 70	94.52 154	99.23 20	99.25 43	95.54 49	99.80 79	96.52 114	99.77 26	99.74 33
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
VPNet		94.99 200	94.19 214	97.40 166	97.16 247	96.57 120	98.71 122	98.97 30	95.67 96	94.84 195	98.24 170	80.36 305	98.67 232	96.46 115	87.32 317	96.96 230

ETH3D cwj APD-0.16		97.96 58	97.52 72	99.29 31	99.05 105	98.52 27	98.33 177	98.68 120	93.18 212	98.68 57	99.13 64	94.62 81	99.83 55	96.45 116	99.55 83	99.52 85

sss		97.39 95	96.98 99	98.61 83	98.60 143	96.61 117	98.22 193	98.93 37	93.97 173	98.01 94	98.48 141	91.98 126	99.85 49	96.45 116	98.15 151	99.39 108

MVS_Test		97.28 100	97.00 97	98.13 120	98.33 160	95.97 147	98.74 113	98.07 241	94.27 161	98.44 72	98.07 180	92.48 111	99.26 163	96.43 118	98.19 150	99.16 137

FIs		96.51 130	96.12 133	97.67 150	97.13 249	97.54 81	99.36 8	99.22 14	95.89 86	94.03 232	98.35 155	91.98 126	98.44 254	96.40 119	92.76 253	97.01 227

test9_res														96.39 120	99.57 75	99.69 51

Anonymous20240529		95.10 194	94.22 212	97.75 142	99.01 108	94.26 223	98.87 87	98.83 68	85.79 329	96.64 157	98.97 87	78.73 313	99.85 49	96.27 121	94.89 216	99.12 142

PMMVS		96.60 125	96.33 126	97.41 164	97.90 191	93.93 230	97.35 271	98.41 180	92.84 227	97.76 109	97.45 233	91.10 148	99.20 170	96.26 122	97.91 157	99.11 143

CLD-MVS		95.62 165	95.34 160	96.46 230	97.52 220	93.75 237	97.27 278	98.46 171	95.53 102	94.42 212	98.00 186	86.21 245	98.97 197	96.25 123	94.37 217	96.66 271
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
Anonymous202405211		95.28 184	94.49 198	97.67 150	99.00 109	93.75 237	98.70 126	97.04 301	90.66 289	96.49 168	98.80 110	78.13 316	99.83 55	96.21 124	95.36 215	99.44 105

RRT_MVS		96.04 145	95.53 151	97.56 158	97.07 253	97.32 87	98.57 148	98.09 237	95.15 124	95.02 191	98.44 144	88.20 205	98.58 242	96.17 125	93.09 250	96.79 252

ZD-MVS					99.46 51	98.70 19		98.79 91	93.21 211	98.67 58	98.97 87	95.70 44	99.83 55	96.07 126	99.58 74

HQP_MVS		96.14 142	95.90 139	96.85 195	97.42 228	94.60 211	98.80 104	98.56 149	97.28 29	95.34 185	98.28 164	87.09 229	99.03 193	96.07 126	94.27 219	96.92 233

plane_prior5								98.56 149					99.03 193	96.07 126	94.27 219	96.92 233

CPTT-MVS		97.72 72	97.32 84	98.92 69	99.64 28	97.10 99	99.12 41	98.81 76	92.34 243	98.09 84	99.08 76	93.01 106	99.92 21	96.06 129	99.77 26	99.75 28

DP-MVS Recon		97.86 65	97.46 77	99.06 61	99.53 36	98.35 43	98.33 177	98.89 46	92.62 232	98.05 86	98.94 96	95.34 59	99.65 123	96.04 130	99.42 96	99.19 132

FC-MVSNet-test		96.42 133	96.05 134	97.53 160	96.95 258	97.27 90	99.36 8	99.23 12	95.83 89	93.93 234	98.37 153	92.00 125	98.32 273	96.02 131	92.72 254	97.00 228

Vis-MVSNet		97.42 93	97.11 91	98.34 106	98.66 137	96.23 135	99.22 25	99.00 27	96.63 60	98.04 88	99.21 48	88.05 211	99.35 158	96.01 132	99.21 106	99.45 104
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
ab-mvs		96.42 133	95.71 146	98.55 87	98.63 140	96.75 112	97.88 234	98.74 102	93.84 179	96.54 165	98.18 174	85.34 259	99.75 104	95.93 133	96.35 196	99.15 138

WTY-MVS		97.37 97	96.92 101	98.72 77	98.86 120	96.89 108	98.31 183	98.71 113	95.26 118	97.67 116	98.56 135	92.21 119	99.78 95	95.89 134	96.85 181	99.48 96

XVG-OURS		96.55 129	96.41 123	96.99 184	98.75 127	93.76 235	97.50 260	98.52 158	95.67 96	96.83 149	99.30 38	88.95 190	99.53 142	95.88 135	96.26 203	97.69 210

agg_prior2														95.87 136	99.57 75	99.68 57

UniMVSNet_NR-MVSNet		95.71 160	95.15 169	97.40 166	96.84 266	96.97 102	98.74 113	99.24 10	95.16 123	93.88 237	97.72 211	91.68 131	98.31 275	95.81 137	87.25 319	96.92 233

DU-MVS		95.42 173	94.76 186	97.40 166	96.53 281	96.97 102	98.66 135	98.99 29	95.43 107	93.88 237	97.69 213	88.57 196	98.31 275	95.81 137	87.25 319	96.92 233

testtj		98.33 50	97.95 55	99.47 11	99.49 45	98.70 19	98.83 94	98.86 61	95.48 104	98.91 45	99.17 56	95.48 50	99.93 15	95.80 139	99.53 85	99.76 26

UniMVSNet (Re)		95.78 157	95.19 168	97.58 156	96.99 257	97.47 83	98.79 108	99.18 16	95.60 99	93.92 235	97.04 266	91.68 131	98.48 248	95.80 139	87.66 313	96.79 252

cascas		94.63 221	93.86 237	96.93 191	96.91 262	94.27 222	96.00 325	98.51 161	85.55 330	94.54 203	96.23 307	84.20 278	98.87 214	95.80 139	96.98 180	97.66 211

Effi-MVS+-dtu		96.29 137	96.56 118	95.51 269	97.89 192	90.22 302	98.80 104	98.10 234	96.57 62	96.45 171	96.66 291	90.81 151	98.91 207	95.72 142	97.99 155	97.40 215

mvs-test1		96.60 125	96.68 115	96.37 235	97.89 192	91.81 275	98.56 149	98.10 234	96.57 62	96.52 167	97.94 190	90.81 151	99.45 152	95.72 142	98.01 154	97.86 204

LPG-MVS_test		95.62 165	95.34 160	96.47 227	97.46 223	93.54 244	98.99 64	98.54 154	94.67 147	94.36 214	98.77 114	85.39 256	99.11 182	95.71 144	94.15 225	96.76 256

LGP-MVS_train				96.47 227	97.46 223	93.54 244		98.54 154	94.67 147	94.36 214	98.77 114	85.39 256	99.11 182	95.71 144	94.15 225	96.76 256

旧先验2							97.57 258		91.30 277	98.67 58			99.80 79	95.70 146

LCM-MVSNet-Re		95.22 187	95.32 163	94.91 287	98.18 173	87.85 330	98.75 110	95.66 329	95.11 127	88.96 319	96.85 284	90.26 164	97.65 315	95.65 147	98.44 141	99.22 128

anonymousdsp		95.42 173	94.91 181	96.94 190	95.10 324	95.90 156	99.14 36	98.41 180	93.75 182	93.16 263	97.46 231	87.50 224	98.41 264	95.63 148	94.03 229	96.50 295

CDPH-MVS		97.94 62	97.49 75	99.28 35	99.47 48	98.44 31	97.91 229	98.67 128	92.57 235	98.77 51	98.85 104	95.93 38	99.72 108	95.56 149	99.69 52	99.68 57

CostFormer		94.95 204	94.73 188	95.60 268	97.28 236	89.06 316	97.53 259	96.89 312	89.66 305	96.82 151	96.72 289	86.05 248	98.95 204	95.53 150	96.13 208	98.79 167

ACMM	93.85 9	95.69 162	95.38 158	96.61 211	97.61 208	93.84 233	98.91 77	98.44 175	95.25 119	94.28 218	98.47 142	86.04 250	99.12 179	95.50 151	93.95 232	96.87 244
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
ACMP	93.49 10	95.34 181	94.98 178	96.43 232	97.67 204	93.48 247	98.73 117	98.44 175	94.94 138	92.53 283	98.53 136	84.50 272	99.14 177	95.48 152	94.00 230	96.66 271
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
testing_2		90.61 300	88.50 306	96.95 189	90.08 345	95.57 165	97.69 249	98.06 245	93.02 218	76.55 341	92.48 337	61.18 348	98.44 254	95.45 153	91.98 260	96.84 248

tttt0517		96.07 143	95.51 153	97.78 139	98.41 152	94.84 197	99.28 16	94.33 341	94.26 162	97.64 120	98.64 126	84.05 280	99.47 150	95.34 154	97.60 170	99.03 150

TAMVS		97.02 112	96.79 106	97.70 147	98.06 182	95.31 178	98.52 153	98.31 196	93.95 174	97.05 140	98.61 127	93.49 101	98.52 246	95.33 155	97.81 161	99.29 122

BP-MVS														95.30 156

HQP-MVS		95.72 159	95.40 154	96.69 204	97.20 242	94.25 224	98.05 217	98.46 171	96.43 67	94.45 207	97.73 209	86.75 235	98.96 200	95.30 156	94.18 223	96.86 246

thisisatest0530		96.01 146	95.36 159	97.97 129	98.38 153	95.52 169	98.88 85	94.19 343	94.04 167	97.64 120	98.31 162	83.82 287	99.46 151	95.29 158	97.70 167	98.93 160

WR-MVS		95.15 191	94.46 201	97.22 171	96.67 276	96.45 125	98.21 194	98.81 76	94.15 163	93.16 263	97.69 213	87.51 222	98.30 277	95.29 158	88.62 304	96.90 240

tpmrst		95.63 164	95.69 148	95.44 273	97.54 217	88.54 323	96.97 294	97.56 270	93.50 200	97.52 127	96.93 279	89.49 170	99.16 173	95.25 160	96.42 195	98.64 180

CDS-MVSNet		96.99 113	96.69 113	97.90 133	98.05 183	95.98 142	98.20 197	98.33 193	93.67 194	96.95 142	98.49 140	93.54 100	98.42 257	95.24 161	97.74 165	99.31 117
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
OPM-MVS		95.69 162	95.33 162	96.76 199	96.16 298	94.63 206	98.43 167	98.39 184	96.64 59	95.02 191	98.78 112	85.15 261	99.05 189	95.21 162	94.20 222	96.60 276

OMC-MVS		97.55 85	97.34 83	98.20 115	99.33 65	95.92 154	98.28 188	98.59 141	95.52 103	97.97 97	99.10 69	93.28 104	99.49 145	95.09 163	98.88 119	99.19 132

UniMVSNet_ETH3D		94.24 245	93.33 261	96.97 187	97.19 245	93.38 252	98.74 113	98.57 147	91.21 283	93.81 241	98.58 132	72.85 339	98.77 225	95.05 164	93.93 233	98.77 169

CANet_DTU		96.96 114	96.55 119	98.21 114	98.17 175	96.07 141	97.98 224	98.21 212	97.24 35	97.13 134	98.93 97	86.88 234	99.91 30	95.00 165	99.37 101	98.66 178

UA-Net		97.96 58	97.62 64	98.98 65	98.86 120	97.47 83	98.89 82	99.08 21	96.67 58	98.72 56	99.54 1	93.15 105	99.81 70	94.87 166	98.83 123	99.65 67

114514_t		96.93 115	96.27 128	98.92 69	99.50 41	97.63 77	98.85 90	98.90 44	84.80 332	97.77 108	99.11 67	92.84 107	99.66 122	94.85 167	99.77 26	99.47 98

Anonymous20231211		94.10 255	93.26 264	96.61 211	99.11 104	94.28 221	99.01 60	98.88 49	86.43 323	92.81 273	97.57 225	81.66 296	98.68 231	94.83 168	89.02 300	96.88 242

XXY-MVS		95.20 189	94.45 203	97.46 161	96.75 271	96.56 121	98.86 89	98.65 135	93.30 209	93.27 260	98.27 167	84.85 266	98.87 214	94.82 169	91.26 270	96.96 230

MG-MVS		97.81 67	97.60 66	98.44 98	99.12 103	95.97 147	97.75 245	98.78 94	96.89 50	98.46 68	99.22 47	93.90 99	99.68 120	94.81 170	99.52 87	99.67 61

EI-MVSNet		95.96 148	95.83 141	96.36 236	97.93 189	93.70 241	98.12 211	98.27 205	93.70 189	95.07 189	99.02 80	92.23 118	98.54 244	94.68 171	93.46 241	96.84 248

thisisatest0515		95.61 167	94.89 182	97.76 141	98.15 176	95.15 183	96.77 310	94.41 339	92.95 222	97.18 133	97.43 235	84.78 267	99.45 152	94.63 172	97.73 166	98.68 175

IterMVS-LS		95.46 169	95.21 167	96.22 244	98.12 177	93.72 240	98.32 182	98.13 229	93.71 187	94.26 219	97.31 241	92.24 117	98.10 291	94.63 172	90.12 282	96.84 248
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
1314		96.25 141	95.73 142	97.79 138	97.13 249	95.55 168	98.19 201	98.59 141	93.47 201	92.03 295	97.82 204	91.33 142	99.49 145	94.62 174	98.44 141	98.32 194

baseline1		95.84 154	95.12 171	98.01 127	98.49 149	95.98 142	98.73 117	97.03 302	95.37 112	96.22 175	98.19 173	89.96 167	99.16 173	94.60 175	87.48 314	98.90 162

IS-MVSNet		97.22 102	96.88 102	98.25 112	98.85 122	96.36 130	99.19 31	97.97 251	95.39 109	97.23 131	98.99 86	91.11 147	98.93 205	94.60 175	98.59 133	99.47 98

NR-MVSNet		94.98 202	94.16 217	97.44 162	96.53 281	97.22 95	98.74 113	98.95 34	94.96 135	89.25 318	97.69 213	89.32 175	98.18 285	94.59 177	87.40 316	96.92 233

IB-MVS	91.98 17	93.27 272	91.97 283	97.19 173	97.47 222	93.41 250	97.09 289	95.99 324	93.32 207	92.47 286	95.73 317	78.06 317	99.53 142	94.59 177	82.98 333	98.62 181
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
DWT-MVSNet_test		94.82 209	94.36 208	96.20 245	97.35 233	90.79 294	98.34 176	96.57 322	92.91 224	95.33 187	96.44 301	82.00 293	99.12 179	94.52 179	95.78 213	98.70 172

HY-MVS	93.96 8	96.82 120	96.23 131	98.57 85	98.46 150	97.00 101	98.14 208	98.21 212	93.95 174	96.72 155	97.99 187	91.58 133	99.76 102	94.51 180	96.54 191	98.95 159

D2MVS		95.18 190	95.08 173	95.48 270	97.10 251	92.07 271	98.30 185	99.13 19	94.02 169	92.90 271	96.73 288	89.48 171	98.73 227	94.48 181	93.60 240	95.65 321

test_part1		92.87 280	91.72 286	96.32 240	97.55 216	93.50 246	99.04 53	98.74 102	83.31 335	90.81 307	97.70 212	76.61 325	98.60 239	94.43 182	87.30 318	96.85 247

Baseline_NR-MVSNet		94.35 238	93.81 239	95.96 254	96.20 294	94.05 228	98.61 140	96.67 320	91.44 270	93.85 239	97.60 222	88.57 196	98.14 288	94.39 183	86.93 322	95.68 320

AdaColmap		97.15 108	96.70 112	98.48 95	99.16 99	96.69 114	98.01 221	98.89 46	94.44 158	96.83 149	98.68 121	90.69 156	99.76 102	94.36 184	99.29 105	98.98 155

1112_ss		96.63 124	96.00 137	98.50 93	98.56 144	96.37 129	98.18 205	98.10 234	92.92 223	94.84 195	98.43 145	92.14 121	99.58 133	94.35 185	96.51 192	99.56 84

CP-MVSNet		94.94 206	94.30 210	96.83 196	96.72 273	95.56 166	99.11 42	98.95 34	93.89 176	92.42 288	97.90 193	87.19 228	98.12 290	94.32 186	88.21 307	96.82 251

CNLPA		97.45 90	97.03 95	98.73 76	99.05 105	97.44 85	98.07 215	98.53 156	95.32 115	96.80 153	98.53 136	93.32 103	99.72 108	94.31 187	99.31 104	99.02 151

testdata				98.26 111	99.20 97	95.36 174		98.68 120	91.89 257	98.60 64	99.10 69	94.44 90	99.82 63	94.27 188	99.44 95	99.58 82

PVSNet	91.96 18	96.35 135	96.15 132	96.96 188	99.17 98	92.05 272	96.08 321	98.68 120	93.69 190	97.75 110	97.80 206	88.86 191	99.69 119	94.26 189	99.01 113	99.15 138

miper_enhance_ethall		95.10 194	94.75 187	96.12 249	97.53 219	93.73 239	96.61 316	98.08 239	92.20 251	93.89 236	96.65 293	92.44 112	98.30 277	94.21 190	91.16 271	96.34 303

Test_1112_low_res		96.34 136	95.66 150	98.36 105	98.56 144	95.94 150	97.71 247	98.07 241	92.10 252	94.79 199	97.29 242	91.75 130	99.56 136	94.17 191	96.50 193	99.58 82

TranMVSNet+NR-MVSNet		95.14 192	94.48 199	97.11 179	96.45 286	96.36 130	99.03 56	99.03 25	95.04 131	93.58 247	97.93 191	88.27 203	98.03 298	94.13 192	86.90 324	96.95 232

API-MVS		97.41 94	97.25 86	97.91 132	98.70 133	96.80 109	98.82 97	98.69 117	94.53 152	98.11 83	98.28 164	94.50 88	99.57 134	94.12 193	99.49 88	97.37 218

ETH3 D test6400		97.59 81	97.01 96	99.34 23	99.40 59	98.56 25	98.20 197	98.81 76	91.63 265	98.44 72	98.85 104	93.98 98	99.82 63	94.11 194	99.69 52	99.64 70

cl-mvsnet2		94.68 216	94.19 214	96.13 248	98.11 178	93.60 242	96.94 296	98.31 196	92.43 240	93.32 259	96.87 283	86.51 238	98.28 281	94.10 195	91.16 271	96.51 293

PLC	95.07 4	97.20 105	96.78 107	98.44 98	99.29 78	96.31 134	98.14 208	98.76 98	92.41 241	96.39 172	98.31 162	94.92 76	99.78 95	94.06 196	98.77 126	99.23 127
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
XVG-ACMP-BASELINE		94.54 228	94.14 219	95.75 264	96.55 280	91.65 281	98.11 213	98.44 175	94.96 135	94.22 222	97.90 193	79.18 311	99.11 182	94.05 197	93.85 234	96.48 297

F-COLMAP		97.09 111	96.80 104	97.97 129	99.45 55	94.95 194	98.55 151	98.62 138	93.02 218	96.17 177	98.58 132	94.01 96	99.81 70	93.95 198	98.90 117	99.14 140

MDTV_nov1_ep13_2view						84.26 338	96.89 304		90.97 287	97.90 104		89.89 168		93.91 199		99.18 136

baseline2		95.11 193	94.52 197	96.87 194	96.65 277	93.56 243	98.27 190	94.10 345	93.45 202	92.02 296	97.43 235	87.45 226	99.19 171	93.88 200	97.41 174	97.87 203

原ACMM1				98.65 81	99.32 68	96.62 115		98.67 128	93.27 210	97.81 107	98.97 87	95.18 68	99.83 55	93.84 201	99.46 93	99.50 91

RPSCF		94.87 208	95.40 154	93.26 314	98.89 117	82.06 344	98.33 177	98.06 245	90.30 296	96.56 161	99.26 42	87.09 229	99.49 145	93.82 202	96.32 198	98.24 195

PAPM_NR		97.46 87	97.11 91	98.50 93	99.50 41	96.41 128	98.63 137	98.60 139	95.18 122	97.06 139	98.06 181	94.26 93	99.57 134	93.80 203	98.87 121	99.52 85

ACMH	92.88 16	94.55 227	93.95 231	96.34 238	97.63 207	93.26 256	98.81 103	98.49 170	93.43 203	89.74 314	98.53 136	81.91 294	99.08 187	93.69 204	93.30 247	96.70 265
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
miper_ehance_all_eth		95.01 198	94.69 190	95.97 253	97.70 203	93.31 254	97.02 292	98.07 241	92.23 248	93.51 252	96.96 275	91.85 128	98.15 287	93.68 205	91.16 271	96.44 300

MAR-MVS		96.91 116	96.40 124	98.45 97	98.69 135	96.90 106	98.66 135	98.68 120	92.40 242	97.07 138	97.96 188	91.54 137	99.75 104	93.68 205	98.92 116	98.69 174
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
Vis-MVSNet (Re-imp)		96.87 118	96.55 119	97.83 136	98.73 128	95.46 171	99.20 29	98.30 202	94.96 135	96.60 160	98.87 103	90.05 165	98.59 240	93.67 207	98.60 132	99.46 102

LS3D		97.16 107	96.66 116	98.68 79	98.53 147	97.19 96	98.93 75	98.90 44	92.83 228	95.99 181	99.37 22	92.12 122	99.87 44	93.67 207	99.57 75	98.97 156

PS-CasMVS		94.67 219	93.99 229	96.71 201	96.68 275	95.26 179	99.13 39	99.03 25	93.68 192	92.33 289	97.95 189	85.35 258	98.10 291	93.59 209	88.16 309	96.79 252

cl_fuxian		94.79 211	94.43 205	95.89 258	97.75 198	93.12 261	97.16 286	98.03 248	92.23 248	93.46 255	97.05 265	91.39 139	98.01 299	93.58 210	89.21 296	96.53 287

CVMVSNet		95.43 172	96.04 135	93.57 310	97.93 189	83.62 339	98.12 211	98.59 141	95.68 95	96.56 161	99.02 80	87.51 222	97.51 321	93.56 211	97.44 172	99.60 78

OurMVSNet-221017-0		94.21 246	94.00 227	94.85 290	95.60 314	89.22 314	98.89 82	97.43 285	95.29 116	92.18 292	98.52 139	82.86 290	98.59 240	93.46 212	91.76 263	96.74 258

eth_miper_zixun_eth		94.68 216	94.41 206	95.47 271	97.64 206	91.71 280	96.73 313	98.07 241	92.71 230	93.64 245	97.21 248	90.54 158	98.17 286	93.38 213	89.76 286	96.54 285

OpenMVS	93.04 13	95.83 155	95.00 176	98.32 107	97.18 246	97.32 87	99.21 28	98.97 30	89.96 301	91.14 303	99.05 79	86.64 237	99.92 21	93.38 213	99.47 90	97.73 208

无先验							97.58 257	98.72 109	91.38 271				99.87 44	93.36 215		99.60 78

1121		97.37 97	96.77 111	99.16 50	99.34 62	97.99 65	98.19 201	98.68 120	90.14 299	98.01 94	98.97 87	94.80 79	99.87 44	93.36 215	99.46 93	99.61 75

gm-plane-assit					95.88 307	87.47 331			89.74 304		96.94 278		99.19 171	93.32 217

WR-MVS_H		95.05 197	94.46 201	96.81 197	96.86 265	95.82 158	99.24 20	99.24 10	93.87 178	92.53 283	96.84 285	90.37 160	98.24 283	93.24 218	87.93 310	96.38 302

tpm		94.13 252	93.80 240	95.12 281	96.50 283	87.91 329	97.44 261	95.89 328	92.62 232	96.37 173	96.30 304	84.13 279	98.30 277	93.24 218	91.66 265	99.14 140

Fast-Effi-MVS+-dtu		95.87 152	95.85 140	95.91 256	97.74 201	91.74 279	98.69 128	98.15 226	95.56 101	94.92 193	97.68 216	88.98 188	98.79 223	93.19 220	97.78 163	97.20 222

pmmvs5		93.65 266	92.97 268	95.68 265	95.49 318	92.37 267	98.20 197	97.28 292	89.66 305	92.58 281	97.26 243	82.14 292	98.09 293	93.18 221	90.95 275	96.58 278

TESTMET0.1,1		94.18 250	93.69 249	95.63 267	96.92 260	89.12 315	96.91 299	94.78 336	93.17 213	94.88 194	96.45 300	78.52 314	98.92 206	93.09 222	98.50 138	98.85 163

test-LLR		95.10 194	94.87 183	95.80 261	96.77 268	89.70 306	96.91 299	95.21 331	95.11 127	94.83 197	95.72 319	87.71 218	98.97 197	93.06 223	98.50 138	98.72 170

test-mter		94.08 257	93.51 256	95.80 261	96.77 268	89.70 306	96.91 299	95.21 331	92.89 225	94.83 197	95.72 319	77.69 319	98.97 197	93.06 223	98.50 138	98.72 170

BH-untuned		95.95 149	95.72 143	96.65 206	98.55 146	92.26 268	98.23 192	97.79 260	93.73 185	94.62 201	98.01 185	88.97 189	99.00 196	93.04 225	98.51 137	98.68 175

EPMVS		94.99 200	94.48 199	96.52 223	97.22 240	91.75 278	97.23 279	91.66 349	94.11 164	97.28 129	96.81 286	85.70 253	98.84 217	93.04 225	97.28 175	98.97 156

pmmvs4		94.69 214	93.99 229	96.81 197	95.74 310	95.94 150	97.40 264	97.67 265	90.42 294	93.37 257	97.59 223	89.08 183	98.20 284	92.97 227	91.67 264	96.30 307

v2v482		94.69 214	94.03 223	96.65 206	96.17 296	94.79 202	98.67 132	98.08 239	92.72 229	94.00 233	97.16 251	87.69 221	98.45 252	92.91 228	88.87 302	96.72 261

Fast-Effi-MVS+		96.28 139	95.70 147	98.03 126	98.29 164	95.97 147	98.58 143	98.25 210	91.74 260	95.29 188	97.23 246	91.03 150	99.15 176	92.90 229	97.96 156	98.97 156

V42		94.78 212	94.14 219	96.70 203	96.33 291	95.22 180	98.97 68	98.09 237	92.32 245	94.31 217	97.06 263	88.39 201	98.55 243	92.90 229	88.87 302	96.34 303

DP-MVS		96.59 127	95.93 138	98.57 85	99.34 62	96.19 138	98.70 126	98.39 184	89.45 308	94.52 204	99.35 28	91.85 128	99.85 49	92.89 231	98.88 119	99.68 57

TDRefinement		91.06 295	89.68 300	95.21 278	85.35 348	91.49 282	98.51 157	97.07 299	91.47 268	88.83 320	97.84 200	77.31 323	99.09 186	92.79 232	77.98 341	95.04 327

ACMH+	92.99 14	94.30 241	93.77 243	95.88 259	97.81 196	92.04 273	98.71 122	98.37 187	93.99 172	90.60 310	98.47 142	80.86 302	99.05 189	92.75 233	92.40 256	96.55 284

cl-mvsnet_		94.51 230	94.01 226	96.02 250	97.58 211	93.40 251	97.05 290	97.96 253	91.73 262	92.76 275	97.08 259	89.06 184	98.13 289	92.61 234	90.29 281	96.52 290

cl-mvsnet1		94.52 229	94.03 223	95.99 251	97.57 215	93.38 252	97.05 290	97.94 254	91.74 260	92.81 273	97.10 253	89.12 181	98.07 295	92.60 235	90.30 280	96.53 287

DPM-MVS		97.55 85	96.99 98	99.23 42	99.04 107	98.55 26	97.17 285	98.35 190	94.85 140	97.93 102	98.58 132	95.07 72	99.71 113	92.60 235	99.34 102	99.43 106

test_post1							96.68 314				30.43 356	87.85 217	98.69 228	92.59 237

SCA		95.46 169	95.13 170	96.46 230	97.67 204	91.29 287	97.33 273	97.60 268	94.68 146	96.92 146	97.10 253	83.97 282	98.89 211	92.59 237	98.32 148	99.20 129

v148		94.29 242	93.76 245	95.91 256	96.10 299	92.93 263	98.58 143	97.97 251	92.59 234	93.47 254	96.95 277	88.53 199	98.32 273	92.56 239	87.06 321	96.49 296

PEN-MVS		94.42 235	93.73 247	96.49 225	96.28 292	94.84 197	99.17 33	99.00 27	93.51 199	92.23 291	97.83 203	86.10 247	97.90 307	92.55 240	86.92 323	96.74 258

Patchmatch-RL test		91.49 291	90.85 292	93.41 311	91.37 341	84.40 337	92.81 342	95.93 327	91.87 258	87.25 325	94.87 327	88.99 185	96.53 335	92.54 241	82.00 335	99.30 120

miper_lstm_enhance		94.33 239	94.07 222	95.11 282	97.75 198	90.97 291	97.22 280	98.03 248	91.67 264	92.76 275	96.97 273	90.03 166	97.78 313	92.51 242	89.64 288	96.56 282

IterMVS		94.09 256	93.85 238	94.80 293	97.99 186	90.35 301	97.18 283	98.12 230	93.68 192	92.46 287	97.34 238	84.05 280	97.41 322	92.51 242	91.33 267	96.62 274
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
IterMVS-SCA-FT		94.11 254	93.87 236	94.85 290	97.98 188	90.56 299	97.18 283	98.11 232	93.75 182	92.58 281	97.48 230	83.97 282	97.41 322	92.48 244	91.30 268	96.58 278

tpm2		94.19 248	93.76 245	95.46 272	97.23 239	89.04 317	97.31 275	96.85 315	87.08 320	96.21 176	96.79 287	83.75 288	98.74 226	92.43 245	96.23 205	98.59 182

PVSNet_0	88.72 19	91.28 293	90.03 298	95.00 285	97.99 186	87.29 333	94.84 335	98.50 166	92.06 253	89.86 313	95.19 324	79.81 307	99.39 155	92.27 246	69.79 346	98.33 193

gg-mvs-nofinetune		92.21 287	90.58 294	97.13 177	96.75 271	95.09 185	95.85 326	89.40 352	85.43 331	94.50 205	81.98 346	80.80 303	98.40 270	92.16 247	98.33 147	97.88 202

pm-mvs1		93.94 262	93.06 266	96.59 214	96.49 284	95.16 181	98.95 72	98.03 248	92.32 245	91.08 304	97.84 200	84.54 271	98.41 264	92.16 247	86.13 330	96.19 310

K. test v3		92.55 284	91.91 285	94.48 300	95.64 313	89.24 313	99.07 50	94.88 335	94.04 167	86.78 327	97.59 223	77.64 322	97.64 316	92.08 249	89.43 293	96.57 280

GBi-Net		94.49 231	93.80 240	96.56 218	98.21 168	95.00 188	98.82 97	98.18 218	92.46 236	94.09 228	97.07 260	81.16 297	97.95 303	92.08 249	92.14 257	96.72 261

test1		94.49 231	93.80 240	96.56 218	98.21 168	95.00 188	98.82 97	98.18 218	92.46 236	94.09 228	97.07 260	81.16 297	97.95 303	92.08 249	92.14 257	96.72 261

FMVSNet3		94.97 203	94.26 211	97.11 179	98.18 173	96.62 115	98.56 149	98.26 209	93.67 194	94.09 228	97.10 253	84.25 275	98.01 299	92.08 249	92.14 257	96.70 265

PatchmatchNet		95.71 160	95.52 152	96.29 242	97.58 211	90.72 296	96.84 308	97.52 276	94.06 166	97.08 136	96.96 275	89.24 178	98.90 210	92.03 253	98.37 144	99.26 125
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
QAPM		96.29 137	95.40 154	98.96 67	97.85 194	97.60 79	99.23 21	98.93 37	89.76 303	93.11 267	99.02 80	89.11 182	99.93 15	91.99 254	99.62 66	99.34 111

新几何1				99.16 50	99.34 62	98.01 62		98.69 117	90.06 300	98.13 82	98.95 95	94.60 82	99.89 35	91.97 255	99.47 90	99.59 80

MDTV_nov1_ep13			95.40 154		97.48 221	88.34 325	96.85 307	97.29 291	93.74 184	97.48 128	97.26 243	89.18 179	99.05 189	91.92 256	97.43 173

EU-MVSNet		93.66 264	94.14 219	92.25 319	95.96 305	83.38 340	98.52 153	98.12 230	94.69 145	92.61 280	98.13 177	87.36 227	96.39 337	91.82 257	90.00 284	96.98 229

GA-MVS		94.81 210	94.03 223	97.14 176	97.15 248	93.86 232	96.76 311	97.58 269	94.00 171	94.76 200	97.04 266	80.91 300	98.48 248	91.79 258	96.25 204	99.09 144

PatchMatch-RL		96.59 127	96.03 136	98.27 109	99.31 70	96.51 123	97.91 229	99.06 22	93.72 186	96.92 146	98.06 181	88.50 200	99.65 123	91.77 259	99.00 114	98.66 178

v1144		94.59 224	93.92 232	96.60 213	96.21 293	94.78 203	98.59 141	98.14 228	91.86 259	94.21 223	97.02 268	87.97 212	98.41 264	91.72 260	89.57 289	96.61 275

v8		94.47 233	93.77 243	96.57 217	96.36 289	94.83 199	99.05 52	98.19 215	91.92 256	93.16 263	96.97 273	88.82 193	98.48 248	91.69 261	87.79 311	96.39 301

testdata2													99.89 35	91.65 262

BH-w/o		95.38 176	95.08 173	96.26 243	98.34 159	91.79 276	97.70 248	97.43 285	92.87 226	94.24 221	97.22 247	88.66 194	98.84 217	91.55 263	97.70 167	98.16 197

LF4IMVS		93.14 277	92.79 271	94.20 305	95.88 307	88.67 321	97.66 252	97.07 299	93.81 181	91.71 298	97.65 217	77.96 318	98.81 221	91.47 264	91.92 262	95.12 324

JIA-IIPM		93.35 269	92.49 276	95.92 255	96.48 285	90.65 297	95.01 331	96.96 306	85.93 327	96.08 178	87.33 343	87.70 220	98.78 224	91.35 265	95.58 214	98.34 192

FMVSNet2		94.47 233	93.61 252	97.04 182	98.21 168	96.43 127	98.79 108	98.27 205	92.46 236	93.50 253	97.09 257	81.16 297	98.00 301	91.09 266	91.93 261	96.70 265

v144192		94.39 237	93.70 248	96.48 226	96.06 301	94.35 220	98.58 143	98.16 225	91.45 269	94.33 216	97.02 268	87.50 224	98.45 252	91.08 267	89.11 297	96.63 273

tpmvs		94.60 222	94.36 208	95.33 276	97.46 223	88.60 322	96.88 305	97.68 264	91.29 278	93.80 242	96.42 302	88.58 195	99.24 166	91.06 268	96.04 210	98.17 196

LTVRE_ROB	92.95 15	94.60 222	93.90 234	96.68 205	97.41 231	94.42 216	98.52 153	98.59 141	91.69 263	91.21 302	98.35 155	84.87 265	99.04 192	91.06 268	93.44 244	96.60 276
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
PAPR		96.84 119	96.24 130	98.65 81	98.72 132	96.92 105	97.36 270	98.57 147	93.33 206	96.67 156	97.57 225	94.30 92	99.56 136	91.05 270	98.59 133	99.47 98

SixPastTwentyTwo		93.34 270	92.86 269	94.75 294	95.67 312	89.41 312	98.75 110	96.67 320	93.89 176	90.15 312	98.25 169	80.87 301	98.27 282	90.90 271	90.64 277	96.57 280

MVS_0304		92.81 281	92.01 282	95.23 277	97.46 223	91.33 285	98.17 206	98.81 76	91.13 285	93.80 242	95.68 322	66.08 345	98.06 296	90.79 272	96.13 208	96.32 306

COLMAP_ROB	93.27 12	95.33 182	94.87 183	96.71 201	99.29 78	93.24 257	98.58 143	98.11 232	89.92 302	93.57 248	99.10 69	86.37 243	99.79 91	90.78 273	98.10 153	97.09 223
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
pmmvs6		91.77 289	90.63 293	95.17 280	94.69 331	91.24 288	98.67 132	97.92 255	86.14 325	89.62 315	97.56 227	75.79 329	98.34 271	90.75 274	84.56 332	95.94 316

BH-RMVSNet		95.92 151	95.32 163	97.69 148	98.32 162	94.64 205	98.19 201	97.45 283	94.56 151	96.03 179	98.61 127	85.02 262	99.12 179	90.68 275	99.06 112	99.30 120

DTE-MVSNet		93.98 261	93.26 264	96.14 247	96.06 301	94.39 218	99.20 29	98.86 61	93.06 216	91.78 297	97.81 205	85.87 251	97.58 318	90.53 276	86.17 328	96.46 299

v10		94.29 242	93.55 254	96.51 224	96.39 288	94.80 201	98.99 64	98.19 215	91.35 274	93.02 269	96.99 271	88.09 209	98.41 264	90.50 277	88.41 306	96.33 305

ambc				89.49 324	86.66 347	75.78 347	92.66 343	96.72 317		86.55 329	92.50 336	46.01 350	97.90 307	90.32 278	82.09 334	94.80 330

lessismore_v0				94.45 303	94.93 327	88.44 324		91.03 350		86.77 328	97.64 219	76.23 327	98.42 257	90.31 279	85.64 331	96.51 293

v1192		94.32 240	93.58 253	96.53 222	96.10 299	94.45 215	98.50 158	98.17 223	91.54 267	94.19 224	97.06 263	86.95 233	98.43 256	90.14 280	89.57 289	96.70 265

MVS		94.67 219	93.54 255	98.08 123	96.88 264	96.56 121	98.19 201	98.50 166	78.05 342	92.69 278	98.02 183	91.07 149	99.63 128	90.09 281	98.36 146	98.04 199

ADS-MVSNet2		94.58 225	94.40 207	95.11 282	98.00 184	88.74 320	96.04 322	97.30 290	90.15 297	96.47 169	96.64 294	87.89 214	97.56 319	90.08 282	97.06 177	99.02 151

ADS-MVSNet		95.00 199	94.45 203	96.63 209	98.00 184	91.91 274	96.04 322	97.74 263	90.15 297	96.47 169	96.64 294	87.89 214	98.96 200	90.08 282	97.06 177	99.02 151

MSDG		95.93 150	95.30 165	97.83 136	98.90 116	95.36 174	96.83 309	98.37 187	91.32 276	94.43 211	98.73 118	90.27 163	99.60 131	90.05 284	98.82 124	98.52 185

v1921920		94.20 247	93.47 258	96.40 234	95.98 304	94.08 227	98.52 153	98.15 226	91.33 275	94.25 220	97.20 249	86.41 242	98.42 257	90.04 285	89.39 294	96.69 270

dp		94.15 251	93.90 234	94.90 288	97.31 235	86.82 335	96.97 294	97.19 296	91.22 282	96.02 180	96.61 296	85.51 255	99.02 195	90.00 286	94.30 218	98.85 163

CMPMVS	66.06 21	89.70 304	89.67 301	89.78 323	93.19 337	76.56 346	97.00 293	98.35 190	80.97 339	81.57 339	97.75 208	74.75 333	98.61 235	89.85 287	93.63 238	94.17 333
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
TR-MVS		94.94 206	94.20 213	97.17 175	97.75 198	94.14 226	97.59 256	97.02 304	92.28 247	95.75 183	97.64 219	83.88 284	98.96 200	89.77 288	96.15 207	98.40 189

MS-PatchMatch		93.84 263	93.63 251	94.46 302	96.18 295	89.45 310	97.76 244	98.27 205	92.23 248	92.13 293	97.49 229	79.50 308	98.69 228	89.75 289	99.38 100	95.25 323

ITE_SJBPF				95.44 273	97.42 228	91.32 286		97.50 278	95.09 130	93.59 246	98.35 155	81.70 295	98.88 213	89.71 290	93.39 245	96.12 311

MVP-Stereo		94.28 244	93.92 232	95.35 275	94.95 326	92.60 266	97.97 225	97.65 266	91.61 266	90.68 309	97.09 257	86.32 244	98.42 257	89.70 291	99.34 102	95.02 328
Qingsong Yan: MVP-Stereo: A Parallel Multi-View Patchmatch Stereo Method with Dilation Matching for Photogrammetric Application.
AllTest		95.24 186	94.65 191	96.99 184	99.25 86	93.21 258	98.59 141	98.18 218	91.36 272	93.52 250	98.77 114	84.67 268	99.72 108	89.70 291	97.87 159	98.02 200

TestCases				96.99 184	99.25 86	93.21 258		98.18 218	91.36 272	93.52 250	98.77 114	84.67 268	99.72 108	89.70 291	97.87 159	98.02 200

GG-mvs-BLEND				96.59 214	96.34 290	94.98 191	96.51 319	88.58 353		93.10 268	94.34 330	80.34 306	98.05 297	89.53 294	96.99 179	96.74 258

USDC		93.33 271	92.71 272	95.21 278	96.83 267	90.83 293	96.91 299	97.50 278	93.84 179	90.72 308	98.14 176	77.69 319	98.82 220	89.51 295	93.21 249	95.97 315

v7n		94.19 248	93.43 259	96.47 227	95.90 306	94.38 219	99.26 18	98.34 192	91.99 254	92.76 275	97.13 252	88.31 202	98.52 246	89.48 296	87.70 312	96.52 290

PM-MVS		87.77 309	86.55 313	91.40 322	91.03 343	83.36 341	96.92 297	95.18 333	91.28 279	86.48 330	93.42 332	53.27 349	96.74 329	89.43 297	81.97 336	94.11 334

FMVSNet1		93.19 276	92.07 281	96.56 218	97.54 217	95.00 188	98.82 97	98.18 218	90.38 295	92.27 290	97.07 260	73.68 337	97.95 303	89.36 298	91.30 268	96.72 261

tpm cat1		93.36 268	92.80 270	95.07 284	97.58 211	87.97 328	96.76 311	97.86 258	82.17 338	93.53 249	96.04 313	86.13 246	99.13 178	89.24 299	95.87 211	98.10 198

UnsupCasMVSNet_eth		90.99 296	89.92 299	94.19 306	94.08 334	89.83 304	97.13 288	98.67 128	93.69 190	85.83 332	96.19 310	75.15 331	96.74 329	89.14 300	79.41 340	96.00 314

v1240		94.06 259	93.29 263	96.34 238	96.03 303	93.90 231	98.44 165	98.17 223	91.18 284	94.13 227	97.01 270	86.05 248	98.42 257	89.13 301	89.50 292	96.70 265

tmp_tt		68.90 318	66.97 320	74.68 333	50.78 358	59.95 355	87.13 347	83.47 356	38.80 353	62.21 349	96.23 307	64.70 346	76.91 354	88.91 302	30.49 352	87.19 344

pmmvs-eth3d		90.36 301	89.05 304	94.32 304	91.10 342	92.12 269	97.63 255	96.95 307	88.86 313	84.91 335	93.13 333	78.32 315	96.74 329	88.70 303	81.81 337	94.09 335

thres600view7		95.49 168	94.77 185	97.67 150	98.98 112	95.02 187	98.85 90	96.90 310	95.38 110	96.63 158	96.90 280	84.29 273	99.59 132	88.65 304	96.33 197	98.40 189

thres100view900		95.38 176	94.70 189	97.41 164	98.98 112	94.92 195	98.87 87	96.90 310	95.38 110	96.61 159	96.88 281	84.29 273	99.56 136	88.11 305	96.29 199	97.76 205

tfpn200view9		95.32 183	94.62 192	97.43 163	98.94 114	94.98 191	98.68 129	96.93 308	95.33 113	96.55 163	96.53 297	84.23 276	99.56 136	88.11 305	96.29 199	97.76 205

thres400		95.38 176	94.62 192	97.65 153	98.94 114	94.98 191	98.68 129	96.93 308	95.33 113	96.55 163	96.53 297	84.23 276	99.56 136	88.11 305	96.29 199	98.40 189

our_test_3		93.65 266	93.30 262	94.69 295	95.45 320	89.68 308	96.91 299	97.65 266	91.97 255	91.66 299	96.88 281	89.67 169	97.93 306	88.02 308	91.49 266	96.48 297

thres200		95.25 185	94.57 194	97.28 169	98.81 124	94.92 195	98.20 197	97.11 297	95.24 121	96.54 165	96.22 309	84.58 270	99.53 142	87.93 309	96.50 193	97.39 216

EG-PatchMatch MVS		91.13 294	90.12 297	94.17 307	94.73 330	89.00 318	98.13 210	97.81 259	89.22 311	85.32 334	96.46 299	67.71 342	98.42 257	87.89 310	93.82 235	95.08 326

CR-MVSNet		94.76 213	94.15 218	96.59 214	97.00 255	93.43 248	94.96 332	97.56 270	92.46 236	96.93 144	96.24 305	88.15 207	97.88 311	87.38 311	96.65 187	98.46 187

Patchmtry		93.22 274	92.35 278	95.84 260	96.77 268	93.09 262	94.66 337	97.56 270	87.37 319	92.90 271	96.24 305	88.15 207	97.90 307	87.37 312	90.10 283	96.53 287

test0.0.03 1		94.08 257	93.51 256	95.80 261	95.53 317	92.89 264	97.38 266	95.97 325	95.11 127	92.51 285	96.66 291	87.71 218	96.94 328	87.03 313	93.67 236	97.57 212

TinyColmap		92.31 286	91.53 287	94.65 297	96.92 260	89.75 305	96.92 297	96.68 319	90.45 293	89.62 315	97.85 199	76.06 328	98.81 221	86.74 314	92.51 255	95.41 322

MIMVSNet		93.26 273	92.21 280	96.41 233	97.73 202	93.13 260	95.65 329	97.03 302	91.27 280	94.04 231	96.06 312	75.33 330	97.19 325	86.56 315	96.23 205	98.92 161

TransMVSNet (Re)		92.67 283	91.51 288	96.15 246	96.58 279	94.65 204	98.90 78	96.73 316	90.86 288	89.46 317	97.86 197	85.62 254	98.09 293	86.45 316	81.12 338	95.71 319

DSMNet-mixed		92.52 285	92.58 275	92.33 318	94.15 333	82.65 342	98.30 185	94.26 342	89.08 312	92.65 279	95.73 317	85.01 263	95.76 338	86.24 317	97.76 164	98.59 182

testgi		93.06 278	92.45 277	94.88 289	96.43 287	89.90 303	98.75 110	97.54 275	95.60 99	91.63 300	97.91 192	74.46 335	97.02 327	86.10 318	93.67 236	97.72 209

YYNet1		90.70 299	89.39 302	94.62 298	94.79 329	90.65 297	97.20 281	97.46 281	87.54 318	72.54 345	95.74 316	86.51 238	96.66 333	86.00 319	86.76 326	96.54 285

MDA-MVSNet_test_wron		90.71 298	89.38 303	94.68 296	94.83 328	90.78 295	97.19 282	97.46 281	87.60 317	72.41 346	95.72 319	86.51 238	96.71 332	85.92 320	86.80 325	96.56 282

UnsupCasMVSNet_bld		87.17 310	85.12 314	93.31 313	91.94 340	88.77 319	94.92 334	98.30 202	84.30 334	82.30 338	90.04 340	63.96 347	97.25 324	85.85 321	74.47 345	93.93 338

EPNet_dtu		95.21 188	94.95 180	95.99 251	96.17 296	90.45 300	98.16 207	97.27 293	96.77 53	93.14 266	98.33 160	90.34 161	98.42 257	85.57 322	98.81 125	99.09 144
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
FMVSNet5		91.81 288	90.92 291	94.49 299	97.21 241	92.09 270	98.00 223	97.55 274	89.31 310	90.86 306	95.61 323	74.48 334	95.32 340	85.57 322	89.70 287	96.07 313

tfpnnormal		93.66 264	92.70 273	96.55 221	96.94 259	95.94 150	98.97 68	99.19 15	91.04 286	91.38 301	97.34 238	84.94 264	98.61 235	85.45 324	89.02 300	95.11 325

Patchmatch-test		94.42 235	93.68 250	96.63 209	97.60 209	91.76 277	94.83 336	97.49 280	89.45 308	94.14 226	97.10 253	88.99 185	98.83 219	85.37 325	98.13 152	99.29 122

ppachtmachnet_test		93.22 274	92.63 274	94.97 286	95.45 320	90.84 292	96.88 305	97.88 257	90.60 290	92.08 294	97.26 243	88.08 210	97.86 312	85.12 326	90.33 279	96.22 308

PCF-MVS	93.45 11	94.68 216	93.43 259	98.42 101	98.62 141	96.77 111	95.48 330	98.20 214	84.63 333	93.34 258	98.32 161	88.55 198	99.81 70	84.80 327	98.96 115	98.68 175
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
MDA-MVSNet-bldmvs		89.97 303	88.35 308	94.83 292	95.21 323	91.34 283	97.64 253	97.51 277	88.36 315	71.17 347	96.13 311	79.22 310	96.63 334	83.65 328	86.27 327	96.52 290

MVS-HIRNet		89.46 306	88.40 307	92.64 316	97.58 211	82.15 343	94.16 341	93.05 348	75.73 344	90.90 305	82.52 345	79.42 309	98.33 272	83.53 329	98.68 127	97.43 213

new-patchmatchnet		88.50 308	87.45 311	91.67 321	90.31 344	85.89 336	97.16 286	97.33 289	89.47 307	83.63 337	92.77 334	76.38 326	95.06 342	82.70 330	77.29 342	94.06 336

PAPM		94.95 204	94.00 227	97.78 139	97.04 254	95.65 162	96.03 324	98.25 210	91.23 281	94.19 224	97.80 206	91.27 144	98.86 216	82.61 331	97.61 169	98.84 165

LCM-MVSNet		78.70 313	76.24 318	86.08 326	77.26 354	71.99 350	94.34 339	96.72 317	61.62 348	76.53 342	89.33 341	33.91 356	92.78 346	81.85 332	74.60 344	93.46 339

new_pmnet		90.06 302	89.00 305	93.22 315	94.18 332	88.32 326	96.42 320	96.89 312	86.19 324	85.67 333	93.62 331	77.18 324	97.10 326	81.61 333	89.29 295	94.23 332

pmmvs3		86.67 312	84.86 315	92.11 320	88.16 346	87.19 334	96.63 315	94.75 337	79.88 340	87.22 326	92.75 335	66.56 344	95.20 341	81.24 334	76.56 343	93.96 337

N_pmnet		87.12 311	87.77 310	85.17 328	95.46 319	61.92 353	97.37 268	70.66 358	85.83 328	88.73 321	96.04 313	85.33 260	97.76 314	80.02 335	90.48 278	95.84 317

TAPA-MVS	93.98 7	95.35 180	94.56 195	97.74 143	99.13 102	94.83 199	98.33 177	98.64 136	86.62 321	96.29 174	98.61 127	94.00 97	99.29 162	80.00 336	99.41 97	99.09 144
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
DeepMVS_CX				86.78 325	97.09 252	72.30 349		95.17 334	75.92 343	84.34 336	95.19 324	70.58 340	95.35 339	79.98 337	89.04 299	92.68 341

Anonymous20231206		91.66 290	91.10 290	93.33 312	94.02 335	87.35 332	98.58 143	97.26 294	90.48 291	90.16 311	96.31 303	83.83 286	96.53 335	79.36 338	89.90 285	96.12 311

test20.03		90.89 297	90.38 295	92.43 317	93.48 336	88.14 327	98.33 177	97.56 270	93.40 204	87.96 323	96.71 290	80.69 304	94.13 344	79.15 339	86.17 328	95.01 329

PatchT		93.06 278	91.97 283	96.35 237	96.69 274	92.67 265	94.48 338	97.08 298	86.62 321	97.08 136	92.23 338	87.94 213	97.90 307	78.89 340	96.69 185	98.49 186

MIMVSNet1		89.67 305	88.28 309	93.82 308	92.81 339	91.08 290	98.01 221	97.45 283	87.95 316	87.90 324	95.87 315	67.63 343	94.56 343	78.73 341	88.18 308	95.83 318

test_0402		91.32 292	90.27 296	94.48 300	96.60 278	91.12 289	98.50 158	97.22 295	86.10 326	88.30 322	96.98 272	77.65 321	97.99 302	78.13 342	92.94 252	94.34 331

OpenMVS_ROB	86.42 20	89.00 307	87.43 312	93.69 309	93.08 338	89.42 311	97.91 229	96.89 312	78.58 341	85.86 331	94.69 328	69.48 341	98.29 280	77.13 343	93.29 248	93.36 340

RPMNet		92.81 281	91.34 289	97.24 170	97.00 255	93.43 248	94.96 332	98.80 86	82.27 337	96.93 144	92.12 339	86.98 232	99.82 63	76.32 344	96.65 187	98.46 187

PMMVS2		77.95 315	75.44 319	85.46 327	82.54 349	74.95 348	94.23 340	93.08 347	72.80 345	74.68 343	87.38 342	36.36 355	91.56 347	73.95 345	63.94 347	89.87 342

FPMVS		77.62 316	77.14 316	79.05 331	79.25 352	60.97 354	95.79 327	95.94 326	65.96 346	67.93 348	94.40 329	37.73 354	88.88 349	68.83 346	88.46 305	87.29 343

Gipuma		78.40 314	76.75 317	83.38 329	95.54 316	80.43 345	79.42 350	97.40 287	64.67 347	73.46 344	80.82 347	45.65 351	93.14 345	66.32 347	87.43 315	76.56 348
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
ANet_high		69.08 317	65.37 321	80.22 330	65.99 356	71.96 351	90.91 346	90.09 351	82.62 336	49.93 353	78.39 348	29.36 357	81.75 350	62.49 348	38.52 351	86.95 345

PMVS	61.03 23	65.95 319	63.57 323	73.09 334	57.90 357	51.22 358	85.05 349	93.93 346	54.45 349	44.32 354	83.57 344	13.22 358	89.15 348	58.68 349	81.00 339	78.91 347
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
MVE	62.14 22	63.28 322	59.38 325	74.99 332	74.33 355	65.47 352	85.55 348	80.50 357	52.02 351	51.10 352	75.00 351	10.91 361	80.50 351	51.60 350	53.40 348	78.99 346
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
E-PMN		64.94 320	64.25 322	67.02 335	82.28 350	59.36 356	91.83 345	85.63 354	52.69 350	60.22 350	77.28 349	41.06 353	80.12 352	46.15 351	41.14 349	61.57 350

EMVS		64.07 321	63.26 324	66.53 336	81.73 351	58.81 357	91.85 344	84.75 355	51.93 352	59.09 351	75.13 350	43.32 352	79.09 353	42.03 352	39.47 350	61.69 349

wuyk23d		30.17 323	30.18 327	30.16 337	78.61 353	43.29 359	66.79 351	14.21 359	17.31 354	14.82 357	11.93 357	11.55 360	41.43 355	37.08 353	19.30 353	5.76 353

test123		20.95 326	23.72 329	12.64 338	13.54 360	8.19 360	96.55 318	6.13 361	7.48 356	16.74 356	37.98 354	12.97 359	6.05 356	16.69 354	5.43 355	23.68 351

testmvs		21.48 325	24.95 328	11.09 339	14.89 359	6.47 361	96.56 317	9.87 360	7.55 355	17.93 355	39.02 353	9.43 362	5.90 357	16.56 355	12.72 354	20.91 352

uanet_test		0.00 329	0.00 332	0.00 340	0.00 361	0.00 362	0.00 352	0.00 362	0.00 357	0.00 358	0.00 358	0.00 363	0.00 358	0.00 356	0.00 356	0.00 354

cdsmvs_eth3d_5k		23.98 324	31.98 326	0.00 340	0.00 361	0.00 362	0.00 352	98.59 141	0.00 357	0.00 358	98.61 127	90.60 157	0.00 358	0.00 356	0.00 356	0.00 354

pcd_1.5k_mvsjas		7.88 328	10.50 331	0.00 340	0.00 361	0.00 362	0.00 352	0.00 362	0.00 357	0.00 358	0.00 358	94.51 85	0.00 358	0.00 356	0.00 356	0.00 354

sosnet-low-res		0.00 329	0.00 332	0.00 340	0.00 361	0.00 362	0.00 352	0.00 362	0.00 357	0.00 358	0.00 358	0.00 363	0.00 358	0.00 356	0.00 356	0.00 354

sosnet		0.00 329	0.00 332	0.00 340	0.00 361	0.00 362	0.00 352	0.00 362	0.00 357	0.00 358	0.00 358	0.00 363	0.00 358	0.00 356	0.00 356	0.00 354

uncertanet		0.00 329	0.00 332	0.00 340	0.00 361	0.00 362	0.00 352	0.00 362	0.00 357	0.00 358	0.00 358	0.00 363	0.00 358	0.00 356	0.00 356	0.00 354

Regformer		0.00 329	0.00 332	0.00 340	0.00 361	0.00 362	0.00 352	0.00 362	0.00 357	0.00 358	0.00 358	0.00 363	0.00 358	0.00 356	0.00 356	0.00 354

ab-mvs-re		8.20 327	10.94 330	0.00 340	0.00 361	0.00 362	0.00 352	0.00 362	0.00 357	0.00 358	98.43 145	0.00 363	0.00 358	0.00 356	0.00 356	0.00 354

uanet		0.00 329	0.00 332	0.00 340	0.00 361	0.00 362	0.00 352	0.00 362	0.00 357	0.00 358	0.00 358	0.00 363	0.00 358	0.00 356	0.00 356	0.00 354

test_241102_ONE					99.71 20	99.24 4		98.87 55	97.62 11	99.73 1	99.39 14	97.53 4	99.74 106

save fliter					99.46 51	98.38 35	98.21 194	98.71 113	97.95 3

test0726					99.72 12	99.25 2	99.06 51	98.88 49	97.62 11	99.56 5	99.50 4	97.42 6

GSMVS																99.20 129

test_part2					99.63 29	99.18 8				99.27 17

sam_mvs1												89.45 172				99.20 129

sam_mvs												88.99 185

MTGPA								98.74 102

test_post											31.83 355	88.83 192	98.91 207

patchmatchnet-post											95.10 326	89.42 173	98.89 211

MTMP							98.89 82	94.14 344

TEST9					99.31 70	98.50 29	97.92 227	98.73 107	92.63 231	97.74 111	98.68 121	96.20 23	99.80 79

test_8					99.29 78	98.44 31	97.89 233	98.72 109	92.98 220	97.70 114	98.66 124	96.20 23	99.80 79

agg_prior					99.30 75	98.38 35		98.72 109		97.57 125			99.81 70

test_prior4						98.01 62	97.86 236

test_prior				99.19 43	99.31 70	98.22 50		98.84 65					99.70 114			99.65 67

新几何2							97.64 253

旧先验1					99.29 78	97.48 82		98.70 116			99.09 74	95.56 47			99.47 90	99.61 75

原ACMM2							97.67 251

test222					99.23 93	97.17 97	97.40 264	98.66 131	88.68 314	98.05 86	98.96 93	94.14 94			99.53 85	99.61 75

segment_acmp												96.85 11

testdata1							97.32 274		96.34 71

test12				99.18 47	99.16 99	98.19 52		98.53 156		98.07 85		95.13 70	99.72 108		99.56 80	99.63 73

plane_prior7					97.42 228	94.63 206

plane_prior6					97.35 233	94.61 209						87.09 229

plane_prior4											98.28 164

plane_prior3						94.61 209			97.02 47	95.34 185

plane_prior2							98.80 104		97.28 29

plane_prior1					97.37 232

plane_prior						94.60 211	98.44 165		96.74 55						94.22 221

n2								0.00 362

nn								0.00 362

door-mid								94.37 340

test11								98.66 131

door								94.64 338

HQP5-MVS						94.25 224

HQP-NCC					97.20 242		98.05 217		96.43 67	94.45 207

ACMP_Plane					97.20 242		98.05 217		96.43 67	94.45 207

HQP4-MVS										94.45 207			98.96 200			96.87 244

HQP3-MVS								98.46 171							94.18 223

HQP2-MVS												86.75 235

NP-MVS					97.28 236	94.51 214					97.73 209

ACMMP++_ref															92.97 251

ACMMP++															93.61 239

Test By Simon												94.64 80

High-resolution multi-view benchmark