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.

MVS_0304		97.70 59	97.25 67	99.07 45	98.90 101	97.83 51	98.20 199	98.74 81	97.51 8	98.03 66	99.06 59	86.12 230	99.93 9	99.02 1	99.64 48	99.44 87

APDe-MVS		99.02 1	98.84 1	99.55 3	99.57 25	98.96 5	99.39 5	98.93 36	97.38 17	99.41 3	99.54 1	96.66 7	99.84 46	98.86 2	99.85 2	99.87 1

CANet		98.05 45	97.76 47	98.90 57	98.73 120	97.27 69	98.35 182	98.78 73	97.37 19	97.72 86	98.96 73	91.53 114	99.92 15	98.79 3	99.65 46	99.51 72

Regformer-4		98.64 11	98.53 7	98.99 49	99.43 38	97.37 66	98.40 178	98.79 71	97.46 12	99.09 16	99.31 22	95.86 34	99.80 62	98.64 4	99.76 26	99.79 4

VDD-MVS		95.82 134	95.23 144	97.61 140	98.84 115	93.98 232	98.68 138	97.40 267	95.02 115	97.95 73	99.34 20	74.37 333	99.78 79	98.64 4	96.80 158	99.08 126

EI-MVSNet-Vis-set		98.47 30	98.39 15	98.69 64	99.46 35	96.49 100	98.30 191	98.69 96	97.21 28	98.84 30	99.36 17	95.41 42	99.78 79	98.62 6	99.65 46	99.80 3

Regformer-3		98.59 17	98.50 11	98.86 59	99.43 38	97.05 77	98.40 178	98.68 99	97.43 13	99.06 17	99.31 22	95.80 35	99.77 84	98.62 6	99.76 26	99.78 7

EI-MVSNet-UG-set		98.41 32	98.34 22	98.61 69	99.45 36	96.32 108	98.28 193	98.68 99	97.17 31	98.74 37	99.37 13	95.25 48	99.79 74	98.57 8	99.54 67	99.73 30

CHOSEN 280x420		97.18 87	97.18 71	97.20 162	98.81 116	93.27 250	95.78 328	99.15 18	95.25 104	96.79 127	98.11 149	92.29 92	99.07 175	98.56 9	99.85 2	99.25 104

xiu_mvs_v1_base_debu		97.60 63	97.56 53	97.72 123	98.35 139	95.98 116	97.86 242	98.51 134	97.13 34	99.01 20	98.40 122	91.56 110	99.80 62	98.53 10	98.68 105	97.37 203

xiu_mvs_v1_base		97.60 63	97.56 53	97.72 123	98.35 139	95.98 116	97.86 242	98.51 134	97.13 34	99.01 20	98.40 122	91.56 110	99.80 62	98.53 10	98.68 105	97.37 203

xiu_mvs_v1_base_debi		97.60 63	97.56 53	97.72 123	98.35 139	95.98 116	97.86 242	98.51 134	97.13 34	99.01 20	98.40 122	91.56 110	99.80 62	98.53 10	98.68 105	97.37 203

VNet		97.79 56	97.40 63	98.96 53	98.88 110	97.55 60	98.63 145	98.93 36	96.74 46	99.02 19	98.84 84	90.33 131	99.83 47	98.53 10	96.66 160	99.50 74

MSLP-MVS++		98.56 23	98.57 5	98.55 73	99.26 66	96.80 86	98.71 129	99.05 23	97.28 21	98.84 30	99.28 28	96.47 12	99.40 137	98.52 14	99.70 40	99.47 80

TSAR-MVS + GP.		98.38 34	98.24 32	98.81 60	99.22 74	97.25 72	98.11 214	98.29 170	97.19 30	98.99 23	99.02 61	96.22 14	99.67 101	98.52 14	98.56 113	99.51 72

Regformer-1		98.66 9	98.51 10	99.12 42	99.35 40	97.81 53	98.37 180	98.76 77	97.49 10	99.20 13	99.21 35	96.08 22	99.79 74	98.42 16	99.73 37	99.75 22

DELS-MVS		98.40 33	98.20 36	98.99 49	99.00 90	97.66 55	97.75 251	98.89 44	97.71 6	98.33 56	98.97 68	94.97 55	99.88 37	98.42 16	99.76 26	99.42 89
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 8	98.52 8	99.19 30	99.35 40	98.01 44	98.37 180	98.81 62	97.48 11	99.21 12	99.21 35	96.13 19	99.80 62	98.40 18	99.73 37	99.75 22

alignmvs		97.56 67	97.07 77	99.01 48	98.66 127	98.37 23	98.83 96	98.06 219	96.74 46	98.00 71	97.65 188	90.80 125	99.48 135	98.37 19	96.56 164	99.19 110

TSAR-MVS + MP.		98.78 3	98.62 4	99.24 27	99.69 17	98.28 30	99.14 45	98.66 109	96.84 43	99.56 2	99.31 22	96.34 13	99.70 96	98.32 20	99.73 37	99.73 30

DeepPCF-MVS	96.37 2	97.93 50	98.48 13	96.30 236	99.00 90	89.54 300	97.43 270	98.87 49	98.16 2	99.26 9	99.38 12	96.12 20	99.64 105	98.30 21	99.77 20	99.72 33

canonicalmvs		97.67 61	97.23 69	98.98 51	98.70 123	98.38 20	99.34 11	98.39 157	96.76 45	97.67 89	97.40 203	92.26 93	99.49 131	98.28 22	96.28 182	99.08 126

SD-MVS		98.64 11	98.68 3	98.53 75	99.33 45	98.36 24	98.90 76	98.85 53	97.28 21	99.72 1	99.39 8	96.63 9	97.60 303	98.17 23	99.85 2	99.64 56

MP-MVS-pluss		98.31 41	97.92 44	99.49 6	99.72 11	98.88 7	98.43 175	98.78 73	94.10 144	97.69 88	99.42 6	95.25 48	99.92 15	98.09 24	99.80 10	99.67 49
MP-MVS-pluss: MP-MVS-pluss. MP-MVS-pluss
SMA-MVS		98.57 21	98.24 32	99.56 2	99.48 33	99.04 4	98.95 71	98.80 69	93.67 174	99.37 5	99.50 3	96.52 11	99.89 29	98.06 25	99.81 8	99.75 22

CNVR-MVS		98.78 3	98.56 6	99.45 10	99.32 48	98.87 8	98.47 171	98.81 62	97.72 4	98.76 36	99.16 45	97.05 4	99.78 79	98.06 25	99.66 45	99.69 38

MVS_111021_HR		98.47 30	98.34 22	98.88 58	99.22 74	97.32 67	97.91 234	99.58 3	97.20 29	98.33 56	99.00 66	95.99 27	99.64 105	98.05 27	99.76 26	99.69 38

VDDNet		95.36 170	94.53 183	97.86 114	98.10 159	95.13 165	98.85 92	97.75 233	90.46 275	98.36 54	99.39 8	73.27 335	99.64 105	97.98 28	96.58 163	98.81 144

MCST-MVS		98.65 10	98.37 18	99.48 7	99.60 24	98.87 8	98.41 177	98.68 99	97.04 38	98.52 47	98.80 88	96.78 6	99.83 47	97.93 29	99.61 51	99.74 28

zzz-MVS		98.55 24	98.25 30	99.46 8	99.76 1	98.64 11	98.55 158	98.74 81	97.27 25	98.02 67	99.39 8	94.81 57	99.96 1	97.91 30	99.79 11	99.77 14

MTAPA		98.58 19	98.29 27	99.46 8	99.76 1	98.64 11	98.90 76	98.74 81	97.27 25	98.02 67	99.39 8	94.81 57	99.96 1	97.91 30	99.79 11	99.77 14

MVS_111021_LR		98.34 38	98.23 34	98.67 66	99.27 64	96.90 83	97.95 229	99.58 3	97.14 33	98.44 52	99.01 65	95.03 54	99.62 110	97.91 30	99.75 32	99.50 74

ACMMP_Plus		98.61 14	98.30 26	99.55 3	99.62 23	98.95 6	98.82 98	98.81 62	95.80 74	99.16 15	99.47 5	95.37 43	99.92 15	97.89 33	99.75 32	99.79 4

PS-MVSNAJ		97.73 57	97.77 46	97.62 135	98.68 126	95.58 147	97.34 279	98.51 134	97.29 20	98.66 40	97.88 167	94.51 63	99.90 27	97.87 34	99.17 89	97.39 201

XVS		98.70 5	98.49 12	99.34 15	99.70 15	98.35 25	99.29 15	98.88 47	97.40 14	98.46 48	99.20 38	95.90 32	99.89 29	97.85 35	99.74 35	99.78 7

X-MVStestdata		94.06 248	92.30 268	99.34 15	99.70 15	98.35 25	99.29 15	98.88 47	97.40 14	98.46 48	43.50 358	95.90 32	99.89 29	97.85 35	99.74 35	99.78 7

xiu_mvs_v2_base		97.66 62	97.70 49	97.56 143	98.61 132	95.46 153	97.44 268	98.46 144	97.15 32	98.65 41	98.15 146	94.33 69	99.80 62	97.84 37	98.66 109	97.41 199

DeepC-MVS	95.98 3	97.88 51	97.58 52	98.77 61	99.25 67	96.93 81	98.83 96	98.75 80	96.96 41	96.89 119	99.50 3	90.46 128	99.87 38	97.84 37	99.76 26	99.52 69
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
HSP-MVS		98.70 5	98.52 8	99.24 27	99.75 3	98.23 31	99.26 18	98.58 122	97.52 7	99.41 3	98.78 90	96.00 26	99.79 74	97.79 39	99.59 55	99.69 38

CP-MVS		98.57 21	98.36 19	99.19 30	99.66 19	97.86 49	99.34 11	98.87 49	95.96 70	98.60 44	99.13 47	96.05 25	99.94 3	97.77 40	99.86 1	99.77 14

SteuartSystems-ACMMP		98.90 2	98.75 2	99.36 14	99.22 74	98.43 19	99.10 52	98.87 49	97.38 17	99.35 6	99.40 7	97.78 1	99.87 38	97.77 40	99.85 2	99.78 7
Skip Steuart: Steuart Systems R&D Blog.
APD-MVS_3200maxsize		98.53 27	98.33 25	99.15 39	99.50 29	97.92 48	99.15 44	98.81 62	96.24 60	99.20 13	99.37 13	95.30 46	99.80 62	97.73 42	99.67 42	99.72 33

LFMVS		95.86 132	94.98 154	98.47 80	98.87 111	96.32 108	98.84 95	96.02 320	93.40 189	98.62 42	99.20 38	74.99 328	99.63 108	97.72 43	97.20 152	99.46 84

casdiffmvs		97.42 75	97.12 72	98.31 90	98.35 139	96.55 98	99.05 58	98.20 184	94.97 118	97.55 98	98.11 149	92.33 91	99.18 159	97.70 44	97.85 139	99.18 114

PHI-MVS		98.34 38	98.06 39	99.18 34	99.15 81	98.12 40	99.04 61	99.09 19	93.32 192	98.83 32	99.10 51	96.54 10	99.83 47	97.70 44	99.76 26	99.59 64

HPM-MVS		98.36 36	98.10 38	99.13 40	99.74 7	97.82 52	99.53 1	98.80 69	94.63 131	98.61 43	98.97 68	95.13 52	99.77 84	97.65 46	99.83 7	99.79 4
Chunlin Ren, Qingshan Xu, Shikun Zhang, Jiaqi Yang: Hierarchical Prior Mining for Non-local Multi-View Stereo. ICCV 2023
HFP-MVS		98.63 13	98.40 14	99.32 18	99.72 11	98.29 28	99.23 23	98.96 31	96.10 67	98.94 24	99.17 42	96.06 23	99.92 15	97.62 47	99.78 15	99.75 22

ACMMPR		98.59 17	98.36 19	99.29 20	99.74 7	98.15 38	99.23 23	98.95 33	96.10 67	98.93 28	99.19 41	95.70 36	99.94 3	97.62 47	99.79 11	99.78 7

jason		97.32 82	97.08 76	98.06 107	97.45 199	95.59 146	97.87 241	97.91 227	94.79 124	98.55 46	98.83 85	91.12 118	99.23 149	97.58 49	99.60 52	99.34 92
jason: jason.
lupinMVS		97.44 73	97.22 70	98.12 101	98.07 160	95.76 142	97.68 256	97.76 232	94.50 135	98.79 33	98.61 105	92.34 89	99.30 143	97.58 49	99.59 55	99.31 95

HPM-MVS_fast		98.38 34	98.13 37	99.12 42	99.75 3	97.86 49	99.44 4	98.82 59	94.46 138	98.94 24	99.20 38	95.16 51	99.74 90	97.58 49	99.85 2	99.77 14

region2R		98.61 14	98.38 17	99.29 20	99.74 7	98.16 37	99.23 23	98.93 36	96.15 62	98.94 24	99.17 42	95.91 31	99.94 3	97.55 52	99.79 11	99.78 7

DeepC-MVS_fast	96.70 1	98.55 24	98.34 22	99.18 34	99.25 67	98.04 42	98.50 168	98.78 73	97.72 4	98.92 29	99.28 28	95.27 47	99.82 53	97.55 52	99.77 20	99.69 38
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 19	98.25 30	99.55 3	99.50 29	99.08 3	98.72 128	98.66 109	97.51 8	98.15 58	98.83 85	95.70 36	99.92 15	97.53 54	99.67 42	99.66 51

nrg030		96.28 121	95.72 123	97.96 111	96.90 232	98.15 38	99.39 5	98.31 165	95.47 86	94.42 199	98.35 128	92.09 100	98.69 213	97.50 55	89.05 281	97.04 215

CSCG		97.85 54	97.74 48	98.20 95	99.67 18	95.16 163	99.22 29	99.32 7	93.04 200	97.02 111	98.92 79	95.36 44	99.91 24	97.43 56	99.64 48	99.52 69

mPP-MVS		98.51 28	98.26 29	99.25 26	99.75 3	98.04 42	99.28 17	98.81 62	96.24 60	98.35 55	99.23 32	95.46 41	99.94 3	97.42 57	99.81 8	99.77 14

mvs_anonymous		96.70 104	96.53 100	97.18 164	98.19 153	93.78 237	98.31 189	98.19 186	94.01 148	94.47 190	98.27 139	92.08 101	98.46 243	97.39 58	97.91 135	99.31 95

NCCC		98.61 14	98.35 21	99.38 12	99.28 63	98.61 13	98.45 172	98.76 77	97.82 3	98.45 51	98.93 77	96.65 8	99.83 47	97.38 59	99.41 79	99.71 35

VPA-MVSNet		95.75 136	95.11 148	97.69 129	97.24 210	97.27 69	98.94 73	99.23 12	95.13 109	95.51 169	97.32 210	85.73 243	98.91 195	97.33 60	89.55 274	96.89 230

3Dnovator	94.51 5	97.46 69	96.93 81	99.07 45	97.78 177	97.64 56	99.35 10	99.06 21	97.02 39	93.75 233	99.16 45	89.25 145	99.92 15	97.22 61	99.75 32	99.64 56

#test#		98.54 26	98.27 28	99.32 18	99.72 11	98.29 28	98.98 68	98.96 31	95.65 80	98.94 24	99.17 42	96.06 23	99.92 15	97.21 62	99.78 15	99.75 22

ACMMP		98.23 43	97.95 43	99.09 44	99.74 7	97.62 58	99.03 62	99.41 6	95.98 69	97.60 94	99.36 17	94.45 67	99.93 9	97.14 63	98.85 100	99.70 37
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 59	97.46 60	98.44 82	99.27 64	95.91 136	98.63 145	99.16 17	94.48 137	97.67 89	98.88 81	92.80 85	99.91 24	97.11 64	99.12 90	99.50 74

mvs_tets		95.41 165	95.00 152	96.65 201	95.58 306	94.42 218	99.00 64	98.55 126	95.73 76	93.21 246	98.38 125	83.45 285	98.63 218	97.09 65	94.00 221	96.91 227

EPNet		97.28 83	96.87 84	98.51 76	94.98 318	96.14 113	98.90 76	97.02 290	98.28 1	95.99 166	99.11 49	91.36 115	99.89 29	96.98 66	99.19 88	99.50 74
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
HyFIR lowres test		96.90 98	96.49 101	98.14 98	99.33 45	95.56 149	97.38 273	99.65 2	92.34 231	97.61 93	98.20 144	89.29 144	99.10 172	96.97 67	97.60 148	99.77 14

3Dnovator+	94.38 6	97.43 74	96.78 88	99.38 12	97.83 175	98.52 14	99.37 7	98.71 93	97.09 37	92.99 254	99.13 47	89.36 142	99.89 29	96.97 67	99.57 58	99.71 35

abl_6		98.30 42	98.03 40	99.13 40	99.56 26	97.76 54	99.13 48	98.82 59	96.14 63	99.26 9	99.37 13	93.33 79	99.93 9	96.96 69	99.67 42	99.69 38

jajsoiax		95.45 160	95.03 151	96.73 189	95.42 313	94.63 208	99.14 45	98.52 132	95.74 75	93.22 245	98.36 127	83.87 282	98.65 217	96.95 70	94.04 219	96.91 227

MVSFormer		97.57 66	97.49 58	97.84 115	98.07 160	95.76 142	99.47 2	98.40 155	94.98 116	98.79 33	98.83 85	92.34 89	98.41 258	96.91 71	99.59 55	99.34 92

test_djsdf		96.00 126	95.69 128	96.93 181	95.72 302	95.49 152	99.47 2	98.40 155	94.98 116	94.58 186	97.86 168	89.16 148	98.41 258	96.91 71	94.12 218	96.88 232

test_prior3		98.22 44	97.90 45	99.19 30	99.31 50	98.22 33	97.80 247	98.84 54	96.12 65	97.89 78	98.69 97	95.96 28	99.70 96	96.89 73	99.60 52	99.65 53

test_prior2							97.80 247		96.12 65	97.89 78	98.69 97	95.96 28		96.89 73	99.60 52

EPP-MVSNet		97.46 69	97.28 66	97.99 109	98.64 129	95.38 155	99.33 13	98.31 165	93.61 177	97.19 103	99.07 58	94.05 73	99.23 149	96.89 73	98.43 120	99.37 91

PS-MVSNAJss		96.43 113	96.26 108	96.92 183	95.84 298	95.08 167	99.16 43	98.50 139	95.87 72	93.84 231	98.34 132	94.51 63	98.61 219	96.88 76	93.45 233	97.06 213

PVSNet_BlendedMVS		96.73 103	96.60 96	97.12 168	99.25 67	95.35 158	98.26 195	99.26 8	94.28 140	97.94 74	97.46 199	92.74 86	99.81 55	96.88 76	93.32 236	96.20 296

PVSNet_Blended		97.38 79	97.12 72	98.14 98	99.25 67	95.35 158	97.28 283	99.26 8	93.13 198	97.94 74	98.21 143	92.74 86	99.81 55	96.88 76	99.40 81	99.27 102

Effi-MVS+		97.12 90	96.69 92	98.39 86	98.19 153	96.72 90	97.37 275	98.43 152	93.71 167	97.65 92	98.02 155	92.20 97	99.25 147	96.87 79	97.79 142	99.19 110

CHOSEN 1792x2688		97.12 90	96.80 85	98.08 104	99.30 55	94.56 215	98.05 219	99.71 1	93.57 178	97.09 105	98.91 80	88.17 186	99.89 29	96.87 79	99.56 64	99.81 2

PGM-MVS		98.49 29	98.23 34	99.27 25	99.72 11	98.08 41	98.99 65	99.49 5	95.43 88	99.03 18	99.32 21	95.56 38	99.94 3	96.80 81	99.77 20	99.78 7

XVG-OURS-SEG-HR		96.51 111	96.34 104	97.02 174	98.77 118	93.76 238	97.79 249	98.50 139	95.45 87	96.94 114	99.09 55	87.87 198	99.55 127	96.76 82	95.83 200	97.74 190

MP-MVS		98.33 40	98.01 41	99.28 22	99.75 3	98.18 36	99.22 29	98.79 71	96.13 64	97.92 76	99.23 32	94.54 62	99.94 3	96.74 83	99.78 15	99.73 30
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
test_normal		94.72 209	93.59 240	98.11 102	95.30 315	95.95 123	97.91 234	97.39 269	94.64 130	85.70 319	95.88 298	80.52 301	99.36 141	96.69 84	98.30 125	99.01 132

DI_MVS_plusplus_test		94.74 208	93.62 238	98.09 103	95.34 314	95.92 134	98.09 217	97.34 271	94.66 129	85.89 316	95.91 297	80.49 302	99.38 140	96.66 85	98.22 126	98.97 134

agg_prior1		97.95 48	97.51 57	99.28 22	99.30 55	98.38 20	97.81 246	98.72 88	93.16 197	97.57 96	98.66 102	96.14 18	99.81 55	96.63 86	99.56 64	99.66 51

train_agg		97.97 46	97.52 56	99.33 17	99.31 50	98.50 15	97.92 231	98.73 86	92.98 203	97.74 84	98.68 99	96.20 15	99.80 62	96.59 87	99.57 58	99.68 44

agg_prior3		97.87 52	97.42 62	99.23 29	99.29 58	98.23 31	97.92 231	98.72 88	92.38 230	97.59 95	98.64 104	96.09 21	99.79 74	96.59 87	99.57 58	99.68 44

MVSTER		96.06 125	95.72 123	97.08 172	98.23 149	95.93 127	98.73 126	98.27 171	94.86 123	95.07 174	98.09 151	88.21 185	98.54 227	96.59 87	93.46 231	96.79 240

UGNet		96.78 102	96.30 106	98.19 97	98.24 148	95.89 138	98.88 83	98.93 36	97.39 16	96.81 125	97.84 171	82.60 288	99.90 27	96.53 90	99.49 70	98.79 145
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 37	98.00 42	99.42 11	99.51 28	98.72 10	98.80 107	98.82 59	94.52 134	99.23 11	99.25 31	95.54 40	99.80 62	96.52 91	99.77 20	99.74 28
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
VPNet		94.99 188	94.19 201	97.40 156	97.16 218	96.57 95	98.71 129	98.97 29	95.67 78	94.84 179	98.24 142	80.36 303	98.67 216	96.46 92	87.32 306	96.96 219

sss		97.39 78	96.98 80	98.61 69	98.60 133	96.61 94	98.22 197	98.93 36	93.97 152	98.01 69	98.48 117	91.98 103	99.85 43	96.45 93	98.15 129	99.39 90

MVS_Test		97.28 83	97.00 79	98.13 100	98.33 144	95.97 120	98.74 123	98.07 217	94.27 141	98.44 52	98.07 152	92.48 88	99.26 146	96.43 94	98.19 128	99.16 116

FIs		96.51 111	96.12 112	97.67 131	97.13 220	97.54 61	99.36 8	99.22 14	95.89 71	94.03 224	98.35 128	91.98 103	98.44 248	96.40 95	92.76 243	97.01 216

test9_res														96.39 96	99.57 58	99.69 38

test_part3							98.55 158		96.40 57		99.31 22		99.93 9	96.37 97

ESAPD		98.70 5	98.39 15	99.62 1	99.63 21	99.18 1	98.55 158	98.84 54	96.40 57	99.27 7	99.31 22	97.38 2	99.93 9	96.37 97	99.78 15	99.76 20

Anonymous20240529		95.10 184	94.22 196	97.75 121	99.01 89	94.26 226	98.87 84	98.83 58	85.79 325	96.64 130	98.97 68	78.73 310	99.85 43	96.27 99	94.89 207	99.12 121

PMMVS		96.60 106	96.33 105	97.41 154	97.90 171	93.93 233	97.35 278	98.41 153	92.84 210	97.76 82	97.45 201	91.10 120	99.20 157	96.26 100	97.91 135	99.11 122

CLD-MVS		95.62 144	95.34 137	96.46 226	97.52 193	93.75 240	97.27 284	98.46 144	95.53 84	94.42 199	98.00 158	86.21 228	98.97 185	96.25 101	94.37 208	96.66 261
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
Anonymous202405211		95.28 176	94.49 184	97.67 131	99.00 90	93.75 240	98.70 132	97.04 288	90.66 272	96.49 153	98.80 88	78.13 313	99.83 47	96.21 102	95.36 204	99.44 87

HQP_MVS		96.14 124	95.90 118	96.85 184	97.42 200	94.60 213	98.80 107	98.56 124	97.28 21	95.34 170	98.28 136	87.09 214	99.03 181	96.07 103	94.27 210	96.92 222

plane_prior5								98.56 124					99.03 181	96.07 103	94.27 210	96.92 222

CPTT-MVS		97.72 58	97.32 65	98.92 55	99.64 20	97.10 76	99.12 50	98.81 62	92.34 231	98.09 61	99.08 57	93.01 83	99.92 15	96.06 105	99.77 20	99.75 22

DP-MVS Recon		97.86 53	97.46 60	99.06 47	99.53 27	98.35 25	98.33 184	98.89 44	92.62 214	98.05 63	98.94 76	95.34 45	99.65 103	96.04 106	99.42 78	99.19 110

FC-MVSNet-test		96.42 114	96.05 113	97.53 144	96.95 227	97.27 69	99.36 8	99.23 12	95.83 73	93.93 226	98.37 126	92.00 102	98.32 267	96.02 107	92.72 244	97.00 217

Vis-MVSNet		97.42 75	97.11 74	98.34 88	98.66 127	96.23 111	99.22 29	99.00 26	96.63 51	98.04 65	99.21 35	88.05 192	99.35 142	96.01 108	99.21 87	99.45 86
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
ab-mvs		96.42 114	95.71 126	98.55 73	98.63 130	96.75 89	97.88 240	98.74 81	93.84 158	96.54 140	98.18 145	85.34 251	99.75 88	95.93 109	96.35 174	99.15 117

WTY-MVS		97.37 80	96.92 82	98.72 63	98.86 112	96.89 85	98.31 189	98.71 93	95.26 103	97.67 89	98.56 111	92.21 96	99.78 79	95.89 110	96.85 157	99.48 79

XVG-OURS		96.55 110	96.41 102	96.99 175	98.75 119	93.76 238	97.50 267	98.52 132	95.67 78	96.83 122	99.30 27	88.95 156	99.53 128	95.88 111	96.26 183	97.69 194

agg_prior2														95.87 112	99.57 58	99.68 44

UniMVSNet_NR-MVSNet		95.71 139	95.15 147	97.40 156	96.84 235	96.97 79	98.74 123	99.24 10	95.16 108	93.88 228	97.72 183	91.68 107	98.31 269	95.81 113	87.25 308	96.92 222

DU-MVS		95.42 163	94.76 173	97.40 156	96.53 250	96.97 79	98.66 143	98.99 28	95.43 88	93.88 228	97.69 184	88.57 176	98.31 269	95.81 113	87.25 308	96.92 222

UniMVSNet (Re)		95.78 135	95.19 146	97.58 141	96.99 226	97.47 63	98.79 112	99.18 16	95.60 81	93.92 227	97.04 240	91.68 107	98.48 238	95.80 115	87.66 303	96.79 240

cascas		94.63 216	93.86 223	96.93 181	96.91 231	94.27 225	96.00 324	98.51 134	85.55 326	94.54 187	96.23 288	84.20 276	98.87 201	95.80 115	96.98 156	97.66 195

Effi-MVS+-dtu		96.29 119	96.56 97	95.51 261	97.89 172	90.22 293	98.80 107	98.10 212	96.57 52	96.45 156	96.66 272	90.81 123	98.91 195	95.72 117	97.99 133	97.40 200

mvs-test1		96.60 106	96.68 94	96.37 230	97.89 172	91.81 269	98.56 156	98.10 212	96.57 52	96.52 142	97.94 162	90.81 123	99.45 136	95.72 117	98.01 132	97.86 187

LPG-MVS_test		95.62 144	95.34 137	96.47 223	97.46 196	93.54 244	98.99 65	98.54 127	94.67 127	94.36 201	98.77 92	85.39 248	99.11 169	95.71 119	94.15 216	96.76 243

LGP-MVS_train				96.47 223	97.46 196	93.54 244		98.54 127	94.67 127	94.36 201	98.77 92	85.39 248	99.11 169	95.71 119	94.15 216	96.76 243

旧先验2							97.57 264		91.30 262	98.67 39			99.80 62	95.70 121

LCM-MVSNet-Re		95.22 179	95.32 140	94.91 285	98.18 155	87.85 323	98.75 119	95.66 332	95.11 110	88.96 304	96.85 265	90.26 133	97.65 301	95.65 122	98.44 118	99.22 107

anonymousdsp		95.42 163	94.91 162	96.94 180	95.10 317	95.90 137	99.14 45	98.41 153	93.75 162	93.16 247	97.46 199	87.50 210	98.41 258	95.63 123	94.03 220	96.50 283

CDPH-MVS		97.94 49	97.49 58	99.28 22	99.47 34	98.44 17	97.91 234	98.67 106	92.57 217	98.77 35	98.85 83	95.93 30	99.72 91	95.56 124	99.69 41	99.68 44

CostFormer		94.95 192	94.73 175	95.60 260	97.28 208	89.06 307	97.53 265	96.89 304	89.66 297	96.82 124	96.72 270	86.05 238	98.95 192	95.53 125	96.13 189	98.79 145

ACMM	93.85 9	95.69 141	95.38 136	96.61 207	97.61 185	93.84 236	98.91 75	98.44 148	95.25 104	94.28 209	98.47 118	86.04 240	99.12 165	95.50 126	93.95 223	96.87 233
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
ACMP	93.49 10	95.34 172	94.98 154	96.43 227	97.67 181	93.48 246	98.73 126	98.44 148	94.94 122	92.53 264	98.53 112	84.50 267	99.14 163	95.48 127	94.00 221	96.66 261
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
testing_2		90.61 300	88.50 307	96.95 179	90.08 340	95.57 148	97.69 255	98.06 219	93.02 201	76.55 342	92.48 338	61.18 349	98.44 248	95.45 128	91.98 250	96.84 236

Test4		92.21 276	90.34 292	97.82 118	92.83 332	95.87 140	97.94 230	98.05 222	94.50 135	82.12 335	94.48 314	59.54 350	98.54 227	95.39 129	98.22 126	99.06 128

TAMVS		97.02 93	96.79 87	97.70 128	98.06 162	95.31 160	98.52 163	98.31 165	93.95 153	97.05 110	98.61 105	93.49 78	98.52 234	95.33 130	97.81 141	99.29 100

BP-MVS														95.30 131

HQP-MVS		95.72 137	95.40 132	96.69 193	97.20 214	94.25 227	98.05 219	98.46 144	96.43 54	94.45 191	97.73 180	86.75 220	98.96 188	95.30 131	94.18 214	96.86 235

WR-MVS		95.15 182	94.46 187	97.22 161	96.67 245	96.45 102	98.21 198	98.81 62	94.15 142	93.16 247	97.69 184	87.51 208	98.30 271	95.29 133	88.62 292	96.90 229

PatchFormer-LS_test		95.47 158	95.27 143	96.08 245	97.59 187	90.66 287	98.10 216	97.34 271	93.98 151	96.08 162	96.15 292	87.65 206	99.12 165	95.27 134	95.24 205	98.44 164

tpmrst		95.63 143	95.69 128	95.44 267	97.54 191	88.54 316	96.97 293	97.56 241	93.50 180	97.52 99	96.93 256	89.49 139	99.16 160	95.25 135	96.42 169	98.64 155

CDS-MVSNet		96.99 94	96.69 92	97.90 113	98.05 163	95.98 116	98.20 199	98.33 164	93.67 174	96.95 112	98.49 116	93.54 77	98.42 251	95.24 136	97.74 145	99.31 95
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
OPM-MVS		95.69 141	95.33 139	96.76 188	96.16 285	94.63 208	98.43 175	98.39 157	96.64 50	95.02 176	98.78 90	85.15 253	99.05 176	95.21 137	94.20 213	96.60 270

OMC-MVS		97.55 68	97.34 64	98.20 95	99.33 45	95.92 134	98.28 193	98.59 117	95.52 85	97.97 72	99.10 51	93.28 81	99.49 131	95.09 138	98.88 97	99.19 110

CANet_DTU		96.96 95	96.55 98	98.21 94	98.17 157	96.07 115	97.98 226	98.21 181	97.24 27	97.13 104	98.93 77	86.88 219	99.91 24	95.00 139	99.37 83	98.66 153

UA-Net		97.96 47	97.62 50	98.98 51	98.86 112	97.47 63	98.89 80	99.08 20	96.67 49	98.72 38	99.54 1	93.15 82	99.81 55	94.87 140	98.83 101	99.65 53

114514_t		96.93 96	96.27 107	98.92 55	99.50 29	97.63 57	98.85 92	98.90 42	84.80 330	97.77 81	99.11 49	92.84 84	99.66 102	94.85 141	99.77 20	99.47 80

XXY-MVS		95.20 181	94.45 189	97.46 151	96.75 240	96.56 96	98.86 91	98.65 113	93.30 194	93.27 244	98.27 139	84.85 258	98.87 201	94.82 142	91.26 261	96.96 219

MG-MVS		97.81 55	97.60 51	98.44 82	99.12 83	95.97 120	97.75 251	98.78 73	96.89 42	98.46 48	99.22 34	93.90 76	99.68 100	94.81 143	99.52 69	99.67 49

EI-MVSNet		95.96 127	95.83 120	96.36 231	97.93 169	93.70 243	98.12 212	98.27 171	93.70 169	95.07 174	99.02 61	92.23 95	98.54 227	94.68 144	93.46 231	96.84 236

IterMVS-LS		95.46 159	95.21 145	96.22 239	98.12 158	93.72 242	98.32 188	98.13 200	93.71 167	94.26 210	97.31 211	92.24 94	98.10 280	94.63 145	90.12 266	96.84 236
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
1314		96.25 123	95.73 122	97.79 119	97.13 220	95.55 151	98.19 203	98.59 117	93.47 181	92.03 277	97.82 175	91.33 116	99.49 131	94.62 146	98.44 118	98.32 174

IS-MVSNet		97.22 85	96.88 83	98.25 93	98.85 114	96.36 106	99.19 35	97.97 224	95.39 90	97.23 102	98.99 67	91.11 119	98.93 193	94.60 147	98.59 111	99.47 80

NR-MVSNet		94.98 190	94.16 202	97.44 152	96.53 250	97.22 73	98.74 123	98.95 33	94.96 119	89.25 302	97.69 184	89.32 143	98.18 277	94.59 148	87.40 305	96.92 222

IB-MVS	91.98 17	93.27 263	91.97 271	97.19 163	97.47 195	93.41 249	97.09 291	95.99 321	93.32 192	92.47 267	95.73 301	78.06 314	99.53 128	94.59 148	82.98 325	98.62 156
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		94.82 200	94.36 192	96.20 240	97.35 205	90.79 284	98.34 183	96.57 315	92.91 206	95.33 172	96.44 282	82.00 290	99.12 165	94.52 150	95.78 201	98.70 149

HY-MVS	93.96 8	96.82 101	96.23 110	98.57 71	98.46 138	97.00 78	98.14 209	98.21 181	93.95 153	96.72 128	97.99 159	91.58 109	99.76 86	94.51 151	96.54 165	98.95 138

Baseline_NR-MVSNet		94.35 230	93.81 225	95.96 247	96.20 280	94.05 231	98.61 148	96.67 312	91.44 253	93.85 230	97.60 192	88.57 176	98.14 278	94.39 152	86.93 311	95.68 309

AdaColmap		97.15 89	96.70 91	98.48 79	99.16 79	96.69 91	98.01 223	98.89 44	94.44 139	96.83 122	98.68 99	90.69 126	99.76 86	94.36 153	99.29 86	98.98 133

1112_ss		96.63 105	96.00 116	98.50 77	98.56 134	96.37 105	98.18 207	98.10 212	92.92 205	94.84 179	98.43 120	92.14 98	99.58 118	94.35 154	96.51 166	99.56 68

CP-MVSNet		94.94 194	94.30 194	96.83 185	96.72 242	95.56 149	99.11 51	98.95 33	93.89 155	92.42 269	97.90 165	87.19 213	98.12 279	94.32 155	88.21 295	96.82 239

CNLPA		97.45 72	97.03 78	98.73 62	99.05 84	97.44 65	98.07 218	98.53 130	95.32 101	96.80 126	98.53 112	93.32 80	99.72 91	94.31 156	99.31 85	99.02 129

testdata				98.26 92	99.20 77	95.36 156		98.68 99	91.89 242	98.60 44	99.10 51	94.44 68	99.82 53	94.27 157	99.44 77	99.58 66

PVSNet	91.96 18	96.35 116	96.15 111	96.96 178	99.17 78	92.05 266	96.08 320	98.68 99	93.69 170	97.75 83	97.80 177	88.86 159	99.69 99	94.26 158	99.01 92	99.15 117

Test_1112_low_res		96.34 117	95.66 130	98.36 87	98.56 134	95.94 124	97.71 253	98.07 217	92.10 237	94.79 183	97.29 212	91.75 106	99.56 121	94.17 159	96.50 167	99.58 66

TranMVSNet+NR-MVSNet		95.14 183	94.48 185	97.11 169	96.45 255	96.36 106	99.03 62	99.03 24	95.04 114	93.58 235	97.93 163	88.27 184	98.03 285	94.13 160	86.90 313	96.95 221

API-MVS		97.41 77	97.25 67	97.91 112	98.70 123	96.80 86	98.82 98	98.69 96	94.53 133	98.11 60	98.28 136	94.50 66	99.57 119	94.12 161	99.49 70	97.37 203

PLC	95.07 4	97.20 86	96.78 88	98.44 82	99.29 58	96.31 110	98.14 209	98.76 77	92.41 228	96.39 157	98.31 135	94.92 56	99.78 79	94.06 162	98.77 104	99.23 106
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
XVG-ACMP-BASELINE		94.54 222	94.14 204	95.75 257	96.55 249	91.65 274	98.11 214	98.44 148	94.96 119	94.22 213	97.90 165	79.18 309	99.11 169	94.05 163	93.85 224	96.48 285

F-COLMAP		97.09 92	96.80 85	97.97 110	99.45 36	94.95 175	98.55 158	98.62 115	93.02 201	96.17 161	98.58 110	94.01 74	99.81 55	93.95 164	98.90 96	99.14 119

MDTV_nov1_ep13_2view						84.26 331	96.89 302		90.97 270	97.90 77		89.89 136		93.91 165		99.18 114

diffmvs		96.32 118	95.74 121	98.07 106	98.26 147	96.14 113	98.53 162	98.23 179	90.10 283	96.88 120	97.73 180	90.16 134	99.15 161	93.90 166	97.85 139	98.91 140

原ACMM1				98.65 67	99.32 48	96.62 92		98.67 106	93.27 195	97.81 80	98.97 68	95.18 50	99.83 47	93.84 167	99.46 75	99.50 74

RPSCF		94.87 196	95.40 132	93.26 313	98.89 109	82.06 338	98.33 184	98.06 219	90.30 279	96.56 136	99.26 30	87.09 214	99.49 131	93.82 168	96.32 176	98.24 175

PAPM_NR		97.46 69	97.11 74	98.50 77	99.50 29	96.41 104	98.63 145	98.60 116	95.18 107	97.06 109	98.06 153	94.26 71	99.57 119	93.80 169	98.87 99	99.52 69

ACMH	92.88 16	94.55 221	93.95 218	96.34 234	97.63 183	93.26 251	98.81 104	98.49 143	93.43 182	89.74 297	98.53 112	81.91 291	99.08 174	93.69 170	93.30 237	96.70 252
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
MAR-MVS		96.91 97	96.40 103	98.45 81	98.69 125	96.90 83	98.66 143	98.68 99	92.40 229	97.07 108	97.96 160	91.54 113	99.75 88	93.68 171	98.92 95	98.69 150
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 99	96.55 98	97.83 116	98.73 120	95.46 153	99.20 33	98.30 168	94.96 119	96.60 135	98.87 82	90.05 135	98.59 222	93.67 172	98.60 110	99.46 84

LS3D		97.16 88	96.66 95	98.68 65	98.53 137	97.19 74	98.93 74	98.90 42	92.83 211	95.99 166	99.37 13	92.12 99	99.87 38	93.67 172	99.57 58	98.97 134

PS-CasMVS		94.67 214	93.99 216	96.71 190	96.68 244	95.26 161	99.13 48	99.03 24	93.68 172	92.33 270	97.95 161	85.35 250	98.10 280	93.59 174	88.16 297	96.79 240

CVMVSNet		95.43 161	96.04 114	93.57 309	97.93 169	83.62 332	98.12 212	98.59 117	95.68 77	96.56 136	99.02 61	87.51 208	97.51 306	93.56 175	97.44 149	99.60 62

OurMVSNet-221017-0		94.21 236	94.00 214	94.85 288	95.60 305	89.22 305	98.89 80	97.43 264	95.29 102	92.18 274	98.52 115	82.86 287	98.59 222	93.46 176	91.76 254	96.74 245

OpenMVS	93.04 13	95.83 133	95.00 152	98.32 89	97.18 217	97.32 67	99.21 32	98.97 29	89.96 286	91.14 285	99.05 60	86.64 222	99.92 15	93.38 177	99.47 72	97.73 191

无先验							97.58 263	98.72 88	91.38 256				99.87 38	93.36 178		99.60 62

1121		97.37 80	96.77 90	99.16 37	99.34 42	97.99 47	98.19 203	98.68 99	90.14 282	98.01 69	98.97 68	94.80 59	99.87 38	93.36 178	99.46 75	99.61 59

gm-plane-assit					95.88 296	87.47 324			89.74 295		96.94 252		99.19 158	93.32 180

WR-MVS_H		95.05 186	94.46 187	96.81 186	96.86 234	95.82 141	99.24 21	99.24 10	93.87 157	92.53 264	96.84 266	90.37 129	98.24 275	93.24 181	87.93 298	96.38 289

tpm		94.13 244	93.80 226	95.12 280	96.50 252	87.91 322	97.44 268	95.89 325	92.62 214	96.37 158	96.30 285	84.13 277	98.30 271	93.24 181	91.66 256	99.14 119

Fast-Effi-MVS+-dtu		95.87 131	95.85 119	95.91 249	97.74 179	91.74 273	98.69 134	98.15 197	95.56 83	94.92 177	97.68 187	88.98 154	98.79 210	93.19 183	97.78 143	97.20 211

pmmvs5		93.65 257	92.97 257	95.68 258	95.49 309	92.37 262	98.20 199	97.28 277	89.66 297	92.58 262	97.26 213	82.14 289	98.09 282	93.18 184	90.95 262	96.58 272

TESTMET0.1,1		94.18 240	93.69 235	95.63 259	96.92 229	89.12 306	96.91 297	94.78 341	93.17 196	94.88 178	96.45 281	78.52 311	98.92 194	93.09 185	98.50 115	98.85 141

test-LLR		95.10 184	94.87 164	95.80 254	96.77 237	89.70 297	96.91 297	95.21 336	95.11 110	94.83 181	95.72 303	87.71 202	98.97 185	93.06 186	98.50 115	98.72 147

test-mter		94.08 246	93.51 246	95.80 254	96.77 237	89.70 297	96.91 297	95.21 336	92.89 207	94.83 181	95.72 303	77.69 316	98.97 185	93.06 186	98.50 115	98.72 147

BH-untuned		95.95 128	95.72 123	96.65 201	98.55 136	92.26 263	98.23 196	97.79 231	93.73 165	94.62 185	98.01 157	88.97 155	99.00 184	93.04 188	98.51 114	98.68 151

EPMVS		94.99 188	94.48 185	96.52 219	97.22 212	91.75 272	97.23 285	91.66 353	94.11 143	97.28 101	96.81 267	85.70 244	98.84 204	93.04 188	97.28 151	98.97 134

pmmvs4		94.69 210	93.99 216	96.81 186	95.74 300	95.94 124	97.40 271	97.67 236	90.42 277	93.37 242	97.59 193	89.08 150	98.20 276	92.97 190	91.67 255	96.30 293

v6		94.83 197	94.21 199	96.69 193	96.36 262	94.85 182	98.87 84	98.11 207	92.46 218	94.44 197	97.05 239	88.76 170	98.57 225	92.95 191	88.92 284	96.65 263

v1neww		94.83 197	94.22 196	96.68 196	96.39 258	94.85 182	98.87 84	98.11 207	92.45 223	94.45 191	97.06 235	88.82 164	98.54 227	92.93 192	88.91 285	96.65 263

v7new		94.83 197	94.22 196	96.68 196	96.39 258	94.85 182	98.87 84	98.11 207	92.45 223	94.45 191	97.06 235	88.82 164	98.54 227	92.93 192	88.91 285	96.65 263

v2v482		94.69 210	94.03 211	96.65 201	96.17 282	94.79 201	98.67 141	98.08 216	92.72 212	94.00 225	97.16 219	87.69 205	98.45 245	92.91 194	88.87 287	96.72 248

Fast-Effi-MVS+		96.28 121	95.70 127	98.03 108	98.29 146	95.97 120	98.58 151	98.25 176	91.74 246	95.29 173	97.23 216	91.03 122	99.15 161	92.90 195	97.96 134	98.97 134

V42		94.78 203	94.14 204	96.70 192	96.33 269	95.22 162	98.97 69	98.09 215	92.32 233	94.31 205	97.06 235	88.39 182	98.55 226	92.90 195	88.87 287	96.34 291

DP-MVS		96.59 108	95.93 117	98.57 71	99.34 42	96.19 112	98.70 132	98.39 157	89.45 301	94.52 188	99.35 19	91.85 105	99.85 43	92.89 197	98.88 97	99.68 44

TDRefinement		91.06 295	89.68 298	95.21 277	85.35 348	91.49 275	98.51 167	97.07 285	91.47 251	88.83 305	97.84 171	77.31 320	99.09 173	92.79 198	77.98 342	95.04 318

ACMH+	92.99 14	94.30 232	93.77 229	95.88 251	97.81 176	92.04 267	98.71 129	98.37 160	93.99 150	90.60 292	98.47 118	80.86 298	99.05 176	92.75 199	92.40 246	96.55 277

divwei89l23v2f112		94.76 204	94.12 207	96.67 199	96.28 275	94.85 182	98.69 134	98.12 202	92.44 225	94.29 208	96.94 252	88.85 161	98.48 238	92.67 200	88.79 291	96.67 258

v1		94.75 206	94.11 208	96.69 193	96.27 277	94.87 180	98.69 134	98.12 202	92.43 226	94.32 204	96.94 252	88.71 173	98.54 227	92.66 201	88.84 290	96.67 258

v1141		94.75 206	94.11 208	96.67 199	96.27 277	94.86 181	98.69 134	98.12 202	92.43 226	94.31 205	96.94 252	88.78 169	98.48 238	92.63 202	88.85 289	96.67 258

test_post1							96.68 310				30.43 362	87.85 199	98.69 213	92.59 203

v148		94.29 233	93.76 231	95.91 249	96.10 286	92.93 257	98.58 151	97.97 224	92.59 216	93.47 241	96.95 250	88.53 179	98.32 267	92.56 204	87.06 310	96.49 284

PEN-MVS		94.42 227	93.73 233	96.49 221	96.28 275	94.84 191	99.17 36	99.00 26	93.51 179	92.23 272	97.83 174	86.10 237	97.90 293	92.55 205	86.92 312	96.74 245

Patchmatch-RL test		91.49 290	90.85 282	93.41 310	91.37 336	84.40 330	92.81 346	95.93 324	91.87 244	87.25 310	94.87 311	88.99 151	96.53 330	92.54 206	82.00 327	99.30 98

IterMVS		94.09 245	93.85 224	94.80 291	97.99 166	90.35 292	97.18 288	98.12 202	93.68 172	92.46 268	97.34 208	84.05 278	97.41 308	92.51 207	91.33 258	96.62 267
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
semantic-postprocess				94.85 288	97.98 168	90.56 290		98.11 207	93.75 162	92.58 262	97.48 198	83.91 280	97.41 308	92.48 208	91.30 259	96.58 272

tpm2		94.19 238	93.76 231	95.46 265	97.23 211	89.04 308	97.31 282	96.85 307	87.08 316	96.21 160	96.79 268	83.75 284	98.74 212	92.43 209	96.23 185	98.59 157

PVSNet_0	88.72 19	91.28 292	90.03 295	95.00 283	97.99 166	87.29 326	94.84 337	98.50 139	92.06 238	89.86 296	95.19 307	79.81 305	99.39 139	92.27 210	69.79 349	98.33 173

gg-mvs-nofinetune		92.21 276	90.58 290	97.13 167	96.75 240	95.09 166	95.85 326	89.40 356	85.43 327	94.50 189	81.98 349	80.80 299	98.40 264	92.16 211	98.33 123	97.88 186

pm-mvs1		93.94 251	93.06 255	96.59 209	96.49 253	95.16 163	98.95 71	98.03 223	92.32 233	91.08 286	97.84 171	84.54 266	98.41 258	92.16 211	86.13 319	96.19 297

K. test v3		92.55 272	91.91 273	94.48 299	95.64 304	89.24 304	99.07 57	94.88 340	94.04 147	86.78 312	97.59 193	77.64 319	97.64 302	92.08 213	89.43 276	96.57 274

GBi-Net		94.49 223	93.80 226	96.56 214	98.21 150	95.00 169	98.82 98	98.18 189	92.46 218	94.09 220	97.07 232	81.16 293	97.95 289	92.08 213	92.14 247	96.72 248

test1		94.49 223	93.80 226	96.56 214	98.21 150	95.00 169	98.82 98	98.18 189	92.46 218	94.09 220	97.07 232	81.16 293	97.95 289	92.08 213	92.14 247	96.72 248

FMVSNet3		94.97 191	94.26 195	97.11 169	98.18 155	96.62 92	98.56 156	98.26 175	93.67 174	94.09 220	97.10 228	84.25 273	98.01 286	92.08 213	92.14 247	96.70 252

Anonymous20240521		94.80 202	94.03 211	97.11 169	96.56 248	96.46 101	99.30 14	98.44 148	92.86 209	91.21 283	97.01 244	89.59 138	98.58 224	92.03 217	89.23 279	96.30 293

PatchmatchNet		95.71 139	95.52 131	96.29 237	97.58 188	90.72 286	96.84 306	97.52 247	94.06 146	97.08 106	96.96 249	89.24 146	98.90 198	92.03 217	98.37 121	99.26 103
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
QAPM		96.29 119	95.40 132	98.96 53	97.85 174	97.60 59	99.23 23	98.93 36	89.76 293	93.11 251	99.02 61	89.11 149	99.93 9	91.99 219	99.62 50	99.34 92

新几何1				99.16 37	99.34 42	98.01 44		98.69 96	90.06 284	98.13 59	98.95 75	94.60 61	99.89 29	91.97 220	99.47 72	99.59 64

v7		94.69 210	94.04 210	96.62 206	96.41 257	94.79 201	98.78 114	98.13 200	91.89 242	94.30 207	97.16 219	88.13 189	98.45 245	91.96 221	89.65 271	96.61 268

MDTV_nov1_ep13			95.40 132		97.48 194	88.34 318	96.85 305	97.29 276	93.74 164	97.48 100	97.26 213	89.18 147	99.05 176	91.92 222	97.43 150

EU-MVSNet		93.66 255	94.14 204	92.25 318	95.96 292	83.38 333	98.52 163	98.12 202	94.69 125	92.61 261	98.13 148	87.36 212	96.39 332	91.82 223	90.00 268	96.98 218

GA-MVS		94.81 201	94.03 211	97.14 166	97.15 219	93.86 235	96.76 308	97.58 240	94.00 149	94.76 184	97.04 240	80.91 296	98.48 238	91.79 224	96.25 184	99.09 123

tfpn1000		95.72 137	95.11 148	97.58 141	99.00 90	95.73 144	99.24 21	95.49 334	94.08 145	96.87 121	97.45 201	85.81 242	99.30 143	91.78 225	96.22 187	97.71 193

PatchMatch-RL		96.59 108	96.03 115	98.27 91	99.31 50	96.51 99	97.91 234	99.06 21	93.72 166	96.92 117	98.06 153	88.50 181	99.65 103	91.77 226	99.00 93	98.66 153

v1144		94.59 219	93.92 219	96.60 208	96.21 279	94.78 203	98.59 149	98.14 199	91.86 245	94.21 214	97.02 242	87.97 193	98.41 258	91.72 227	89.57 272	96.61 268

v8		94.47 225	93.77 229	96.57 213	96.36 262	94.83 193	99.05 58	98.19 186	91.92 241	93.16 247	96.97 248	88.82 164	98.48 238	91.69 228	87.79 301	96.39 288

testdata2													99.89 29	91.65 229

BH-w/o		95.38 167	95.08 150	96.26 238	98.34 143	91.79 270	97.70 254	97.43 264	92.87 208	94.24 212	97.22 217	88.66 174	98.84 204	91.55 230	97.70 146	98.16 177

tfpn_ndepth		95.53 152	94.90 163	97.39 159	98.96 98	95.88 139	99.05 58	95.27 335	93.80 161	96.95 112	96.93 256	85.53 246	99.40 137	91.54 231	96.10 190	96.89 230

v52		94.18 240	93.52 244	96.13 243	95.95 293	94.29 224	99.23 23	98.21 181	91.42 254	92.84 256	96.89 259	87.85 199	98.53 233	91.51 232	87.81 299	95.57 312

V4		94.18 240	93.52 244	96.13 243	95.89 295	94.31 223	99.23 23	98.22 180	91.42 254	92.82 257	96.89 259	87.93 195	98.52 234	91.51 232	87.81 299	95.58 311

LF4IMVS		93.14 268	92.79 260	94.20 304	95.88 296	88.67 313	97.66 258	97.07 285	93.81 160	91.71 279	97.65 188	77.96 315	98.81 208	91.47 234	91.92 252	95.12 315

JIA-IIPM		93.35 260	92.49 265	95.92 248	96.48 254	90.65 288	95.01 333	96.96 296	85.93 323	96.08 162	87.33 345	87.70 204	98.78 211	91.35 235	95.58 202	98.34 172

Patchmatch-test1		95.32 174	94.97 156	96.35 232	97.67 181	91.29 278	97.33 280	97.60 239	94.68 126	96.92 117	96.95 250	83.97 279	98.50 237	91.33 236	98.32 124	99.25 104

FMVSNet2		94.47 225	93.61 239	97.04 173	98.21 150	96.43 103	98.79 112	98.27 171	92.46 218	93.50 240	97.09 230	81.16 293	98.00 287	91.09 237	91.93 251	96.70 252

v144192		94.39 229	93.70 234	96.48 222	96.06 288	94.35 222	98.58 151	98.16 196	91.45 252	94.33 203	97.02 242	87.50 210	98.45 245	91.08 238	89.11 280	96.63 266

tpmvs		94.60 217	94.36 192	95.33 276	97.46 196	88.60 314	96.88 303	97.68 235	91.29 263	93.80 232	96.42 283	88.58 175	99.24 148	91.06 239	96.04 197	98.17 176

LTVRE_ROB	92.95 15	94.60 217	93.90 221	96.68 196	97.41 203	94.42 218	98.52 163	98.59 117	91.69 247	91.21 283	98.35 128	84.87 257	99.04 180	91.06 239	93.44 234	96.60 270
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		96.84 100	96.24 109	98.65 67	98.72 122	96.92 82	97.36 277	98.57 123	93.33 191	96.67 129	97.57 195	94.30 70	99.56 121	91.05 241	98.59 111	99.47 80

SixPastTwentyTwo		93.34 261	92.86 258	94.75 292	95.67 303	89.41 303	98.75 119	96.67 312	93.89 155	90.15 295	98.25 141	80.87 297	98.27 274	90.90 242	90.64 263	96.57 274

COLMAP_ROB	93.27 12	95.33 173	94.87 164	96.71 190	99.29 58	93.24 252	98.58 151	98.11 207	89.92 289	93.57 236	99.10 51	86.37 226	99.79 74	90.78 243	98.10 131	97.09 212
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
pmmvs6		91.77 288	90.63 289	95.17 279	94.69 324	91.24 279	98.67 141	97.92 226	86.14 320	89.62 298	97.56 196	75.79 325	98.34 265	90.75 244	84.56 324	95.94 303

BH-RMVSNet		95.92 130	95.32 140	97.69 129	98.32 145	94.64 207	98.19 203	97.45 262	94.56 132	96.03 164	98.61 105	85.02 254	99.12 165	90.68 245	99.06 91	99.30 98

v748		93.75 254	93.06 255	95.82 253	95.73 301	92.64 260	99.25 20	98.24 178	91.60 249	92.22 273	96.52 279	87.60 207	98.46 243	90.64 246	85.72 320	96.36 290

tpmp4_e23		93.91 252	93.42 251	95.38 273	97.62 184	88.59 315	97.52 266	97.34 271	87.94 312	94.17 217	96.79 268	82.91 286	99.05 176	90.62 247	95.91 198	98.50 160

DTE-MVSNet		93.98 250	93.26 254	96.14 242	96.06 288	94.39 220	99.20 33	98.86 52	93.06 199	91.78 278	97.81 176	85.87 241	97.58 304	90.53 248	86.17 317	96.46 287

conf0.01		95.56 150	94.84 166	97.72 123	98.90 101	95.93 127	99.17 36	95.70 326	93.42 183	96.50 147	97.16 219	86.12 230	99.22 151	90.51 249	96.06 191	98.02 180

conf0.002		95.56 150	94.84 166	97.72 123	98.90 101	95.93 127	99.17 36	95.70 326	93.42 183	96.50 147	97.16 219	86.12 230	99.22 151	90.51 249	96.06 191	98.02 180

thresconf0.02		95.50 153	94.84 166	97.51 145	98.90 101	95.93 127	99.17 36	95.70 326	93.42 183	96.50 147	97.16 219	86.12 230	99.22 151	90.51 249	96.06 191	97.37 203

tfpn_n400		95.50 153	94.84 166	97.51 145	98.90 101	95.93 127	99.17 36	95.70 326	93.42 183	96.50 147	97.16 219	86.12 230	99.22 151	90.51 249	96.06 191	97.37 203

tfpnconf		95.50 153	94.84 166	97.51 145	98.90 101	95.93 127	99.17 36	95.70 326	93.42 183	96.50 147	97.16 219	86.12 230	99.22 151	90.51 249	96.06 191	97.37 203

tfpnview11		95.50 153	94.84 166	97.51 145	98.90 101	95.93 127	99.17 36	95.70 326	93.42 183	96.50 147	97.16 219	86.12 230	99.22 151	90.51 249	96.06 191	97.37 203

v10		94.29 233	93.55 242	96.51 220	96.39 258	94.80 198	98.99 65	98.19 186	91.35 259	93.02 253	96.99 246	88.09 190	98.41 258	90.50 255	88.41 294	96.33 292

ambc				89.49 324	86.66 347	75.78 345	92.66 347	96.72 309		86.55 314	92.50 337	46.01 354	97.90 293	90.32 256	82.09 326	94.80 321

lessismore_v0				94.45 302	94.93 320	88.44 317		91.03 354		86.77 313	97.64 190	76.23 323	98.42 251	90.31 257	85.64 321	96.51 282

v1192		94.32 231	93.58 241	96.53 218	96.10 286	94.45 217	98.50 168	98.17 194	91.54 250	94.19 215	97.06 235	86.95 218	98.43 250	90.14 258	89.57 272	96.70 252

MVS		94.67 214	93.54 243	98.08 104	96.88 233	96.56 96	98.19 203	98.50 139	78.05 344	92.69 259	98.02 155	91.07 121	99.63 108	90.09 259	98.36 122	98.04 179

ADS-MVSNet2		94.58 220	94.40 191	95.11 281	98.00 164	88.74 311	96.04 321	97.30 275	90.15 280	96.47 154	96.64 274	87.89 196	97.56 305	90.08 260	97.06 153	99.02 129

ADS-MVSNet		95.00 187	94.45 189	96.63 204	98.00 164	91.91 268	96.04 321	97.74 234	90.15 280	96.47 154	96.64 274	87.89 196	98.96 188	90.08 260	97.06 153	99.02 129

MSDG		95.93 129	95.30 142	97.83 116	98.90 101	95.36 156	96.83 307	98.37 160	91.32 261	94.43 198	98.73 96	90.27 132	99.60 111	90.05 262	98.82 102	98.52 159

v1921920		94.20 237	93.47 248	96.40 229	95.98 291	94.08 230	98.52 163	98.15 197	91.33 260	94.25 211	97.20 218	86.41 225	98.42 251	90.04 263	89.39 277	96.69 257

dp		94.15 243	93.90 221	94.90 286	97.31 207	86.82 328	96.97 293	97.19 282	91.22 267	96.02 165	96.61 276	85.51 247	99.02 183	90.00 264	94.30 209	98.85 141

CMPMVS	66.06 21	89.70 304	89.67 299	89.78 323	93.19 330	76.56 343	97.00 292	98.35 162	80.97 340	81.57 337	97.75 179	74.75 330	98.61 219	89.85 265	93.63 228	94.17 334
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
testpf		88.74 309	89.09 302	87.69 327	95.78 299	83.16 335	84.05 356	94.13 349	85.22 328	90.30 293	94.39 316	74.92 329	95.80 334	89.77 266	93.28 239	84.10 351

TR-MVS		94.94 194	94.20 200	97.17 165	97.75 178	94.14 229	97.59 262	97.02 290	92.28 235	95.75 168	97.64 190	83.88 281	98.96 188	89.77 266	96.15 188	98.40 165

MS-PatchMatch		93.84 253	93.63 237	94.46 301	96.18 281	89.45 301	97.76 250	98.27 171	92.23 236	92.13 275	97.49 197	79.50 306	98.69 213	89.75 268	99.38 82	95.25 314

ITE_SJBPF				95.44 267	97.42 200	91.32 277		97.50 253	95.09 113	93.59 234	98.35 128	81.70 292	98.88 200	89.71 269	93.39 235	96.12 298

MVP-Stereo		94.28 235	93.92 219	95.35 275	94.95 319	92.60 261	97.97 227	97.65 237	91.61 248	90.68 291	97.09 230	86.32 227	98.42 251	89.70 270	99.34 84	95.02 319
Qingsong Yan: MVP-Stereo: A Parallel Multi-View Patchmatch Stereo Method with Dilation Matching for Photogrammetric Application.
AllTest		95.24 178	94.65 178	96.99 175	99.25 67	93.21 253	98.59 149	98.18 189	91.36 257	93.52 238	98.77 92	84.67 259	99.72 91	89.70 270	97.87 137	98.02 180

TestCases				96.99 175	99.25 67	93.21 253		98.18 189	91.36 257	93.52 238	98.77 92	84.67 259	99.72 91	89.70 270	97.87 137	98.02 180

GG-mvs-BLEND				96.59 209	96.34 265	94.98 172	96.51 318	88.58 357		93.10 252	94.34 317	80.34 304	98.05 284	89.53 273	96.99 155	96.74 245

USDC		93.33 262	92.71 261	95.21 277	96.83 236	90.83 283	96.91 297	97.50 253	93.84 158	90.72 290	98.14 147	77.69 316	98.82 207	89.51 274	93.21 240	95.97 302

v7n		94.19 238	93.43 249	96.47 223	95.90 294	94.38 221	99.26 18	98.34 163	91.99 239	92.76 258	97.13 227	88.31 183	98.52 234	89.48 275	87.70 302	96.52 280

PM-MVS		87.77 312	86.55 314	91.40 321	91.03 338	83.36 334	96.92 295	95.18 338	91.28 264	86.48 315	93.42 320	53.27 351	96.74 324	89.43 276	81.97 328	94.11 335

FMVSNet1		93.19 267	92.07 270	96.56 214	97.54 191	95.00 169	98.82 98	98.18 189	90.38 278	92.27 271	97.07 232	73.68 334	97.95 289	89.36 277	91.30 259	96.72 248

tpm cat1		93.36 259	92.80 259	95.07 282	97.58 188	87.97 321	96.76 308	97.86 229	82.17 338	93.53 237	96.04 295	86.13 229	99.13 164	89.24 278	95.87 199	98.10 178

UnsupCasMVSNet_eth		90.99 296	89.92 297	94.19 305	94.08 327	89.83 295	97.13 290	98.67 106	93.69 170	85.83 318	96.19 291	75.15 327	96.74 324	89.14 279	79.41 336	96.00 301

v1240		94.06 248	93.29 253	96.34 234	96.03 290	93.90 234	98.44 173	98.17 194	91.18 268	94.13 219	97.01 244	86.05 238	98.42 251	89.13 280	89.50 275	96.70 252

view600		95.60 146	94.93 158	97.62 135	99.05 84	94.85 182	99.09 53	97.01 292	95.36 95	96.52 142	97.37 204	84.55 262	99.59 112	89.07 281	96.39 170	98.40 165

view800		95.60 146	94.93 158	97.62 135	99.05 84	94.85 182	99.09 53	97.01 292	95.36 95	96.52 142	97.37 204	84.55 262	99.59 112	89.07 281	96.39 170	98.40 165

conf0.05thres1000		95.60 146	94.93 158	97.62 135	99.05 84	94.85 182	99.09 53	97.01 292	95.36 95	96.52 142	97.37 204	84.55 262	99.59 112	89.07 281	96.39 170	98.40 165

tfpn		95.60 146	94.93 158	97.62 135	99.05 84	94.85 182	99.09 53	97.01 292	95.36 95	96.52 142	97.37 204	84.55 262	99.59 112	89.07 281	96.39 170	98.40 165

tmp_tt		68.90 328	66.97 328	74.68 343	50.78 363	59.95 359	87.13 352	83.47 362	38.80 358	62.21 352	96.23 288	64.70 347	76.91 361	88.91 285	30.49 358	87.19 348

v18		92.10 278	90.97 278	95.50 262	96.34 265	94.85 182	98.82 98	97.52 247	89.99 285	85.31 323	93.26 322	88.90 158	96.92 315	88.82 286	79.77 334	94.73 322

v17		92.08 279	90.94 279	95.48 264	96.34 265	94.83 193	98.81 104	97.52 247	89.95 287	85.32 321	93.24 323	88.91 157	96.91 316	88.76 287	79.63 335	94.71 324

pmmvs-eth3d		90.36 301	89.05 304	94.32 303	91.10 337	92.12 264	97.63 261	96.95 297	88.86 307	84.91 331	93.13 324	78.32 312	96.74 324	88.70 288	81.81 329	94.09 336

v16		92.08 279	90.94 279	95.49 263	96.38 261	94.84 191	98.81 104	97.51 250	89.94 288	85.25 324	93.28 321	88.86 159	96.91 316	88.70 288	79.78 333	94.72 323

thres600view7		95.49 157	94.77 172	97.67 131	98.98 94	95.02 168	98.85 92	96.90 300	95.38 91	96.63 131	96.90 258	84.29 269	99.59 112	88.65 290	96.33 175	98.40 165

tfpn111		95.43 161	94.74 174	97.51 145	98.98 94	94.92 176	98.87 84	96.90 300	95.38 91	96.61 132	96.88 261	84.29 269	99.59 112	88.43 291	96.32 176	98.02 180

v15		91.94 281	90.77 283	95.43 269	96.31 273	94.83 193	98.77 115	97.50 253	89.92 289	85.13 325	93.08 326	88.76 170	96.86 318	88.40 292	79.10 337	94.61 328

V14		91.93 282	90.76 284	95.42 272	96.33 269	94.81 197	98.77 115	97.51 250	89.86 291	85.09 326	93.13 324	88.80 168	96.83 320	88.32 293	79.06 339	94.60 329

V9		91.91 283	90.73 285	95.45 266	96.32 272	94.80 198	98.77 115	97.50 253	89.81 292	85.03 328	93.08 326	88.76 170	96.86 318	88.24 294	79.03 340	94.69 325

v12		91.89 284	90.70 286	95.43 269	96.31 273	94.80 198	98.76 118	97.50 253	89.76 293	84.95 329	93.00 329	88.82 164	96.82 322	88.23 295	79.00 341	94.68 327

v13		91.88 285	90.69 287	95.43 269	96.33 269	94.78 203	98.75 119	97.50 253	89.68 296	84.93 330	92.98 330	88.84 162	96.83 320	88.14 296	79.09 338	94.69 325

conf200view11		95.40 166	94.70 176	97.50 150	98.98 94	94.92 176	98.87 84	96.90 300	95.38 91	96.61 132	96.88 261	84.29 269	99.56 121	88.11 297	96.29 178	98.02 180

thres100view900		95.38 167	94.70 176	97.41 154	98.98 94	94.92 176	98.87 84	96.90 300	95.38 91	96.61 132	96.88 261	84.29 269	99.56 121	88.11 297	96.29 178	97.76 188

tfpn200view9		95.32 174	94.62 179	97.43 153	98.94 99	94.98 172	98.68 138	96.93 298	95.33 99	96.55 138	96.53 277	84.23 274	99.56 121	88.11 297	96.29 178	97.76 188

thres400		95.38 167	94.62 179	97.65 134	98.94 99	94.98 172	98.68 138	96.93 298	95.33 99	96.55 138	96.53 277	84.23 274	99.56 121	88.11 297	96.29 178	98.40 165

our_test_3		93.65 257	93.30 252	94.69 293	95.45 311	89.68 299	96.91 297	97.65 237	91.97 240	91.66 280	96.88 261	89.67 137	97.93 292	88.02 301	91.49 257	96.48 285

thres200		95.25 177	94.57 181	97.28 160	98.81 116	94.92 176	98.20 199	97.11 283	95.24 106	96.54 140	96.22 290	84.58 261	99.53 128	87.93 302	96.50 167	97.39 201

EG-PatchMatch MVS		91.13 293	90.12 294	94.17 306	94.73 323	89.00 309	98.13 211	97.81 230	89.22 305	85.32 321	96.46 280	67.71 343	98.42 251	87.89 303	93.82 225	95.08 317

CR-MVSNet		94.76 204	94.15 203	96.59 209	97.00 224	93.43 247	94.96 334	97.56 241	92.46 218	96.93 115	96.24 286	88.15 187	97.88 297	87.38 304	96.65 161	98.46 162

v11		91.85 286	90.68 288	95.36 274	96.34 265	94.74 205	98.80 107	97.43 264	89.60 299	85.09 326	93.03 328	88.53 179	96.75 323	87.37 305	79.96 332	94.58 330

Patchmtry		93.22 265	92.35 267	95.84 252	96.77 237	93.09 256	94.66 340	97.56 241	87.37 315	92.90 255	96.24 286	88.15 187	97.90 293	87.37 305	90.10 267	96.53 279

test0.0.03 1		94.08 246	93.51 246	95.80 254	95.53 308	92.89 258	97.38 273	95.97 322	95.11 110	92.51 266	96.66 272	87.71 202	96.94 314	87.03 307	93.67 226	97.57 196

TinyColmap		92.31 275	91.53 274	94.65 295	96.92 229	89.75 296	96.92 295	96.68 311	90.45 276	89.62 298	97.85 170	76.06 324	98.81 208	86.74 308	92.51 245	95.41 313

MIMVSNet		93.26 264	92.21 269	96.41 228	97.73 180	93.13 255	95.65 329	97.03 289	91.27 265	94.04 223	96.06 294	75.33 326	97.19 311	86.56 309	96.23 185	98.92 139

TransMVSNet (Re)		92.67 271	91.51 275	96.15 241	96.58 247	94.65 206	98.90 76	96.73 308	90.86 271	89.46 300	97.86 168	85.62 245	98.09 282	86.45 310	81.12 330	95.71 308

DSMNet-mixed		92.52 273	92.58 264	92.33 317	94.15 326	82.65 336	98.30 191	94.26 346	89.08 306	92.65 260	95.73 301	85.01 255	95.76 335	86.24 311	97.76 144	98.59 157

testgi		93.06 269	92.45 266	94.88 287	96.43 256	89.90 294	98.75 119	97.54 246	95.60 81	91.63 281	97.91 164	74.46 332	97.02 313	86.10 312	93.67 226	97.72 192

YYNet1		90.70 299	89.39 300	94.62 296	94.79 322	90.65 288	97.20 286	97.46 260	87.54 314	72.54 346	95.74 300	86.51 223	96.66 328	86.00 313	86.76 315	96.54 278

MDA-MVSNet_test_wron		90.71 298	89.38 301	94.68 294	94.83 321	90.78 285	97.19 287	97.46 260	87.60 313	72.41 347	95.72 303	86.51 223	96.71 327	85.92 314	86.80 314	96.56 276

UnsupCasMVSNet_bld		87.17 313	85.12 316	93.31 312	91.94 334	88.77 310	94.92 336	98.30 168	84.30 332	82.30 334	90.04 342	63.96 348	97.25 310	85.85 315	74.47 348	93.93 339

EPNet_dtu		95.21 180	94.95 157	95.99 246	96.17 282	90.45 291	98.16 208	97.27 278	96.77 44	93.14 250	98.33 133	90.34 130	98.42 251	85.57 316	98.81 103	99.09 123
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
FMVSNet5		91.81 287	90.92 281	94.49 298	97.21 213	92.09 265	98.00 225	97.55 245	89.31 304	90.86 289	95.61 306	74.48 331	95.32 337	85.57 316	89.70 270	96.07 300

tfpnnormal		93.66 255	92.70 262	96.55 217	96.94 228	95.94 124	98.97 69	99.19 15	91.04 269	91.38 282	97.34 208	84.94 256	98.61 219	85.45 318	89.02 283	95.11 316

Patchmatch-test		94.42 227	93.68 236	96.63 204	97.60 186	91.76 271	94.83 338	97.49 259	89.45 301	94.14 218	97.10 228	88.99 151	98.83 206	85.37 319	98.13 130	99.29 100

ppachtmachnet_test		93.22 265	92.63 263	94.97 284	95.45 311	90.84 282	96.88 303	97.88 228	90.60 273	92.08 276	97.26 213	88.08 191	97.86 299	85.12 320	90.33 265	96.22 295

PCF-MVS	93.45 11	94.68 213	93.43 249	98.42 85	98.62 131	96.77 88	95.48 330	98.20 184	84.63 331	93.34 243	98.32 134	88.55 178	99.81 55	84.80 321	98.96 94	98.68 151
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
MDA-MVSNet-bldmvs		89.97 303	88.35 309	94.83 290	95.21 316	91.34 276	97.64 259	97.51 250	88.36 310	71.17 348	96.13 293	79.22 308	96.63 329	83.65 322	86.27 316	96.52 280

MVS-HIRNet		89.46 306	88.40 308	92.64 315	97.58 188	82.15 337	94.16 344	93.05 352	75.73 346	90.90 288	82.52 348	79.42 307	98.33 266	83.53 323	98.68 105	97.43 198

new-patchmatchnet		88.50 311	87.45 312	91.67 320	90.31 339	85.89 329	97.16 289	97.33 274	89.47 300	83.63 333	92.77 334	76.38 322	95.06 339	82.70 324	77.29 343	94.06 337

PAPM		94.95 192	94.00 214	97.78 120	97.04 223	95.65 145	96.03 323	98.25 176	91.23 266	94.19 215	97.80 177	91.27 117	98.86 203	82.61 325	97.61 147	98.84 143

LCM-MVSNet		78.70 321	76.24 325	86.08 331	77.26 358	71.99 351	94.34 342	96.72 309	61.62 352	76.53 343	89.33 343	33.91 361	92.78 347	81.85 326	74.60 347	93.46 340

new_pmnet		90.06 302	89.00 305	93.22 314	94.18 325	88.32 319	96.42 319	96.89 304	86.19 319	85.67 320	93.62 319	77.18 321	97.10 312	81.61 327	89.29 278	94.23 333

pmmvs3		86.67 315	84.86 317	92.11 319	88.16 343	87.19 327	96.63 311	94.75 342	79.88 342	87.22 311	92.75 335	66.56 345	95.20 338	81.24 328	76.56 345	93.96 338

N_pmnet		87.12 314	87.77 311	85.17 334	95.46 310	61.92 357	97.37 275	70.66 364	85.83 324	88.73 306	96.04 295	85.33 252	97.76 300	80.02 329	90.48 264	95.84 304

TAPA-MVS	93.98 7	95.35 171	94.56 182	97.74 122	99.13 82	94.83 193	98.33 184	98.64 114	86.62 317	96.29 159	98.61 105	94.00 75	99.29 145	80.00 330	99.41 79	99.09 123
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
DeepMVS_CX				86.78 330	97.09 222	72.30 350		95.17 339	75.92 345	84.34 332	95.19 307	70.58 339	95.35 336	79.98 331	89.04 282	92.68 342

Anonymous20231206		91.66 289	91.10 277	93.33 311	94.02 328	87.35 325	98.58 151	97.26 279	90.48 274	90.16 294	96.31 284	83.83 283	96.53 330	79.36 332	89.90 269	96.12 298

test20.03		90.89 297	90.38 291	92.43 316	93.48 329	88.14 320	98.33 184	97.56 241	93.40 189	87.96 308	96.71 271	80.69 300	94.13 341	79.15 333	86.17 317	95.01 320

PatchT		93.06 269	91.97 271	96.35 232	96.69 243	92.67 259	94.48 341	97.08 284	86.62 317	97.08 106	92.23 340	87.94 194	97.90 293	78.89 334	96.69 159	98.49 161

MIMVSNet1		89.67 305	88.28 310	93.82 307	92.81 333	91.08 281	98.01 223	97.45 262	87.95 311	87.90 309	95.87 299	67.63 344	94.56 340	78.73 335	88.18 296	95.83 305

test_0402		91.32 291	90.27 293	94.48 299	96.60 246	91.12 280	98.50 168	97.22 281	86.10 321	88.30 307	96.98 247	77.65 318	97.99 288	78.13 336	92.94 242	94.34 332

OpenMVS_ROB	86.42 20	89.00 307	87.43 313	93.69 308	93.08 331	89.42 302	97.91 234	96.89 304	78.58 343	85.86 317	94.69 313	69.48 340	98.29 273	77.13 337	93.29 238	93.36 341

testus		88.91 308	89.08 303	88.40 326	91.39 335	76.05 344	96.56 314	96.48 316	89.38 303	89.39 301	95.17 309	70.94 338	93.56 344	77.04 338	95.41 203	95.61 310

RPMNet		92.52 273	91.17 276	96.59 209	97.00 224	93.43 247	94.96 334	97.26 279	82.27 337	96.93 115	92.12 341	86.98 217	97.88 297	76.32 339	96.65 161	98.46 162

Anonymous20231211		83.69 318	81.50 320	90.26 322	89.23 342	80.10 340	97.97 227	97.06 287	72.79 348	82.05 336	92.57 336	50.28 352	96.32 333	76.15 340	75.38 346	94.37 331

test2356		88.68 310	88.61 306	88.87 325	89.90 341	78.23 341	95.11 332	96.66 314	88.66 309	89.06 303	94.33 318	73.14 336	92.56 348	75.56 341	95.11 206	95.81 306

LP		91.12 294	89.99 296	94.53 297	96.35 264	88.70 312	93.86 345	97.35 270	84.88 329	90.98 287	94.77 312	84.40 268	97.43 307	75.41 342	91.89 253	97.47 197

PMMVS2		77.95 323	75.44 326	85.46 332	82.54 350	74.95 349	94.23 343	93.08 351	72.80 347	74.68 344	87.38 344	36.36 359	91.56 350	73.95 343	63.94 350	89.87 344

no-one		74.41 325	70.76 327	85.35 333	79.88 353	76.83 342	94.68 339	94.22 347	80.33 341	63.81 351	79.73 352	35.45 360	93.36 345	71.78 344	36.99 357	85.86 350

test1235678		86.26 316	85.81 315	87.62 328	86.97 346	75.00 348	96.55 316	96.32 319	86.08 322	81.32 338	92.98 330	73.10 337	92.05 349	71.64 345	87.32 306	95.81 306

test12356		83.47 319	83.37 319	83.78 335	84.43 349	70.09 353	95.12 331	95.60 333	82.98 333	78.89 341	92.43 339	64.99 346	91.41 351	70.36 346	85.55 322	89.82 345

FPMVS		77.62 324	77.14 322	79.05 339	79.25 354	60.97 358	95.79 327	95.94 323	65.96 349	67.93 350	94.40 315	37.73 358	88.88 354	68.83 347	88.46 293	87.29 347

1111		84.94 317	84.30 318	86.86 329	87.59 344	75.10 346	96.63 311	96.43 317	82.53 335	80.75 339	92.91 332	68.94 341	93.79 342	68.24 348	84.66 323	91.70 343

.test1245		73.05 326	76.31 324	63.27 347	87.59 344	75.10 346	96.63 311	96.43 317	82.53 335	80.75 339	92.91 332	68.94 341	93.79 342	68.24 348	12.72 360	20.91 360

Gipuma		78.40 322	76.75 323	83.38 336	95.54 307	80.43 339	79.42 357	97.40 267	64.67 350	73.46 345	80.82 351	45.65 355	93.14 346	66.32 350	87.43 304	76.56 356
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
testmv		78.74 320	77.35 321	82.89 337	78.16 357	69.30 354	95.87 325	94.65 343	81.11 339	70.98 349	87.11 346	46.31 353	90.42 352	65.28 351	76.72 344	88.95 346

wuykxyi23d		63.73 333	58.86 335	78.35 340	67.62 360	67.90 355	86.56 353	87.81 359	58.26 353	42.49 359	70.28 357	11.55 366	85.05 355	63.66 352	41.50 353	82.11 353

PNet_i23d		67.70 329	65.07 330	75.60 341	78.61 355	59.61 360	89.14 351	88.24 358	61.83 351	52.37 355	80.89 350	18.91 363	84.91 356	62.70 353	52.93 352	82.28 352

ANet_high		69.08 327	65.37 329	80.22 338	65.99 361	71.96 352	90.91 350	90.09 355	82.62 334	49.93 357	78.39 353	29.36 362	81.75 357	62.49 354	38.52 356	86.95 349

PMVS	61.03 23	65.95 330	63.57 332	73.09 344	57.90 362	51.22 363	85.05 355	93.93 350	54.45 354	44.32 358	83.57 347	13.22 364	89.15 353	58.68 355	81.00 331	78.91 355
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
MVE	62.14 22	63.28 334	59.38 334	74.99 342	74.33 359	65.47 356	85.55 354	80.50 363	52.02 356	51.10 356	75.00 356	10.91 368	80.50 358	51.60 356	53.40 351	78.99 354
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
E-PMN		64.94 331	64.25 331	67.02 345	82.28 351	59.36 361	91.83 349	85.63 360	52.69 355	60.22 353	77.28 354	41.06 357	80.12 359	46.15 357	41.14 354	61.57 358

EMVS		64.07 332	63.26 333	66.53 346	81.73 352	58.81 362	91.85 348	84.75 361	51.93 357	59.09 354	75.13 355	43.32 356	79.09 360	42.03 358	39.47 355	61.69 357

wuyk23d		30.17 336	30.18 338	30.16 349	78.61 355	43.29 364	66.79 358	14.21 365	17.31 359	14.82 362	11.93 363	11.55 366	41.43 362	37.08 359	19.30 359	5.76 362

test123		20.95 339	23.72 340	12.64 350	13.54 365	8.19 365	96.55 316	6.13 367	7.48 361	16.74 361	37.98 360	12.97 365	6.05 363	16.69 360	5.43 362	23.68 359

testmvs		21.48 338	24.95 339	11.09 351	14.89 364	6.47 366	96.56 314	9.87 366	7.55 360	17.93 360	39.02 359	9.43 369	5.90 364	16.56 361	12.72 360	20.91 360

cdsmvs_eth3d_5k		23.98 337	31.98 337	0.00 352	0.00 366	0.00 367	0.00 359	98.59 117	0.00 362	0.00 363	98.61 105	90.60 127	0.00 365	0.00 362	0.00 363	0.00 363

pcd_1.5k_mvsjas		7.88 341	10.50 342	0.00 352	0.00 366	0.00 367	0.00 359	0.00 368	0.00 362	0.00 363	0.00 364	94.51 63	0.00 365	0.00 362	0.00 363	0.00 363

pcd1.5k->3k		39.42 335	41.78 336	32.35 348	96.17 282	0.00 367	0.00 359	98.54 127	0.00 362	0.00 363	0.00 364	87.78 201	0.00 365	0.00 362	93.56 230	97.06 213

sosnet-low-res		0.00 342	0.00 343	0.00 352	0.00 366	0.00 367	0.00 359	0.00 368	0.00 362	0.00 363	0.00 364	0.00 370	0.00 365	0.00 362	0.00 363	0.00 363

sosnet		0.00 342	0.00 343	0.00 352	0.00 366	0.00 367	0.00 359	0.00 368	0.00 362	0.00 363	0.00 364	0.00 370	0.00 365	0.00 362	0.00 363	0.00 363

uncertanet		0.00 342	0.00 343	0.00 352	0.00 366	0.00 367	0.00 359	0.00 368	0.00 362	0.00 363	0.00 364	0.00 370	0.00 365	0.00 362	0.00 363	0.00 363

Regformer		0.00 342	0.00 343	0.00 352	0.00 366	0.00 367	0.00 359	0.00 368	0.00 362	0.00 363	0.00 364	0.00 370	0.00 365	0.00 362	0.00 363	0.00 363

ab-mvs-re		8.20 340	10.94 341	0.00 352	0.00 366	0.00 367	0.00 359	0.00 368	0.00 362	0.00 363	98.43 120	0.00 370	0.00 365	0.00 362	0.00 363	0.00 363

uanet		0.00 342	0.00 343	0.00 352	0.00 366	0.00 367	0.00 359	0.00 368	0.00 362	0.00 363	0.00 364	0.00 370	0.00 365	0.00 362	0.00 363	0.00 363

GSMVS																99.20 108

test_part2					99.63 21	99.18 1				99.27 7

test_part1								98.84 54				97.38 2			99.78 15	99.76 20

sam_mvs1												89.45 140				99.20 108

sam_mvs												88.99 151

MTGPA								98.74 81

test_post											31.83 361	88.83 163	98.91 195

patchmatchnet-post											95.10 310	89.42 141	98.89 199

MTMP							98.89 80	94.14 348

TEST9					99.31 50	98.50 15	97.92 231	98.73 86	92.63 213	97.74 84	98.68 99	96.20 15	99.80 62

test_8					99.29 58	98.44 17	97.89 239	98.72 88	92.98 203	97.70 87	98.66 102	96.20 15	99.80 62

agg_prior					99.30 55	98.38 20		98.72 88		97.57 96			99.81 55

test_prior4						98.01 44	97.86 242

test_prior				99.19 30	99.31 50	98.22 33		98.84 54					99.70 96			99.65 53

新几何2							97.64 259

旧先验1					99.29 58	97.48 62		98.70 95			99.09 55	95.56 38			99.47 72	99.61 59

原ACMM2							97.67 257

test222					99.23 73	97.17 75	97.40 271	98.66 109	88.68 308	98.05 63	98.96 73	94.14 72			99.53 68	99.61 59

segment_acmp												96.85 5

testdata1							97.32 281		96.34 59

test12				99.18 34	99.16 79	98.19 35		98.53 130		98.07 62		95.13 52	99.72 91		99.56 64	99.63 58

plane_prior7					97.42 200	94.63 208

plane_prior6					97.35 205	94.61 211						87.09 214

plane_prior4											98.28 136

plane_prior3						94.61 211			97.02 39	95.34 170

plane_prior2							98.80 107		97.28 21

plane_prior1					97.37 204

plane_prior						94.60 213	98.44 173		96.74 46						94.22 212

n2								0.00 368

nn								0.00 368

door-mid								94.37 345

test11								98.66 109

door								94.64 344

HQP5-MVS						94.25 227

HQP-NCC					97.20 214		98.05 219		96.43 54	94.45 191

ACMP_Plane					97.20 214		98.05 219		96.43 54	94.45 191

HQP4-MVS										94.45 191			98.96 188			96.87 233

HQP3-MVS								98.46 144							94.18 214

HQP2-MVS												86.75 220

NP-MVS					97.28 208	94.51 216					97.73 180

ACMMP++_ref															92.97 241

ACMMP++															93.61 229

Test By Simon												94.64 60

High-resolution multi-view benchmark