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.

CNVR-MVS		98.46 1	98.38 1	98.72 3	99.80 4	96.19 9	99.80 7	97.99 51	97.05 3	99.41 1	99.59 2	92.89 11	100.00 1	98.99 6	99.90 4	99.96 4

SMA-MVS		97.21 13	96.98 16	97.91 21	99.30 44	93.93 48	99.16 58	97.58 98	89.53 107	99.35 2	99.52 3	90.24 39	99.99 4	98.32 21	99.77 20	99.82 22

TSAR-MVS + MP.		97.44 10	97.46 9	97.39 40	99.12 53	93.49 57	98.52 136	97.50 114	94.46 17	98.99 3	98.64 76	91.58 16	99.08 115	98.49 17	99.83 12	99.60 59

PS-MVSNAJ		96.87 25	96.40 30	98.29 11	97.35 108	97.29 1	99.03 77	97.11 148	95.83 9	98.97 4	99.14 29	82.48 146	99.60 77	98.60 11	99.08 60	98.00 154

旧先验2							98.67 118		85.75 193	98.96 5			98.97 118	93.84 86

xiu_mvs_v2_base		96.66 29	96.17 38	98.11 17	97.11 116	96.96 2	99.01 80	97.04 156	95.51 13	98.86 6	99.11 35	82.19 152	99.36 100	98.59 13	98.14 86	98.00 154

NCCC		98.12 3	98.11 3	98.13 15	99.76 6	94.46 39	99.81 5	97.88 57	96.54 4	98.84 7	99.46 6	92.55 13	99.98 9	98.25 23	99.93 1	99.94 6

SD-MVS		97.51 8	97.40 11	97.81 24	99.01 59	93.79 51	99.33 49	97.38 129	93.73 29	98.83 8	99.02 42	90.87 30	99.88 35	98.69 10	99.74 21	99.77 35

HSP-MVS		97.73 5	98.15 2	96.44 92	99.54 27	90.14 128	99.41 38	97.47 117	95.46 14	98.60 9	99.19 19	95.71 4	99.49 87	98.15 24	99.85 9	99.69 47

APD-MVS		96.95 22	96.72 24	97.63 28	99.51 34	93.58 53	99.16 58	97.44 122	90.08 98	98.59 10	99.07 36	89.06 48	99.42 95	97.92 26	99.66 29	99.88 15
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
testdata				95.26 132	98.20 81	87.28 188		97.60 95	85.21 201	98.48 11	99.15 27	88.15 63	98.72 128	90.29 123	99.45 47	99.78 30

TEST9					99.57 23	93.17 60	99.38 40	97.66 83	89.57 105	98.39 12	99.18 21	90.88 29	99.66 66

train_agg		97.20 14	97.08 14	97.57 32	99.57 23	93.17 60	99.38 40	97.66 83	90.18 93	98.39 12	99.18 21	90.94 27	99.66 66	98.58 14	99.85 9	99.88 15

test_8					99.55 26	93.07 64	99.37 43	97.64 88	90.18 93	98.36 14	99.19 19	90.94 27	99.64 72

agg_prior3		97.09 18	96.97 17	97.45 35	99.56 25	92.79 71	99.36 44	97.67 82	89.59 103	98.36 14	99.16 25	90.57 34	99.68 63	98.58 14	99.85 9	99.88 15

HPM-MVS++		97.72 6	97.59 7	98.14 14	99.53 33	94.76 30	99.19 53	97.75 73	95.66 11	98.21 16	99.29 10	91.10 19	99.99 4	97.68 29	99.87 5	99.68 48

test_part2					99.54 27	95.42 14				98.13 17

ESAPD		97.97 4	97.82 6	98.43 10	99.54 27	95.42 14	99.43 33	97.69 79	92.81 44	98.13 17	99.48 4	93.96 6	99.97 14	99.52 1	99.83 12	99.90 9

SteuartSystems-ACMMP		97.25 11	97.34 12	97.01 51	97.38 107	91.46 91	99.75 8	97.66 83	94.14 21	98.13 17	99.26 11	92.16 14	99.66 66	97.91 27	99.64 31	99.90 9
Skip Steuart: Steuart Systems R&D Blog.
test_prior3		97.07 19	97.09 13	97.01 51	99.58 19	91.77 82	99.57 19	97.57 102	91.43 70	98.12 20	98.97 48	90.43 36	99.49 87	98.33 19	99.81 15	99.79 26

test_prior2							99.57 19		91.43 70	98.12 20	98.97 48	90.43 36		98.33 19	99.81 15

PHI-MVS		96.65 30	96.46 29	97.21 45	99.34 40	91.77 82	99.70 10	98.05 47	86.48 186	98.05 22	99.20 18	89.33 46	99.96 18	98.38 18	99.62 35	99.90 9

MVSFormer		94.71 74	94.08 74	96.61 85	95.05 182	94.87 22	97.77 199	96.17 201	86.84 180	98.04 23	98.52 82	85.52 103	95.99 261	89.83 126	98.97 65	98.96 99

lupinMVS		96.32 41	95.94 43	97.44 36	95.05 182	94.87 22	99.86 2	96.50 180	93.82 27	98.04 23	98.77 65	85.52 103	98.09 147	96.98 38	98.97 65	99.37 71

APDe-MVS		97.53 7	97.47 8	97.70 26	99.58 19	93.63 52	99.56 21	97.52 109	93.59 32	98.01 25	99.12 32	90.80 32	99.55 79	99.26 4	99.79 17	99.93 7

ACMMP_Plus		96.59 31	96.18 36	97.81 24	98.82 69	93.55 54	98.88 97	97.59 96	90.66 80	97.98 26	99.14 29	86.59 90	100.00 1	96.47 45	99.46 45	99.89 14

agg_prior1		97.12 16	97.03 15	97.38 41	99.54 27	92.66 72	99.35 46	97.64 88	90.38 88	97.98 26	99.17 23	90.84 31	99.61 75	98.57 16	99.78 19	99.87 19

agg_prior					99.54 27	92.66 72		97.64 88		97.98 26			99.61 75

CDPH-MVS		96.56 32	96.18 36	97.70 26	99.59 18	93.92 49	99.13 69	97.44 122	89.02 120	97.90 29	99.22 16	88.90 51	99.49 87	94.63 78	99.79 17	99.68 48

EPNet		96.82 26	96.68 26	97.25 44	98.65 73	93.10 63	99.48 26	98.76 18	96.54 4	97.84 30	98.22 94	87.49 72	99.66 66	95.35 65	97.78 92	99.00 94
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
MSLP-MVS++		97.50 9	97.45 10	97.63 28	99.65 13	93.21 59	99.70 10	98.13 45	94.61 16	97.78 31	99.46 6	89.85 41	99.81 53	97.97 25	99.91 3	99.88 15

test12				97.83 23	99.33 43	94.45 40		97.55 105		97.56 32		88.60 54	99.50 86		99.71 27	99.55 61

MVS_0304		96.12 46	95.26 56	98.69 4	98.44 78	96.54 7	99.70 10	96.89 165	95.76 10	97.53 33	99.12 32	72.42 231	99.93 25	98.75 8	98.69 77	99.61 58

xiu_mvs_v1_base_debu		94.73 71	93.98 76	96.99 54	95.19 172	95.24 17	98.62 124	96.50 180	92.99 37	97.52 34	98.83 62	72.37 232	99.15 109	97.03 34	96.74 104	96.58 192

xiu_mvs_v1_base		94.73 71	93.98 76	96.99 54	95.19 172	95.24 17	98.62 124	96.50 180	92.99 37	97.52 34	98.83 62	72.37 232	99.15 109	97.03 34	96.74 104	96.58 192

xiu_mvs_v1_base_debi		94.73 71	93.98 76	96.99 54	95.19 172	95.24 17	98.62 124	96.50 180	92.99 37	97.52 34	98.83 62	72.37 232	99.15 109	97.03 34	96.74 104	96.58 192

DeepPCF-MVS	93.56 1	96.55 33	97.84 5	92.68 193	98.71 72	78.11 302	99.70 10	97.71 78	98.18 1	97.36 37	99.76 1	90.37 38	99.94 23	99.27 3	99.54 42	99.99 1

CANet		97.00 20	96.49 28	98.55 6	98.86 68	96.10 10	99.83 4	97.52 109	95.90 8	97.21 38	98.90 58	82.66 143	99.93 25	98.71 9	98.80 74	99.63 55

CANet_DTU		94.31 83	93.35 88	97.20 46	97.03 119	94.71 32	98.62 124	95.54 243	95.61 12	97.21 38	98.47 88	71.88 237	99.84 46	88.38 144	97.46 98	97.04 178

VNet		95.08 65	94.26 69	97.55 33	98.07 85	93.88 50	98.68 116	98.73 21	90.33 90	97.16 40	97.43 115	79.19 168	99.53 81	96.91 40	91.85 159	99.24 83

region2R		96.30 42	96.17 38	96.70 77	99.70 7	90.31 125	99.46 30	97.66 83	90.55 84	97.07 41	99.07 36	86.85 87	99.97 14	95.43 63	99.74 21	99.81 23

原ACMM1				96.18 100	99.03 58	90.08 131		97.63 92	88.98 121	97.00 42	98.97 48	88.14 64	99.71 62	88.23 145	99.62 35	98.76 119

Regformer-1		96.97 21	96.80 22	97.47 34	99.46 37	93.11 62	98.89 95	97.94 53	92.89 41	96.90 43	99.02 42	89.78 42	99.53 81	97.06 33	99.26 57	99.75 36

HFP-MVS		96.42 38	96.26 35	96.90 62	99.69 8	90.96 112	99.47 27	97.81 66	90.54 85	96.88 44	99.05 39	87.57 69	99.96 18	95.65 58	99.72 23	99.78 30

#test#		96.48 35	96.34 33	96.90 62	99.69 8	90.96 112	99.53 24	97.81 66	90.94 78	96.88 44	99.05 39	87.57 69	99.96 18	95.87 57	99.72 23	99.78 30

Regformer-2		96.94 24	96.78 23	97.42 37	99.46 37	92.97 67	98.89 95	97.93 54	92.86 43	96.88 44	99.02 42	89.74 43	99.53 81	97.03 34	99.26 57	99.75 36

XVS		96.47 36	96.37 31	96.77 70	99.62 15	90.66 121	99.43 33	97.58 98	92.41 54	96.86 47	98.96 51	87.37 75	99.87 38	95.65 58	99.43 48	99.78 30

X-MVStestdata		90.69 164	88.66 177	96.77 70	99.62 15	90.66 121	99.43 33	97.58 98	92.41 54	96.86 47	29.59 356	87.37 75	99.87 38	95.65 58	99.43 48	99.78 30

1121		95.19 64	94.45 66	97.42 37	98.88 66	92.58 76	96.22 251	97.75 73	85.50 197	96.86 47	99.01 46	88.59 56	99.90 31	87.64 151	99.60 38	99.79 26

TSAR-MVS + GP.		96.95 22	96.91 18	97.07 48	98.88 66	91.62 87	99.58 18	96.54 179	95.09 15	96.84 50	98.63 77	91.16 17	99.77 58	99.04 5	96.42 109	99.81 23

ACMMPR		96.28 43	96.14 41	96.73 74	99.68 10	90.47 123	99.47 27	97.80 68	90.54 85	96.83 51	99.03 41	86.51 93	99.95 21	95.65 58	99.72 23	99.75 36

PMMVS		93.62 98	93.90 83	92.79 189	96.79 132	81.40 275	98.85 98	96.81 166	91.25 75	96.82 52	98.15 98	77.02 183	98.13 146	93.15 99	96.30 113	98.83 111

PGM-MVS		95.85 52	95.65 51	96.45 91	99.50 35	89.77 140	98.22 173	98.90 17	89.19 114	96.74 53	98.95 53	85.91 101	99.92 27	93.94 83	99.46 45	99.66 51

jason		95.40 62	94.86 62	97.03 50	92.91 227	94.23 45	99.70 10	96.30 191	93.56 33	96.73 54	98.52 82	81.46 157	97.91 155	96.08 54	98.47 83	98.96 99
jason: jason.
新几何1				97.40 39	98.92 64	92.51 78		97.77 72	85.52 195	96.69 55	99.06 38	88.08 65	99.89 34	84.88 174	99.62 35	99.79 26

APD-MVS_3200maxsize		95.64 58	95.65 51	95.62 118	99.24 49	87.80 172	98.42 150	97.22 139	88.93 125	96.64 56	98.98 47	85.49 106	99.36 100	96.68 41	99.27 56	99.70 45

MG-MVS		97.24 12	96.83 21	98.47 9	99.79 5	95.71 12	99.07 72	99.06 15	94.45 18	96.42 57	98.70 73	88.81 52	99.74 61	95.35 65	99.86 8	99.97 3

alignmvs		95.77 56	95.00 61	98.06 18	97.35 108	95.68 13	99.71 9	97.50 114	91.50 68	96.16 58	98.61 78	86.28 97	99.00 117	96.19 51	91.74 161	99.51 64

Regformer-3		96.50 34	96.36 32	96.91 61	99.34 40	91.72 85	98.71 109	97.90 56	92.48 49	96.00 59	98.95 53	88.60 54	99.52 84	96.44 46	98.83 71	99.49 66

CP-MVS		96.22 44	96.15 40	96.42 93	99.67 11	89.62 143	99.70 10	97.61 94	90.07 99	96.00 59	99.16 25	87.43 73	99.92 27	96.03 55	99.72 23	99.70 45

Regformer-4		96.45 37	96.33 34	96.81 69	99.34 40	91.44 92	98.71 109	97.88 57	92.43 50	95.97 61	98.95 53	88.42 58	99.51 85	96.40 47	98.83 71	99.49 66

MCST-MVS		98.18 2	97.95 4	98.86 1	99.85 3	96.60 5	99.70 10	97.98 52	97.18 2	95.96 62	99.33 9	92.62 12	100.00 1	98.99 6	99.93 1	99.98 2

DeepC-MVS_fast	93.52 2	97.16 15	96.84 20	98.13 15	99.61 17	94.45 40	98.85 98	97.64 88	96.51 6	95.88 63	99.39 8	87.35 79	99.99 4	96.61 42	99.69 28	99.96 4
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
test222					98.32 79	91.21 101	98.08 186	97.58 98	83.74 230	95.87 64	99.02 42	86.74 88			99.64 31	99.81 23

canonicalmvs		95.02 66	93.96 79	98.20 12	97.53 101	95.92 11	98.71 109	96.19 200	91.78 64	95.86 65	98.49 86	79.53 165	99.03 116	96.12 52	91.42 167	99.66 51

abl_6		94.63 77	94.48 65	95.09 136	98.61 75	86.96 193	98.06 188	96.97 162	89.31 110	95.86 65	98.56 80	79.82 163	99.64 72	94.53 80	98.65 80	98.66 124

Effi-MVS+		93.87 89	93.15 92	96.02 106	95.79 156	90.76 117	96.70 233	95.78 225	86.98 177	95.71 67	97.17 131	79.58 164	98.01 153	94.57 79	96.09 117	99.31 76

HPM-MVS_fast		94.89 67	94.62 64	95.70 117	99.11 54	88.44 163	99.14 66	97.11 148	85.82 192	95.69 68	98.47 88	83.46 126	99.32 104	93.16 98	99.63 34	99.35 72

HY-MVS	88.56 7	95.29 63	94.23 70	98.48 8	97.72 91	96.41 8	94.03 289	98.74 19	92.42 53	95.65 69	94.76 182	86.52 92	99.49 87	95.29 67	92.97 143	99.53 62

CHOSEN 280x420		96.80 27	96.85 19	96.66 80	97.85 88	94.42 42	94.76 281	98.36 26	92.50 48	95.62 70	97.52 111	97.92 1	97.38 195	98.31 22	98.80 74	98.20 149

MP-MVS		96.00 48	95.82 46	96.54 88	99.47 36	90.13 130	99.36 44	97.41 126	90.64 83	95.49 71	98.95 53	85.51 105	99.98 9	96.00 56	99.59 40	99.52 63

HPM-MVS		95.41 61	95.22 57	95.99 107	99.29 45	89.14 149	99.17 57	97.09 152	87.28 173	95.40 72	98.48 87	84.93 112	99.38 98	95.64 62	99.65 30	99.47 69
Chunlin Ren, Qingshan Xu, Shikun Zhang, Jiaqi Yang: Hierarchical Prior Mining for Non-local Multi-View Stereo. ICCV 2023
UA-Net		93.30 107	92.62 102	95.34 129	96.27 144	88.53 162	95.88 265	96.97 162	90.90 79	95.37 73	97.07 136	82.38 149	99.10 114	83.91 187	94.86 132	98.38 139

sss		94.85 69	93.94 81	97.58 30	96.43 140	94.09 47	98.93 86	99.16 14	89.50 108	95.27 74	97.85 100	81.50 156	99.65 70	92.79 104	94.02 137	98.99 96

WTY-MVS		95.97 49	95.11 59	98.54 7	97.62 94	96.65 4	99.44 31	98.74 19	92.25 57	95.21 75	98.46 90	86.56 91	99.46 94	95.00 71	92.69 147	99.50 65

DELS-MVS		97.12 16	96.60 27	98.68 5	98.03 86	96.57 6	99.84 3	97.84 61	96.36 7	95.20 76	98.24 93	88.17 62	99.83 48	96.11 53	99.60 38	99.64 53
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
MVS_111021_HR		96.69 28	96.69 25	96.72 76	98.58 76	91.00 111	99.14 66	99.45 1	93.86 26	95.15 77	98.73 69	88.48 57	99.76 59	97.23 32	99.56 41	99.40 70

MVS_Test		93.67 96	92.67 101	96.69 78	96.72 134	92.66 72	97.22 216	96.03 206	87.69 163	95.12 78	94.03 189	81.55 155	98.28 142	89.17 139	96.46 107	99.14 88

MVS_111021_LR		95.78 55	95.94 43	95.28 131	98.19 83	87.69 173	98.80 103	99.26 13	93.39 34	95.04 79	98.69 74	84.09 120	99.76 59	96.96 39	99.06 61	98.38 139

CostFormer		92.89 116	92.48 105	94.12 163	94.99 184	85.89 228	92.89 299	97.00 161	86.98 177	95.00 80	90.78 245	90.05 40	97.51 185	92.92 102	91.73 162	98.96 99

mPP-MVS		95.90 51	95.75 49	96.38 95	99.58 19	89.41 148	99.26 51	97.41 126	90.66 80	94.82 81	98.95 53	86.15 99	99.98 9	95.24 68	99.64 31	99.74 39

EI-MVSNet-Vis-set		95.76 57	95.63 53	96.17 102	99.14 52	90.33 124	98.49 142	97.82 63	91.92 61	94.75 82	98.88 60	87.06 83	99.48 92	95.40 64	97.17 102	98.70 122

LFMVS		92.23 132	90.84 146	96.42 93	98.24 80	91.08 109	98.24 171	96.22 198	83.39 242	94.74 83	98.31 92	61.12 300	98.85 119	94.45 81	92.82 144	99.32 75

tpmrst		92.78 117	92.16 112	94.65 149	96.27 144	87.45 180	91.83 308	97.10 151	89.10 119	94.68 84	90.69 250	88.22 61	97.73 173	89.78 128	91.80 160	98.77 118

DP-MVS Recon		95.85 52	95.15 58	97.95 19	99.87 2	94.38 43	99.60 17	97.48 116	86.58 184	94.42 85	99.13 31	87.36 78	99.98 9	93.64 90	98.33 85	99.48 68

zzz-MVS		96.21 45	95.96 42	96.96 59	99.29 45	91.19 102	98.69 113	97.45 119	92.58 46	94.39 86	99.24 14	86.43 95	99.99 4	96.22 49	99.40 51	99.71 43

MTAPA		96.09 47	95.80 48	96.96 59	99.29 45	91.19 102	97.23 215	97.45 119	92.58 46	94.39 86	99.24 14	86.43 95	99.99 4	96.22 49	99.40 51	99.71 43

CPTT-MVS		94.60 78	94.43 67	95.09 136	99.66 12	86.85 196	99.44 31	97.47 117	83.22 244	94.34 88	98.96 51	82.50 144	99.55 79	94.81 74	99.50 43	98.88 107

diffmvs		92.07 138	90.77 150	95.97 109	96.41 141	91.32 100	96.46 240	95.98 207	81.73 266	94.33 89	93.36 207	78.72 173	98.20 143	84.28 179	95.66 125	98.41 135

PVSNet_BlendedMVS		93.36 104	93.20 91	93.84 172	98.77 70	91.61 88	99.47 27	98.04 48	91.44 69	94.21 90	92.63 221	83.50 124	99.87 38	97.41 30	83.37 227	90.05 285

PVSNet_Blended		95.94 50	95.66 50	96.75 72	98.77 70	91.61 88	99.88 1	98.04 48	93.64 31	94.21 90	97.76 104	83.50 124	99.87 38	97.41 30	97.75 93	98.79 114

EI-MVSNet-UG-set		95.43 59	95.29 55	95.86 113	99.07 57	89.87 137	98.43 149	97.80 68	91.78 64	94.11 92	98.77 65	86.25 98	99.48 92	94.95 73	96.45 108	98.22 147

MAR-MVS		94.43 80	94.09 73	95.45 127	99.10 55	87.47 179	98.39 156	97.79 70	88.37 141	94.02 93	99.17 23	78.64 175	99.91 29	92.48 105	98.85 70	98.96 99
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
PAPM		96.35 39	95.94 43	97.58 30	94.10 198	95.25 16	98.93 86	98.17 41	94.26 19	93.94 94	98.72 71	89.68 44	97.88 158	96.36 48	99.29 55	99.62 57

GG-mvs-BLEND				96.98 57	96.53 137	94.81 29	87.20 323	97.74 75		93.91 95	96.40 160	96.56 2	96.94 209	95.08 69	98.95 68	99.20 86

API-MVS		94.78 70	94.18 71	96.59 86	99.21 50	90.06 134	98.80 103	97.78 71	83.59 234	93.85 96	99.21 17	83.79 122	99.97 14	92.37 106	99.00 64	99.74 39

tpm2		91.77 144	91.09 134	93.82 173	94.83 188	85.56 237	92.51 302	97.16 144	84.00 221	93.83 97	90.66 255	87.54 71	97.17 200	87.73 150	91.55 165	98.72 120

PAPR		96.35 39	95.82 46	97.94 20	99.63 14	94.19 46	99.42 37	97.55 105	92.43 50	93.82 98	99.12 32	87.30 80	99.91 29	94.02 82	99.06 61	99.74 39

PVSNet	87.13 12	93.69 93	92.83 98	96.28 98	97.99 87	90.22 127	99.38 40	98.93 16	91.42 72	93.66 99	97.68 107	71.29 243	99.64 72	87.94 148	97.20 101	98.98 97

VDD-MVS		91.24 154	90.18 157	94.45 154	97.08 117	85.84 232	98.40 155	96.10 204	86.99 175	93.36 100	98.16 97	54.27 318	99.20 106	96.59 43	90.63 176	98.31 145

VDDNet		90.08 173	88.54 183	94.69 148	94.41 194	87.68 174	98.21 176	96.40 185	76.21 304	93.33 101	97.75 105	54.93 316	98.77 122	94.71 77	90.96 170	97.61 166

MP-MVS-pluss		95.80 54	95.30 54	97.29 43	98.95 63	92.66 72	98.59 130	97.14 145	88.95 123	93.12 102	99.25 12	85.62 102	99.94 23	96.56 44	99.48 44	99.28 80
MP-MVS-pluss: MP-MVS-pluss. MP-MVS-pluss
MDTV_nov1_ep13_2view						91.17 104	91.38 311		87.45 167	93.08 103		86.67 89		87.02 156		98.95 103

DWT-MVSNet_test		94.36 81	93.95 80	95.62 118	96.99 120	89.47 146	96.62 236	97.38 129	90.96 77	93.07 104	97.27 123	93.73 8	98.09 147	85.86 168	93.65 139	99.29 78

EPNet_dtu		92.28 130	92.15 113	92.70 192	97.29 110	84.84 244	98.64 122	97.82 63	92.91 40	93.02 105	97.02 138	85.48 108	95.70 271	72.25 297	94.89 131	97.55 167
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
gg-mvs-nofinetune		90.00 174	87.71 190	96.89 66	96.15 149	94.69 33	85.15 329	97.74 75	68.32 328	92.97 106	60.16 342	96.10 3	96.84 211	93.89 84	98.87 69	99.14 88

114514_t		94.06 86	93.05 94	97.06 49	99.08 56	92.26 80	98.97 84	97.01 160	82.58 255	92.57 107	98.22 94	80.68 161	99.30 105	89.34 135	99.02 63	99.63 55

OMC-MVS		93.90 88	93.62 86	94.73 147	98.63 74	87.00 192	98.04 189	96.56 177	92.19 58	92.46 108	98.73 69	79.49 166	99.14 112	92.16 109	94.34 135	98.03 153

PAPM_NR		95.43 59	95.05 60	96.57 87	99.42 39	90.14 128	98.58 131	97.51 111	90.65 82	92.44 109	98.90 58	87.77 68	99.90 31	90.88 118	99.32 54	99.68 48

UGNet		91.91 143	90.85 145	95.10 135	97.06 118	88.69 158	98.01 190	98.24 30	92.41 54	92.39 110	93.61 202	60.52 301	99.68 63	88.14 146	97.25 100	96.92 184
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
MDTV_nov1_ep13			90.47 156		96.14 150	88.55 160	91.34 312	97.51 111	89.58 104	92.24 111	90.50 264	86.99 86	97.61 179	77.64 243	92.34 150

PatchFormer-LS_test		94.08 85	93.60 87	95.53 125	96.92 121	89.57 144	96.51 239	97.34 133	91.29 74	92.22 112	97.18 129	91.66 15	98.02 152	87.05 155	92.21 154	99.00 94

Vis-MVSNet		92.64 125	91.85 120	95.03 141	95.12 178	88.23 164	98.48 143	96.81 166	91.61 66	92.16 113	97.22 127	71.58 241	98.00 154	85.85 169	97.81 89	98.88 107
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
TESTMET0.1,1		93.82 90	93.26 90	95.49 126	95.21 171	90.25 126	99.15 63	97.54 108	89.18 116	91.79 114	94.87 180	89.13 47	97.63 177	86.21 162	96.29 114	98.60 125

EPMVS		92.59 127	91.59 127	95.59 120	97.22 112	90.03 135	91.78 309	98.04 48	90.42 87	91.66 115	90.65 256	86.49 94	97.46 186	81.78 209	96.31 112	99.28 80

test-LLR		93.11 114	92.68 100	94.40 155	94.94 186	87.27 189	99.15 63	97.25 135	90.21 91	91.57 116	94.04 187	84.89 113	97.58 180	85.94 165	96.13 115	98.36 142

test-mter		93.27 109	92.89 97	94.40 155	94.94 186	87.27 189	99.15 63	97.25 135	88.95 123	91.57 116	94.04 187	88.03 66	97.58 180	85.94 165	96.13 115	98.36 142

JIA-IIPM		85.97 235	84.85 233	89.33 258	93.23 224	73.68 314	85.05 330	97.13 147	69.62 324	91.56 118	68.03 340	88.03 66	96.96 207	77.89 242	93.12 141	97.34 170

tpmp4_e23		91.05 156	90.07 158	93.97 169	95.77 158	85.30 239	92.64 300	97.09 152	84.42 216	91.53 119	90.31 267	87.38 74	97.82 162	80.86 217	90.62 177	98.79 114

PVSNet_Blended_VisFu		94.67 75	94.11 72	96.34 97	97.14 115	91.10 107	99.32 50	97.43 124	92.10 60	91.53 119	96.38 163	83.29 130	99.68 63	93.42 95	96.37 110	98.25 146

CHOSEN 1792x2688		94.35 82	93.82 84	95.95 110	97.40 106	88.74 157	98.41 152	98.27 28	92.18 59	91.43 121	96.40 160	78.88 169	99.81 53	93.59 91	97.81 89	99.30 77

ACMMP		94.67 75	94.30 68	95.79 114	99.25 48	88.13 166	98.41 152	98.67 23	90.38 88	91.43 121	98.72 71	82.22 151	99.95 21	93.83 87	95.76 123	99.29 78
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
EPP-MVSNet		93.75 92	93.67 85	94.01 167	95.86 155	85.70 234	98.67 118	97.66 83	84.46 214	91.36 123	97.18 129	91.16 17	97.79 164	92.93 101	93.75 138	98.53 130

PLC	91.07 3	94.23 84	94.01 75	94.87 143	99.17 51	87.49 178	99.25 52	96.55 178	88.43 139	91.26 124	98.21 96	85.92 100	99.86 43	89.77 129	97.57 94	97.24 172
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
HyFIR lowres test		93.68 95	93.29 89	94.87 143	97.57 100	88.04 168	98.18 178	98.47 24	87.57 165	91.24 125	95.05 178	85.49 106	97.46 186	93.22 97	92.82 144	99.10 90

thres200		93.69 93	92.59 103	96.97 58	97.76 89	94.74 31	99.35 46	99.36 2	89.23 113	91.21 126	96.97 141	83.42 127	98.77 122	85.08 172	90.96 170	97.39 169

mvs-test1		91.57 146	92.20 111	89.70 250	95.15 176	74.34 311	99.51 25	95.40 254	91.92 61	91.02 127	97.25 124	74.27 208	98.08 150	89.45 131	95.83 122	96.67 185

CDS-MVSNet		93.47 99	93.04 95	94.76 145	94.75 190	89.45 147	98.82 101	97.03 158	87.91 155	90.97 128	96.48 158	89.06 48	96.36 238	89.50 130	92.81 146	98.49 132
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
tfpn200view9		93.43 101	92.27 109	96.90 62	97.68 92	94.84 24	99.18 55	99.36 2	88.45 136	90.79 129	96.90 143	83.31 128	98.75 124	84.11 183	90.69 172	97.12 173

thres400		93.39 103	92.27 109	96.73 74	97.68 92	94.84 24	99.18 55	99.36 2	88.45 136	90.79 129	96.90 143	83.31 128	98.75 124	84.11 183	90.69 172	96.61 186

CR-MVSNet		88.83 191	87.38 194	93.16 182	93.47 217	86.24 215	84.97 331	94.20 284	88.92 126	90.76 131	86.88 304	84.43 117	94.82 291	70.64 302	92.17 156	98.41 135

RPMNet		84.62 250	81.78 263	93.16 182	93.47 217	86.24 215	84.97 331	96.28 195	64.85 334	90.76 131	78.80 333	80.95 160	94.82 291	53.76 332	92.17 156	98.41 135

PatchmatchNet		92.05 142	91.04 135	95.06 139	96.17 148	89.04 151	91.26 313	97.26 134	89.56 106	90.64 133	90.56 262	88.35 60	97.11 202	79.53 226	96.07 119	99.03 93
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
PatchT		85.44 244	83.19 248	92.22 198	93.13 226	83.00 260	83.80 337	96.37 186	70.62 318	90.55 134	79.63 331	84.81 115	94.87 289	58.18 329	91.59 164	98.79 114

tpm		89.67 178	88.95 171	91.82 205	92.54 230	81.43 274	92.95 298	95.92 216	87.81 157	90.50 135	89.44 282	84.99 111	95.65 272	83.67 190	82.71 233	98.38 139

tfpn111		93.20 111	92.00 117	96.83 68	97.62 94	94.84 24	99.06 74	99.36 2	87.96 151	90.47 136	96.78 145	83.29 130	98.71 129	82.93 195	90.47 179	96.94 180

conf200view11		93.32 106	92.15 113	96.84 67	97.62 94	94.84 24	99.06 74	99.36 2	87.96 151	90.47 136	96.78 145	83.29 130	98.75 124	84.11 183	90.69 172	96.94 180

thres100view900		93.34 105	92.15 113	96.90 62	97.62 94	94.84 24	99.06 74	99.36 2	87.96 151	90.47 136	96.78 145	83.29 130	98.75 124	84.11 183	90.69 172	97.12 173

thres600view7		93.18 112	92.00 117	96.75 72	97.62 94	94.92 21	99.07 72	99.36 2	87.96 151	90.47 136	96.78 145	83.29 130	98.71 129	82.93 195	90.47 179	96.61 186

AdaColmap		93.82 90	93.06 93	96.10 105	99.88 1	89.07 150	98.33 158	97.55 105	86.81 182	90.39 140	98.65 75	75.09 191	99.98 9	93.32 96	97.53 96	99.26 82

XVG-OURS-SEG-HR		90.95 158	90.66 153	91.83 204	95.18 175	81.14 281	95.92 262	95.92 216	88.40 140	90.33 141	97.85 100	70.66 246	99.38 98	92.83 103	88.83 196	94.98 200

IS-MVSNet		93.00 115	92.51 104	94.49 152	96.14 150	87.36 186	98.31 161	95.70 231	88.58 132	90.17 142	97.50 112	83.02 139	97.22 198	87.06 154	96.07 119	98.90 106

CSCG		94.87 68	94.71 63	95.36 128	99.54 27	86.49 206	99.34 48	98.15 43	82.71 253	90.15 143	99.25 12	89.48 45	99.86 43	94.97 72	98.82 73	99.72 42

Patchmatch-test1		90.10 171	88.61 178	94.57 151	94.95 185	88.83 153	96.26 247	97.21 140	90.06 100	90.03 144	90.68 252	66.61 275	95.83 268	77.31 244	94.36 134	99.05 92

XVG-OURS		90.83 160	90.49 155	91.86 203	95.23 170	81.25 279	95.79 270	95.92 216	88.96 122	90.02 145	98.03 99	71.60 240	99.35 102	91.06 115	87.78 200	94.98 200

ADS-MVSNet2		87.62 206	86.88 201	89.86 246	96.21 146	79.14 291	87.15 324	92.99 299	83.01 248	89.91 146	87.27 300	78.87 170	92.80 309	74.20 278	92.27 152	97.64 162

ADS-MVSNet		88.99 186	87.30 195	94.07 164	96.21 146	87.56 177	87.15 324	96.78 168	83.01 248	89.91 146	87.27 300	78.87 170	97.01 206	74.20 278	92.27 152	97.64 162

tfpn_ndepth		93.28 108	92.32 106	96.16 103	97.74 90	92.86 70	99.01 80	98.19 39	85.50 197	89.84 148	97.12 133	93.57 9	97.58 180	79.39 229	90.50 178	98.04 152

ab-mvs		91.05 156	89.17 167	96.69 78	95.96 153	91.72 85	92.62 301	97.23 138	85.61 194	89.74 149	93.89 195	68.55 259	99.42 95	91.09 114	87.84 199	98.92 105

TAMVS		92.62 126	92.09 116	94.20 161	94.10 198	87.68 174	98.41 152	96.97 162	87.53 166	89.74 149	96.04 167	84.77 116	96.49 227	88.97 141	92.31 151	98.42 134

Vis-MVSNet (Re-imp)		93.26 110	93.00 96	94.06 165	96.14 150	86.71 202	98.68 116	96.70 169	88.30 143	89.71 151	97.64 108	85.43 109	96.39 236	88.06 147	96.32 111	99.08 91

view600		92.78 117	91.50 129	96.63 81	97.51 102	94.66 34	98.91 89	99.36 2	87.31 169	89.64 152	96.59 152	83.26 135	98.63 133	80.76 218	90.15 182	96.61 186

view800		92.78 117	91.50 129	96.63 81	97.51 102	94.66 34	98.91 89	99.36 2	87.31 169	89.64 152	96.59 152	83.26 135	98.63 133	80.76 218	90.15 182	96.61 186

conf0.05thres1000		92.78 117	91.50 129	96.63 81	97.51 102	94.66 34	98.91 89	99.36 2	87.31 169	89.64 152	96.59 152	83.26 135	98.63 133	80.76 218	90.15 182	96.61 186

tfpn		92.78 117	91.50 129	96.63 81	97.51 102	94.66 34	98.91 89	99.36 2	87.31 169	89.64 152	96.59 152	83.26 135	98.63 133	80.76 218	90.15 182	96.61 186

CNLPA		93.64 97	92.74 99	96.36 96	98.96 62	90.01 136	99.19 53	95.89 222	86.22 189	89.40 156	98.85 61	80.66 162	99.84 46	88.57 143	96.92 103	99.24 83

Fast-Effi-MVS+		91.72 145	90.79 149	94.49 152	95.89 154	87.40 183	99.54 23	95.70 231	85.01 207	89.28 157	95.68 170	77.75 179	97.57 184	83.22 191	95.06 129	98.51 131

PatchMatch-RL		91.47 148	90.54 154	94.26 159	98.20 81	86.36 212	96.94 223	97.14 145	87.75 159	88.98 158	95.75 169	71.80 239	99.40 97	80.92 215	97.39 99	97.02 179

dp		90.16 170	88.83 174	94.14 162	96.38 142	86.42 208	91.57 310	97.06 155	84.76 211	88.81 159	90.19 275	84.29 119	97.43 188	75.05 270	91.35 169	98.56 129

tfpn1000		92.67 124	91.64 126	95.78 115	97.61 99	92.34 79	98.69 113	98.18 40	84.15 219	88.80 160	96.99 140	93.56 10	97.21 199	76.56 254	90.19 181	97.77 161

DeepC-MVS	91.02 4	94.56 79	93.92 82	96.46 90	97.16 114	90.76 117	98.39 156	97.11 148	93.92 22	88.66 161	98.33 91	78.14 177	99.85 45	95.02 70	98.57 81	98.78 117
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
CVMVSNet		90.30 166	90.91 144	88.46 273	94.32 195	73.58 315	97.61 204	97.59 96	90.16 96	88.43 162	97.10 134	76.83 184	92.86 305	82.64 198	93.54 140	98.93 104

TR-MVS		90.77 161	89.44 163	94.76 145	96.31 143	88.02 169	97.92 192	95.96 211	85.52 195	88.22 163	97.23 126	66.80 273	98.09 147	84.58 177	92.38 149	98.17 150

conf0.01		92.06 140	90.99 136	95.24 133	96.84 124	91.39 93	98.31 161	98.20 32	83.57 235	88.08 164	97.34 117	91.05 20	97.40 189	75.80 260	89.74 189	96.94 180

conf0.002		92.06 140	90.99 136	95.24 133	96.84 124	91.39 93	98.31 161	98.20 32	83.57 235	88.08 164	97.34 117	91.05 20	97.40 189	75.80 260	89.74 189	96.94 180

thresconf0.02		92.14 133	90.99 136	95.58 121	96.84 124	91.39 93	98.31 161	98.20 32	83.57 235	88.08 164	97.34 117	91.05 20	97.40 189	75.80 260	89.74 189	97.94 156

tfpn_n400		92.14 133	90.99 136	95.58 121	96.84 124	91.39 93	98.31 161	98.20 32	83.57 235	88.08 164	97.34 117	91.05 20	97.40 189	75.80 260	89.74 189	97.94 156

tfpnconf		92.14 133	90.99 136	95.58 121	96.84 124	91.39 93	98.31 161	98.20 32	83.57 235	88.08 164	97.34 117	91.05 20	97.40 189	75.80 260	89.74 189	97.94 156

tfpnview11		92.14 133	90.99 136	95.58 121	96.84 124	91.39 93	98.31 161	98.20 32	83.57 235	88.08 164	97.34 117	91.05 20	97.40 189	75.80 260	89.74 189	97.94 156

F-COLMAP		92.07 138	91.75 124	93.02 185	98.16 84	82.89 264	98.79 106	95.97 209	86.54 185	87.92 170	97.80 102	78.69 174	99.65 70	85.97 164	95.93 121	96.53 195

BH-RMVSNet		91.25 153	89.99 159	95.03 141	96.75 133	88.55 160	98.65 120	94.95 267	87.74 160	87.74 171	97.80 102	68.27 261	98.14 145	80.53 223	97.49 97	98.41 135

Effi-MVS+-dtu		89.97 175	90.68 152	87.81 286	95.15 176	71.98 320	97.87 196	95.40 254	91.92 61	87.57 172	91.44 234	74.27 208	96.84 211	89.45 131	93.10 142	94.60 202

HQP-NCC					93.95 202		99.16 58		93.92 22	87.57 172

ACMP_Plane					93.95 202		99.16 58		93.92 22	87.57 172

HQP4-MVS										87.57 172			97.77 166			92.72 209

HQP-MVS		91.50 147	91.23 133	92.29 197	93.95 202	86.39 210	99.16 58	96.37 186	93.92 22	87.57 172	96.67 150	73.34 221	97.77 166	93.82 88	86.29 204	92.72 209

TAPA-MVS	87.50 9	90.35 165	89.05 169	94.25 160	98.48 77	85.17 242	98.42 150	96.58 176	82.44 259	87.24 177	98.53 81	82.77 142	98.84 120	59.09 327	97.88 88	98.72 120
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
HQP_MVS		91.26 151	90.95 143	92.16 199	93.84 209	86.07 223	99.02 78	96.30 191	93.38 35	86.99 178	96.52 156	72.92 226	97.75 171	93.46 93	86.17 207	92.67 211

plane_prior3						85.91 227			93.65 30	86.99 178

GA-MVS		90.10 171	88.69 176	94.33 157	92.44 231	87.97 170	99.08 71	96.26 196	89.65 102	86.92 180	93.11 214	68.09 262	96.96 207	82.54 199	90.15 182	98.05 151

1112_ss		92.71 122	91.55 128	96.20 99	95.56 163	91.12 105	98.48 143	94.69 273	88.29 144	86.89 181	98.50 84	87.02 84	98.66 131	84.75 175	89.77 188	98.81 112

Test_1112_low_res		92.27 131	90.97 142	96.18 100	95.53 164	91.10 107	98.47 145	94.66 274	88.28 145	86.83 182	93.50 206	87.00 85	98.65 132	84.69 176	89.74 189	98.80 113

cascas		90.93 159	89.33 166	95.76 116	95.69 160	93.03 66	98.99 83	96.59 173	80.49 276	86.79 183	94.45 185	65.23 283	98.60 137	93.52 92	92.18 155	95.66 199

OPM-MVS		89.76 177	89.15 168	91.57 214	90.53 256	85.58 236	98.11 183	95.93 215	92.88 42	86.05 184	96.47 159	67.06 272	97.87 159	89.29 138	86.08 209	91.26 249

VPA-MVSNet		89.10 185	87.66 191	93.45 177	92.56 229	91.02 110	97.97 191	98.32 27	86.92 179	86.03 185	92.01 226	68.84 258	97.10 204	90.92 117	75.34 262	92.23 221

tpm cat1		88.89 188	87.27 196	93.76 174	95.79 156	85.32 238	90.76 317	97.09 152	76.14 305	85.72 186	88.59 290	82.92 140	98.04 151	76.96 248	91.43 166	97.90 160

IB-MVS	89.43 6	92.12 137	90.83 148	95.98 108	95.40 168	90.78 116	99.81 5	98.06 46	91.23 76	85.63 187	93.66 201	90.63 33	98.78 121	91.22 113	71.85 298	98.36 142
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
EI-MVSNet		89.87 176	89.38 165	91.36 219	94.32 195	85.87 229	97.61 204	96.59 173	85.10 203	85.51 188	97.10 134	81.30 159	96.56 219	83.85 189	83.03 230	91.64 236

MVSTER		92.71 122	92.32 106	93.86 171	97.29 110	92.95 68	99.01 80	96.59 173	90.09 97	85.51 188	94.00 191	94.61 5	96.56 219	90.77 121	83.03 230	92.08 228

RPSCF		85.33 245	85.55 222	84.67 306	94.63 192	62.28 332	93.73 292	93.76 288	74.38 311	85.23 190	97.06 137	64.09 286	98.31 140	80.98 213	86.08 209	93.41 208

BH-w/o		92.32 129	91.79 122	93.91 170	96.85 123	86.18 218	99.11 70	95.74 227	88.13 148	84.81 191	97.00 139	77.26 182	97.91 155	89.16 140	98.03 87	97.64 162

CLD-MVS		91.06 155	90.71 151	92.10 200	94.05 201	86.10 221	99.55 22	96.29 194	94.16 20	84.70 192	97.17 131	69.62 251	97.82 162	94.74 76	86.08 209	92.39 214
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
tpmvs		89.16 184	87.76 188	93.35 178	97.19 113	84.75 246	90.58 319	97.36 131	81.99 262	84.56 193	89.31 285	83.98 121	98.17 144	74.85 273	90.00 187	97.12 173

nrg030		90.23 167	88.87 172	94.32 158	91.53 245	93.54 55	98.79 106	95.89 222	88.12 149	84.55 194	94.61 184	78.80 172	96.88 210	92.35 107	75.21 263	92.53 213

VPNet		88.30 200	86.57 202	93.49 176	91.95 238	91.35 99	98.18 178	97.20 141	88.61 131	84.52 195	94.89 179	62.21 295	96.76 215	89.34 135	72.26 294	92.36 215

MVS		93.92 87	92.28 108	98.83 2	95.69 160	96.82 3	96.22 251	98.17 41	84.89 209	84.34 196	98.61 78	79.32 167	99.83 48	93.88 85	99.43 48	99.86 20

mvs_anonymous		92.50 128	91.65 125	95.06 139	96.60 136	89.64 142	97.06 221	96.44 184	86.64 183	84.14 197	93.93 193	82.49 145	96.17 255	91.47 112	96.08 118	99.35 72

Fast-Effi-MVS+-dtu		88.84 190	88.59 181	89.58 253	93.44 220	78.18 300	98.65 120	94.62 275	88.46 135	84.12 198	95.37 176	68.91 256	96.52 225	82.06 203	91.70 163	94.06 203

LS3D		90.19 169	88.72 175	94.59 150	98.97 60	86.33 213	96.90 225	96.60 172	74.96 308	84.06 199	98.74 68	75.78 188	99.83 48	74.93 271	97.57 94	97.62 165

ACMM	86.95 13	88.77 194	88.22 187	90.43 235	93.61 214	81.34 277	98.50 140	95.92 216	87.88 156	83.85 200	95.20 177	67.20 270	97.89 157	86.90 159	84.90 216	92.06 229
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
BH-untuned		91.46 149	90.84 146	93.33 179	96.51 139	84.83 245	98.84 100	95.50 246	86.44 188	83.50 201	96.70 149	75.49 190	97.77 166	86.78 161	97.81 89	97.40 168

FIs		90.70 163	89.87 160	93.18 181	92.29 232	91.12 105	98.17 181	98.25 29	89.11 118	83.44 202	94.82 181	82.26 150	96.17 255	87.76 149	82.76 232	92.25 219

UniMVSNet (Re)		89.50 181	88.32 185	93.03 184	92.21 234	90.96 112	98.90 94	98.39 25	89.13 117	83.22 203	92.03 224	81.69 154	96.34 244	86.79 160	72.53 289	91.81 233

UniMVSNet_NR-MVSNet		89.60 179	88.55 182	92.75 191	92.17 235	90.07 132	98.74 108	98.15 43	88.37 141	83.21 204	93.98 192	82.86 141	95.93 265	86.95 157	72.47 290	92.25 219

DU-MVS		88.83 191	87.51 192	92.79 189	91.46 246	90.07 132	98.71 109	97.62 93	88.87 127	83.21 204	93.68 199	74.63 196	95.93 265	86.95 157	72.47 290	92.36 215

LPG-MVS_test		88.86 189	88.47 184	90.06 242	93.35 222	80.95 283	98.22 173	95.94 213	87.73 161	83.17 206	96.11 165	66.28 277	97.77 166	90.19 124	85.19 213	91.46 243

LGP-MVS_train				90.06 242	93.35 222	80.95 283		95.94 213	87.73 161	83.17 206	96.11 165	66.28 277	97.77 166	90.19 124	85.19 213	91.46 243

v6		87.27 212	85.86 214	91.50 215	89.97 265	86.84 198	98.45 146	95.67 233	83.85 226	83.11 208	90.97 241	74.46 203	96.58 217	81.97 205	74.34 271	91.09 253

v1neww		87.29 210	85.88 212	91.50 215	90.07 258	86.87 194	98.45 146	95.66 236	83.84 227	83.07 209	90.99 239	74.58 200	96.56 219	81.96 206	74.33 272	91.07 256

v7new		87.29 210	85.88 212	91.50 215	90.07 258	86.87 194	98.45 146	95.66 236	83.84 227	83.07 209	90.99 239	74.58 200	96.56 219	81.96 206	74.33 272	91.07 256

FC-MVSNet-test		90.22 168	89.40 164	92.67 194	91.78 242	89.86 138	97.89 193	98.22 31	88.81 128	82.96 211	94.66 183	81.90 153	95.96 263	85.89 167	82.52 235	92.20 224

v7		86.91 219	85.45 224	91.29 220	90.06 260	86.73 200	98.26 169	95.49 247	83.08 247	82.95 212	90.96 242	73.37 219	96.42 233	79.90 225	74.97 264	90.71 271

v1		87.23 214	85.76 216	91.66 212	89.88 270	87.37 185	98.54 134	95.64 238	83.91 223	82.88 213	90.70 248	74.64 194	96.53 223	81.54 211	74.08 278	91.08 254

divwei89l23v2f112		87.23 214	85.75 218	91.66 212	89.88 270	87.40 183	98.53 135	95.62 239	83.91 223	82.84 214	90.67 253	74.75 192	96.49 227	81.55 210	74.05 280	91.08 254

v1141		87.23 214	85.75 218	91.67 211	89.88 270	87.43 182	98.52 136	95.62 239	83.91 223	82.83 215	90.69 250	74.70 193	96.49 227	81.53 212	74.08 278	91.07 256

PCF-MVS	89.78 5	91.26 151	89.63 161	96.16 103	95.44 166	91.58 90	95.29 277	96.10 204	85.07 205	82.75 216	97.45 114	78.28 176	99.78 57	80.60 222	95.65 126	97.12 173
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
V42		87.00 218	85.68 221	90.98 225	89.91 266	86.08 222	98.32 160	95.61 241	83.67 233	82.72 217	90.67 253	74.00 214	96.53 223	81.94 208	74.28 275	90.32 279

v1144		86.83 221	85.31 226	91.40 218	89.75 275	87.21 191	98.31 161	95.45 251	83.22 244	82.70 218	90.78 245	73.36 220	96.36 238	79.49 227	74.69 268	90.63 274

v144192		86.40 229	84.89 232	90.91 226	89.48 286	85.59 235	98.21 176	95.43 253	82.45 258	82.62 219	90.58 261	72.79 229	96.36 238	78.45 237	74.04 281	90.79 266

3Dnovator	87.35 11	93.17 113	91.77 123	97.37 42	95.41 167	93.07 64	98.82 101	97.85 60	91.53 67	82.56 220	97.58 110	71.97 236	99.82 51	91.01 116	99.23 59	99.22 85

v2v482		87.27 212	85.76 216	91.78 210	89.59 281	87.58 176	98.56 132	95.54 243	84.53 213	82.51 221	91.78 230	73.11 225	96.47 230	82.07 202	74.14 277	91.30 248

Baseline_NR-MVSNet		85.83 238	84.82 234	88.87 266	88.73 294	83.34 257	98.63 123	91.66 322	80.41 277	82.44 222	91.35 235	74.63 196	95.42 278	84.13 182	71.39 301	87.84 303

v1192		86.32 231	84.71 236	91.17 221	89.53 284	86.40 209	98.13 182	95.44 252	82.52 257	82.42 223	90.62 258	71.58 241	96.33 245	77.23 245	74.88 265	90.79 266

test_djsdf		88.26 202	87.73 189	89.84 247	88.05 302	82.21 269	97.77 199	96.17 201	86.84 180	82.41 224	91.95 229	72.07 235	95.99 261	89.83 126	84.50 219	91.32 247

1314		93.44 100	91.98 119	97.84 22	95.24 169	94.38 43	96.22 251	97.92 55	90.18 93	82.28 225	97.71 106	77.63 180	99.80 55	91.94 111	98.67 79	99.34 74

v1921920		86.02 234	84.44 240	90.77 228	89.32 288	85.20 240	98.10 184	95.35 258	82.19 260	82.25 226	90.71 247	70.73 244	96.30 251	76.85 251	74.49 269	90.80 265

v1240		85.77 241	84.11 243	90.73 229	89.26 289	85.15 243	97.88 195	95.23 265	81.89 265	82.16 227	90.55 263	69.60 252	96.31 248	75.59 268	74.87 266	90.72 270

XVG-ACMP-BASELINE		85.86 237	84.95 231	88.57 270	89.90 268	77.12 305	94.30 285	95.60 242	87.40 168	82.12 228	92.99 217	53.42 321	97.66 175	85.02 173	83.83 223	90.92 262

GBi-Net		86.67 224	84.96 229	91.80 206	95.11 179	88.81 154	96.77 228	95.25 261	82.94 250	82.12 228	90.25 269	62.89 292	94.97 286	79.04 231	80.24 241	91.62 238

test1		86.67 224	84.96 229	91.80 206	95.11 179	88.81 154	96.77 228	95.25 261	82.94 250	82.12 228	90.25 269	62.89 292	94.97 286	79.04 231	80.24 241	91.62 238

FMVSNet3		88.81 193	87.08 200	93.99 168	96.52 138	94.59 38	98.08 186	96.20 199	85.85 191	82.12 228	91.60 233	74.05 213	95.40 279	79.04 231	80.24 241	91.99 231

IterMVS-LS		88.34 199	87.44 193	91.04 223	94.10 198	85.85 231	98.10 184	95.48 248	85.12 202	82.03 232	91.21 236	81.35 158	95.63 273	83.86 188	75.73 260	91.63 237
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
MIMVSNet		84.48 254	81.83 262	92.42 196	91.73 243	87.36 186	85.52 327	94.42 279	81.40 269	81.91 233	87.58 296	51.92 323	92.81 308	73.84 283	88.15 198	97.08 177

PS-MVSNAJss		89.54 180	89.05 169	91.00 224	88.77 293	84.36 249	97.39 207	95.97 209	88.47 133	81.88 234	93.80 197	82.48 146	96.50 226	89.34 135	83.34 228	92.15 225

WR-MVS		88.54 198	87.22 198	92.52 195	91.93 240	89.50 145	98.56 132	97.84 61	86.99 175	81.87 235	93.81 196	74.25 210	95.92 267	85.29 170	74.43 270	92.12 226

TranMVSNet+NR-MVSNet		87.75 203	86.31 206	92.07 201	90.81 253	88.56 159	98.33 158	97.18 142	87.76 158	81.87 235	93.90 194	72.45 230	95.43 277	83.13 193	71.30 302	92.23 221

DP-MVS		88.75 195	86.56 203	95.34 129	98.92 64	87.45 180	97.64 203	93.52 293	70.55 319	81.49 237	97.25 124	74.43 205	99.88 35	71.14 301	94.09 136	98.67 123

3Dnovator+	87.72 8	93.43 101	91.84 121	98.17 13	95.73 159	95.08 20	98.92 88	97.04 156	91.42 72	81.48 238	97.60 109	74.60 198	99.79 56	90.84 119	98.97 65	99.64 53

QAPM		91.41 150	89.49 162	97.17 47	95.66 162	93.42 58	98.60 128	97.51 111	80.92 274	81.39 239	97.41 116	72.89 228	99.87 38	82.33 200	98.68 78	98.21 148

XXY-MVS		87.75 203	86.02 209	92.95 187	90.46 257	89.70 141	97.71 201	95.90 220	84.02 220	80.95 240	94.05 186	67.51 268	97.10 204	85.16 171	78.41 250	92.04 230

v148		86.38 230	85.06 228	90.37 237	89.47 287	84.10 251	98.52 136	95.48 248	83.80 229	80.93 241	90.22 272	74.60 198	96.31 248	80.92 215	71.55 300	90.69 272

FMVSNet2		86.90 220	84.79 235	93.24 180	95.11 179	92.54 77	97.67 202	95.86 224	82.94 250	80.55 242	91.17 237	62.89 292	95.29 281	77.23 245	79.71 247	91.90 232

pmmvs4		87.58 207	86.17 208	91.80 206	89.58 282	88.92 152	97.25 213	95.28 260	82.54 256	80.49 243	93.17 213	75.62 189	96.05 260	82.75 197	78.90 248	90.42 277

ACMP	87.39 10	88.71 196	88.24 186	90.12 241	93.91 207	81.06 282	98.50 140	95.67 233	89.43 109	80.37 244	95.55 171	65.67 280	97.83 161	90.55 122	84.51 218	91.47 242
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
pmmvs5		85.87 236	84.40 242	90.30 238	88.53 297	84.23 250	98.60 128	93.71 290	81.53 268	80.29 245	92.02 225	64.51 285	95.52 275	82.04 204	78.34 251	91.15 251

test0.0.03 1		88.96 187	88.61 178	90.03 244	91.09 250	84.43 248	98.97 84	97.02 159	90.21 91	80.29 245	96.31 164	84.89 113	91.93 325	72.98 293	85.70 212	93.73 204

jajsoiax		87.35 208	86.51 204	89.87 245	87.75 307	81.74 272	97.03 222	95.98 207	88.47 133	80.15 247	93.80 197	61.47 297	96.36 238	89.44 133	84.47 220	91.50 241

mvs_tets		87.09 217	86.22 207	89.71 249	87.87 303	81.39 276	96.73 232	95.90 220	88.19 147	79.99 248	93.61 202	59.96 303	96.31 248	89.40 134	84.34 221	91.43 245

ITE_SJBPF				87.93 284	92.26 233	76.44 306		93.47 294	87.67 164	79.95 249	95.49 174	56.50 310	97.38 195	75.24 269	82.33 236	89.98 287

v8		86.11 233	84.45 239	91.10 222	89.99 264	86.85 196	97.24 214	95.36 256	81.99 262	79.89 250	89.86 278	74.53 202	96.39 236	78.83 235	72.32 292	90.05 285

v10		85.73 242	84.01 245	90.87 227	90.03 261	86.73 200	97.20 217	95.22 266	81.25 270	79.85 251	89.75 279	73.30 224	96.28 252	76.87 249	72.64 288	89.61 292

WR-MVS_H		86.53 228	85.49 223	89.66 252	91.04 251	83.31 258	97.53 206	98.20 32	84.95 208	79.64 252	90.90 244	78.01 178	95.33 280	76.29 256	72.81 286	90.35 278

anonymousdsp		86.69 223	85.75 218	89.53 254	86.46 316	82.94 261	96.39 242	95.71 230	83.97 222	79.63 253	90.70 248	68.85 257	95.94 264	86.01 163	84.02 222	89.72 291

Patchmtry		83.61 265	81.64 265	89.50 255	93.36 221	82.84 266	84.10 334	94.20 284	69.47 325	79.57 254	86.88 304	84.43 117	94.78 293	68.48 306	74.30 274	90.88 263

CP-MVSNet		86.54 227	85.45 224	89.79 248	91.02 252	82.78 267	97.38 209	97.56 104	85.37 199	79.53 255	93.03 215	71.86 238	95.25 282	79.92 224	73.43 284	91.34 246

Patchmatch-test		86.25 232	84.06 244	92.82 188	94.42 193	82.88 265	82.88 339	94.23 283	71.58 315	79.39 256	90.62 258	89.00 50	96.42 233	63.03 317	91.37 168	99.16 87

DSMNet-mixed		81.60 276	81.43 267	82.10 312	84.36 321	60.79 333	93.63 294	86.74 343	79.00 284	79.32 257	87.15 302	63.87 288	89.78 330	66.89 310	91.92 158	95.73 198

MSDG		88.29 201	86.37 205	94.04 166	96.90 122	86.15 220	96.52 238	94.36 281	77.89 301	79.22 258	96.95 142	69.72 250	99.59 78	73.20 290	92.58 148	96.37 196

PS-CasMVS		85.81 239	84.58 238	89.49 256	90.77 254	82.11 270	97.20 217	97.36 131	84.83 210	79.12 259	92.84 218	67.42 269	95.16 284	78.39 238	73.25 285	91.21 250

IterMVS		85.81 239	84.67 237	89.22 259	93.51 216	83.67 255	96.32 245	94.80 269	85.09 204	78.69 260	90.17 276	66.57 276	93.17 301	79.48 228	77.42 256	90.81 264
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
PEN-MVS		85.21 246	83.93 246	89.07 263	89.89 269	81.31 278	97.09 220	97.24 137	84.45 215	78.66 261	92.68 220	68.44 260	94.87 289	75.98 258	70.92 303	91.04 259

semantic-postprocess				89.00 264	93.46 219	82.90 263		94.70 272	85.02 206	78.62 262	90.35 265	66.63 274	93.33 300	79.38 230	77.36 257	90.76 268

OpenMVS	85.28 14	90.75 162	88.84 173	96.48 89	93.58 215	93.51 56	98.80 103	97.41 126	82.59 254	78.62 262	97.49 113	68.00 264	99.82 51	84.52 178	98.55 82	96.11 197

PVSNet_0	83.28 16	87.31 209	85.16 227	93.74 175	94.78 189	84.59 247	98.91 89	98.69 22	89.81 101	78.59 264	93.23 211	61.95 296	99.34 103	94.75 75	55.72 337	97.30 171

v52		84.19 258	82.92 251	88.01 282	87.64 309	79.92 287	96.23 249	95.32 259	79.87 280	78.51 265	89.05 286	69.50 254	96.32 246	77.95 241	72.24 295	87.79 306

V4		84.20 257	82.92 251	88.02 281	87.59 310	79.91 288	96.21 254	95.36 256	79.88 279	78.51 265	89.00 287	69.52 253	96.32 246	77.96 240	72.29 293	87.83 305

EU-MVSNet		84.19 258	84.42 241	83.52 309	88.64 296	67.37 329	96.04 259	95.76 226	85.29 200	78.44 267	93.18 212	70.67 245	91.48 328	75.79 266	75.98 258	91.70 235

v7n		84.42 255	82.75 256	89.43 257	88.15 300	81.86 271	96.75 231	95.67 233	80.53 275	78.38 268	89.43 283	69.89 248	96.35 243	73.83 284	72.13 296	90.07 284

FMVSNet1		83.94 262	81.32 269	91.80 206	91.94 239	88.81 154	96.77 228	95.25 261	77.98 296	78.25 269	90.25 269	50.37 327	94.97 286	73.27 289	77.81 254	91.62 238

MS-PatchMatch		86.75 222	85.92 211	89.22 259	91.97 237	82.47 268	96.91 224	96.14 203	83.74 230	77.73 270	93.53 205	58.19 305	97.37 197	76.75 252	98.35 84	87.84 303

v748		83.84 263	82.31 261	88.41 275	87.65 308	79.10 292	96.66 234	95.51 245	80.09 278	77.65 271	88.53 291	69.81 249	96.23 253	75.67 267	69.25 305	89.91 288

DTE-MVSNet		84.14 260	82.80 254	88.14 280	88.95 291	79.87 289	96.81 227	96.24 197	83.50 241	77.60 272	92.52 222	67.89 266	94.24 296	72.64 296	69.05 307	90.32 279

COLMAP_ROB	82.69 18	84.54 253	82.82 253	89.70 250	96.72 134	78.85 294	95.89 263	92.83 308	71.55 316	77.54 273	95.89 168	59.40 304	99.14 112	67.26 308	88.26 197	91.11 252
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
testpf		80.59 286	80.13 273	81.97 314	94.25 197	71.65 321	60.37 349	95.46 250	70.99 317	76.97 274	87.74 294	73.58 218	91.67 326	76.86 250	84.97 215	82.60 337

OurMVSNet-221017-0		84.13 261	83.59 247	85.77 300	87.81 304	70.24 324	94.89 280	93.65 292	86.08 190	76.53 275	93.28 210	61.41 298	96.14 257	80.95 214	77.69 255	90.93 261

tfpnnormal		83.65 264	81.35 268	90.56 232	91.37 248	88.06 167	97.29 211	97.87 59	78.51 290	76.20 276	90.91 243	64.78 284	96.47 230	61.71 320	73.50 282	87.13 313

pm-mvs1		84.68 249	82.78 255	90.40 236	89.58 282	85.18 241	97.31 210	94.73 271	81.93 264	76.05 277	92.01 226	65.48 282	96.11 258	78.75 236	69.14 306	89.91 288

AllTest		84.97 247	83.12 249	90.52 233	96.82 130	78.84 295	95.89 263	92.17 315	77.96 298	75.94 278	95.50 172	55.48 313	99.18 107	71.15 299	87.14 201	93.55 206

TestCases				90.52 233	96.82 130	78.84 295		92.17 315	77.96 298	75.94 278	95.50 172	55.48 313	99.18 107	71.15 299	87.14 201	93.55 206

testgi		82.29 267	81.00 271	86.17 297	87.24 312	74.84 310	97.39 207	91.62 323	88.63 130	75.85 280	95.42 175	46.07 332	91.55 327	66.87 311	79.94 244	92.12 226

MVP-Stereo		86.61 226	85.83 215	88.93 265	88.70 295	83.85 254	96.07 258	94.41 280	82.15 261	75.64 281	91.96 228	67.65 267	96.45 232	77.20 247	98.72 76	86.51 316
Qingsong Yan: MVP-Stereo: A Parallel Multi-View Patchmatch Stereo Method with Dilation Matching for Photogrammetric Application.
LF4IMVS		81.94 271	81.17 270	84.25 307	87.23 313	68.87 328	93.35 296	91.93 320	83.35 243	75.40 282	93.00 216	49.25 329	96.65 216	78.88 234	78.11 252	87.22 312

LTVRE_ROB	81.71 19	84.59 252	82.72 257	90.18 239	92.89 228	83.18 259	93.15 297	94.74 270	78.99 285	75.14 283	92.69 219	65.64 281	97.63 177	69.46 303	81.82 238	89.74 290
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
test2356		80.96 283	81.77 264	78.52 320	81.02 328	62.33 331	98.22 173	94.49 276	79.38 283	74.56 284	90.34 266	70.65 247	85.10 339	60.83 321	86.42 203	88.14 300

Anonymous20231206		80.76 285	79.42 281	84.79 305	84.78 319	72.98 316	96.53 237	92.97 300	79.56 282	74.33 285	88.83 288	61.27 299	92.15 322	60.59 323	75.92 259	89.24 296

FMVSNet5		82.29 267	80.54 272	87.52 288	93.79 212	84.01 252	93.73 292	92.47 312	76.92 303	74.27 286	86.15 308	63.69 289	89.24 331	69.07 304	74.79 267	89.29 295

MVS-HIRNet		79.01 293	75.13 300	90.66 230	93.82 211	81.69 273	85.16 328	93.75 289	54.54 340	74.17 287	59.15 344	57.46 307	96.58 217	63.74 316	94.38 133	93.72 205

ACMH+	83.78 15	84.21 256	82.56 260	89.15 261	93.73 213	79.16 290	96.43 241	94.28 282	81.09 271	74.00 288	94.03 189	54.58 317	97.67 174	76.10 257	78.81 249	90.63 274

NR-MVSNet		87.74 205	86.00 210	92.96 186	91.46 246	90.68 120	96.65 235	97.42 125	88.02 150	73.42 289	93.68 199	77.31 181	95.83 268	84.26 180	71.82 299	92.36 215

USDC		84.74 248	82.93 250	90.16 240	91.73 243	83.54 256	95.00 279	93.30 295	88.77 129	73.19 290	93.30 209	53.62 320	97.65 176	75.88 259	81.54 239	89.30 294

LCM-MVSNet-Re		88.59 197	88.61 178	88.51 272	95.53 164	72.68 318	96.85 226	88.43 341	88.45 136	73.14 291	90.63 257	75.82 187	94.38 295	92.95 100	95.71 124	98.48 133

TDRefinement		78.01 298	75.31 299	86.10 298	70.06 342	73.84 313	93.59 295	91.58 324	74.51 310	73.08 292	91.04 238	49.63 328	97.12 201	74.88 272	59.47 331	87.33 309

TransMVSNet (Re)		81.97 270	79.61 279	89.08 262	89.70 277	84.01 252	97.26 212	91.85 321	78.84 286	73.07 293	91.62 232	67.17 271	95.21 283	67.50 307	59.46 332	88.02 302

SixPastTwentyTwo		82.63 266	81.58 266	85.79 299	88.12 301	71.01 323	95.17 278	92.54 311	84.33 217	72.93 294	92.08 223	60.41 302	95.61 274	74.47 275	74.15 276	90.75 269

testus		77.11 302	76.95 295	77.58 321	80.02 331	58.93 337	97.78 197	90.48 331	79.68 281	72.84 295	90.61 260	37.72 342	86.57 338	60.28 325	83.18 229	87.23 311

pmmvs6		79.90 290	77.31 291	87.67 287	84.17 322	78.13 301	95.86 267	93.68 291	67.94 329	72.67 296	89.62 281	50.98 326	95.75 270	74.80 274	66.04 313	89.14 297

ACMH	83.09 17	84.60 251	82.61 258	90.57 231	93.18 225	82.94 261	96.27 246	94.92 268	81.01 272	72.61 297	93.61 202	56.54 309	97.79 164	74.31 276	81.07 240	90.99 260
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
LP		77.80 301	74.39 303	88.01 282	91.93 240	79.02 293	80.88 341	92.90 305	65.43 332	72.00 298	81.29 323	65.78 279	92.73 314	43.76 342	75.58 261	92.27 218

Patchmatch-RL test		81.90 272	80.13 273	87.23 291	80.71 329	70.12 326	84.07 335	88.19 342	83.16 246	70.57 299	82.18 312	87.18 81	92.59 317	82.28 201	62.78 318	98.98 97

DI_MVS_plusplus_test		89.41 182	87.24 197	95.92 112	89.06 290	90.75 119	98.18 178	96.63 170	89.29 112	70.54 300	90.31 267	63.50 290	98.40 138	92.25 108	95.44 127	98.60 125

test_0402		78.81 295	76.33 297	86.26 296	91.18 249	78.44 299	95.88 265	91.34 326	68.55 326	70.51 301	89.91 277	52.65 322	94.99 285	47.14 337	79.78 246	85.34 330

test_normal		89.37 183	87.18 199	95.93 111	88.94 292	90.83 115	98.24 171	96.62 171	89.31 110	70.38 302	90.20 274	63.50 290	98.37 139	92.06 110	95.41 128	98.59 128

TinyColmap		80.42 288	77.94 288	87.85 285	92.09 236	78.58 297	93.74 291	89.94 335	74.99 307	69.77 303	91.78 230	46.09 331	97.58 180	65.17 315	77.89 253	87.38 308

test20.03		78.51 297	77.48 290	81.62 315	83.07 325	71.03 322	96.11 257	92.83 308	81.66 267	69.31 304	89.68 280	57.53 306	87.29 335	58.65 328	68.47 308	86.53 315

N_pmnet		70.19 311	69.87 310	71.12 326	88.24 299	30.63 357	95.85 268	28.70 358	70.18 322	68.73 305	86.55 306	64.04 287	93.81 297	53.12 333	73.46 283	88.94 298

v18		82.00 269	79.76 277	88.72 267	90.03 261	86.81 199	96.17 256	93.12 296	78.70 287	68.39 306	82.10 313	74.64 194	93.00 302	74.21 277	60.45 325	86.35 317

OpenMVS_ROB	73.86 20	77.99 299	75.06 301	86.77 294	83.81 324	77.94 303	96.38 243	91.53 325	67.54 330	68.38 307	87.13 303	43.94 333	96.08 259	55.03 331	81.83 237	86.29 320

v17		81.87 274	79.61 279	88.64 269	89.91 266	86.64 204	96.01 260	93.08 297	78.54 288	68.27 308	81.96 315	74.44 204	92.95 304	74.03 280	60.22 327	86.34 318

v16		81.90 272	79.65 278	88.65 268	90.02 263	86.66 203	96.01 260	93.07 298	78.53 289	68.27 308	82.05 314	74.39 206	92.96 303	74.02 281	60.48 324	86.33 319

ambc				79.60 318	72.76 340	56.61 341	76.20 343	92.01 319		68.25 310	80.23 329	23.34 347	94.73 294	73.78 285	60.81 323	87.48 307

PM-MVS		74.88 305	72.85 306	80.98 317	78.98 333	64.75 330	90.81 316	85.77 345	80.95 273	68.23 311	82.81 310	29.08 345	92.84 306	76.54 255	62.46 320	85.36 329

v15		81.62 275	79.32 282	88.52 271	89.80 273	86.56 205	95.83 269	92.96 301	78.50 291	67.88 312	81.68 317	74.22 211	92.82 307	73.46 287	59.55 328	86.18 322

V14		81.55 277	79.26 283	88.42 274	89.80 273	86.33 213	95.72 272	92.96 301	78.35 292	67.82 313	81.70 316	74.13 212	92.78 311	73.32 288	59.50 330	86.16 324

v11		81.38 279	79.03 286	88.41 275	89.68 278	86.43 207	95.74 271	92.82 310	78.03 295	67.74 314	81.45 321	73.33 223	92.69 315	72.23 298	60.27 326	86.11 326

pmmvs3		72.86 308	69.76 311	82.17 311	73.86 337	74.19 312	94.20 286	89.01 338	64.23 335	67.72 315	80.91 325	41.48 336	88.65 333	62.40 318	54.02 339	83.68 334

lessismore_v0				85.08 302	85.59 317	69.28 327		90.56 330		67.68 316	90.21 273	54.21 319	95.46 276	73.88 282	62.64 319	90.50 276

V9		81.46 278	79.15 284	88.39 277	89.75 275	86.17 219	95.62 273	92.92 303	78.22 293	67.65 317	81.64 318	73.95 215	92.80 309	73.15 291	59.43 333	86.21 321

v12		81.37 280	79.05 285	88.33 278	89.68 278	86.05 225	95.48 275	92.92 303	78.08 294	67.55 318	81.58 319	73.75 216	92.75 312	73.05 292	59.37 334	86.18 322

K. test v3		81.04 282	79.77 276	84.83 304	87.41 311	70.23 325	95.60 274	93.93 287	83.70 232	67.51 319	89.35 284	55.76 311	93.58 299	76.67 253	68.03 310	90.67 273

MIMVSNet1		75.92 304	73.30 305	83.81 308	81.29 327	75.57 308	92.26 305	92.05 318	73.09 314	67.48 320	86.18 307	40.87 338	87.64 334	55.78 330	70.68 304	88.21 299

v13		81.30 281	78.99 287	88.25 279	89.61 280	85.87 229	95.39 276	92.90 305	77.93 300	67.45 321	81.52 320	73.66 217	92.75 312	72.91 294	59.53 329	86.14 325

pmmvs-eth3d		78.71 296	76.16 298	86.38 295	80.25 330	81.19 280	94.17 287	92.13 317	77.97 297	66.90 322	82.31 311	55.76 311	92.56 318	73.63 286	62.31 321	85.38 328

EG-PatchMatch MVS		79.92 289	77.59 289	86.90 293	87.06 314	77.90 304	96.20 255	94.06 286	74.61 309	66.53 323	88.76 289	40.40 340	96.20 254	67.02 309	83.66 226	86.61 314

UnsupCasMVSNet_eth		78.90 294	76.67 296	85.58 301	82.81 326	74.94 309	91.98 307	96.31 190	84.64 212	65.84 324	87.71 295	51.33 324	92.23 321	72.89 295	56.50 336	89.56 293

Test4		85.71 243	82.59 259	95.07 138	84.45 320	89.84 139	97.20 217	95.73 228	89.19 114	64.59 325	87.58 296	40.59 339	96.77 214	88.95 142	95.01 130	98.60 125

new-patchmatchnet		74.80 306	72.40 307	81.99 313	78.36 335	72.20 319	94.44 282	92.36 313	77.06 302	63.47 326	79.98 330	51.04 325	88.85 332	60.53 324	54.35 338	84.92 331

new_pmnet		76.02 303	73.71 304	82.95 310	83.88 323	72.85 317	91.26 313	92.26 314	70.44 320	62.60 327	81.37 322	47.64 330	92.32 320	61.85 319	72.10 297	83.68 334

UnsupCasMVSNet_bld		73.85 307	70.14 309	84.99 303	79.44 332	75.73 307	88.53 322	95.24 264	70.12 323	61.94 328	74.81 336	41.41 337	93.62 298	68.65 305	51.13 343	85.62 327

CMPMVS	58.40 21	80.48 287	80.11 275	81.59 316	85.10 318	59.56 335	94.14 288	95.95 212	68.54 327	60.71 329	93.31 208	55.35 315	97.87 159	83.06 194	84.85 217	87.33 309
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
1111		72.28 309	71.36 308	75.02 324	73.04 338	57.38 339	92.30 303	90.22 333	62.27 336	59.46 330	80.36 327	76.23 185	87.07 336	44.29 340	64.08 317	80.59 338

.test1245		61.50 315	64.44 314	52.65 339	73.04 338	57.38 339	92.30 303	90.22 333	62.27 336	59.46 330	80.36 327	76.23 185	87.07 336	44.29 340	1.80 354	13.50 354

Anonymous20231211		67.10 312	63.29 315	78.54 319	75.68 336	60.00 334	92.05 306	88.86 339	49.84 341	59.35 332	78.48 334	26.15 346	90.76 329	45.96 339	53.24 340	84.88 332

DeepMVS_CX				76.08 322	90.74 255	51.65 345		90.84 328	86.47 187	57.89 333	87.98 292	35.88 343	92.60 316	65.77 314	65.06 315	83.97 333

test1235678		71.07 310	69.53 312	75.71 323	71.87 341	55.27 343	94.32 283	90.76 329	70.23 321	57.61 334	79.06 332	43.13 334	83.72 341	50.48 334	68.30 309	88.14 300

testing_2		80.92 284	77.24 292	91.98 202	78.88 334	87.83 171	93.96 290	95.72 229	84.27 218	56.20 335	80.42 326	38.64 341	96.40 235	87.20 153	79.85 245	91.72 234

test12356		66.36 313	65.12 313	70.08 329	66.92 343	50.46 346	89.96 320	88.58 340	66.00 331	53.38 336	78.13 335	32.89 344	82.87 342	48.36 336	61.87 322	76.92 339

YYNet1		79.64 292	77.04 294	87.43 290	87.80 305	79.98 286	96.23 249	94.44 277	73.83 313	51.83 337	87.53 298	67.96 265	92.07 324	66.00 313	67.75 312	90.23 281

MDA-MVSNet_test_wron		79.65 291	77.05 293	87.45 289	87.79 306	80.13 285	96.25 248	94.44 277	73.87 312	51.80 338	87.47 299	68.04 263	92.12 323	66.02 312	67.79 311	90.09 282

LCM-MVSNet		60.07 317	56.37 318	71.18 325	54.81 352	48.67 347	82.17 340	89.48 337	37.95 344	49.13 339	69.12 337	13.75 356	81.76 343	59.28 326	51.63 342	83.10 336

MDA-MVSNet-bldmvs		77.82 300	74.75 302	87.03 292	88.33 298	78.52 298	96.34 244	92.85 307	75.57 306	48.87 340	87.89 293	57.32 308	92.49 319	60.79 322	64.80 316	90.08 283

PMMVS2		58.97 318	55.07 319	70.69 328	62.72 344	55.37 342	85.97 326	80.52 349	49.48 342	45.94 341	68.31 339	15.73 354	80.78 345	49.79 335	37.12 344	75.91 341

testmv		60.41 316	57.98 317	67.69 330	58.16 351	47.14 348	89.09 321	86.74 343	61.52 339	44.30 342	68.44 338	20.98 348	79.92 347	40.94 344	51.67 341	76.01 340

FPMVS		61.57 314	60.32 316	65.34 331	60.14 348	42.44 351	91.02 315	89.72 336	44.15 343	42.63 343	80.93 324	19.02 349	80.59 346	42.50 343	72.76 287	73.00 342

Gipuma		54.77 320	52.22 321	62.40 333	86.50 315	59.37 336	50.20 350	90.35 332	36.52 346	41.20 344	49.49 348	18.33 351	81.29 344	32.10 348	65.34 314	46.54 350
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
tmp_tt		53.66 321	52.86 320	56.05 336	32.75 357	41.97 353	73.42 345	76.12 352	21.91 352	39.68 345	96.39 162	42.59 335	65.10 352	78.00 239	14.92 352	61.08 347

no-one		56.69 319	51.89 322	71.08 327	59.35 350	58.65 338	83.78 338	84.81 348	61.73 338	36.46 346	56.52 346	18.15 352	84.78 340	47.03 338	19.19 348	69.81 344

E-PMN		41.02 327	40.93 326	41.29 340	61.97 345	33.83 354	84.00 336	65.17 356	27.17 349	27.56 347	46.72 350	17.63 353	60.41 354	19.32 351	18.82 349	29.61 351

PNet_i23d		48.05 323	44.98 324	57.28 335	60.15 346	42.39 352	80.85 342	73.14 354	36.78 345	27.46 348	56.66 345	6.38 357	68.34 350	36.65 346	26.72 346	61.10 346

ANet_high		50.71 322	46.17 323	64.33 332	44.27 355	52.30 344	76.13 344	78.73 350	64.95 333	27.37 349	55.23 347	14.61 355	67.74 351	36.01 347	18.23 350	72.95 343

EMVS		39.96 328	39.88 327	40.18 341	59.57 349	32.12 356	84.79 333	64.57 357	26.27 350	26.14 350	44.18 353	18.73 350	59.29 355	17.03 352	17.67 351	29.12 352

MVE	44.00 22	41.70 326	37.64 329	53.90 338	49.46 353	43.37 350	65.09 348	66.66 355	26.19 351	25.77 351	48.53 349	3.58 361	63.35 353	26.15 350	27.28 345	54.97 349
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
PMVS	41.42 23	45.67 324	42.50 325	55.17 337	34.28 356	32.37 355	66.24 347	78.71 351	30.72 348	22.04 352	59.59 343	4.59 358	77.85 348	27.49 349	58.84 335	55.29 348
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
testmvs		18.81 331	23.05 332	6.10 345	4.48 358	2.29 360	97.78 197	3.00 360	3.27 354	18.60 353	62.71 341	1.53 363	2.49 358	14.26 354	1.80 354	13.50 354

wuykxyi23d		43.53 325	37.95 328	60.27 334	45.36 354	44.79 349	68.27 346	74.26 353	33.48 347	18.21 354	40.16 355	3.64 359	71.01 349	38.85 345	19.31 347	65.02 345

test123		16.58 333	19.47 333	7.91 344	3.59 359	5.37 359	94.32 283	1.39 361	2.49 355	13.98 355	44.60 352	2.91 362	2.65 357	11.35 355	0.57 356	15.70 353

wuyk23d		16.71 332	16.73 334	16.65 343	60.15 346	25.22 358	41.24 351	5.17 359	6.56 353	5.48 356	3.61 357	3.64 359	22.72 356	15.20 353	9.52 353	1.99 356

cdsmvs_eth3d_5k		22.52 330	30.03 331	0.00 346	0.00 360	0.00 361	0.00 352	97.17 143	0.00 356	0.00 357	98.77 65	74.35 207	0.00 359	0.00 356	0.00 357	0.00 357

pcd_1.5k_mvsjas		6.87 335	9.16 336	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	0.00 358	82.48 146	0.00 359	0.00 356	0.00 357	0.00 357

pcd1.5k->3k		35.91 329	37.64 329	30.74 342	89.49 285	0.00 361	0.00 352	96.36 189	0.00 356	0.00 357	0.00 358	69.17 255	0.00 359	0.00 356	83.71 225	92.21 223

sosnet-low-res		0.00 336	0.00 337	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	0.00 358	0.00 364	0.00 359	0.00 356	0.00 357	0.00 357

sosnet		0.00 336	0.00 337	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	0.00 358	0.00 364	0.00 359	0.00 356	0.00 357	0.00 357

uncertanet		0.00 336	0.00 337	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	0.00 358	0.00 364	0.00 359	0.00 356	0.00 357	0.00 357

Regformer		0.00 336	0.00 337	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	0.00 358	0.00 364	0.00 359	0.00 356	0.00 357	0.00 357

ab-mvs-re		8.21 334	10.94 335	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	98.50 84	0.00 364	0.00 359	0.00 356	0.00 357	0.00 357

uanet		0.00 336	0.00 337	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	0.00 358	0.00 364	0.00 359	0.00 356	0.00 357	0.00 357

GSMVS																98.84 109

test_part3							99.43 33		92.81 44		99.48 4		99.97 14	99.52 1

test_part1								97.69 79				93.96 6			99.83 12	99.90 9

sam_mvs1												88.39 59				98.84 109

sam_mvs												87.08 82

MTGPA								97.45 119

test_post1							90.74 318				41.37 354	85.38 110	96.36 238	83.16 192

test_post											46.00 351	87.37 75	97.11 202

patchmatchnet-post											84.86 309	88.73 53	96.81 213

MTMP								91.09 327

gm-plane-assit					94.69 191	88.14 165			88.22 146		97.20 128		98.29 141	90.79 120

test9_res														98.60 11	99.87 5	99.90 9

agg_prior2														97.84 28	99.87 5	99.91 8

test_prior4						92.00 81	99.41 38

test_prior				97.01 51	99.58 19	91.77 82		97.57 102					99.49 87			99.79 26

新几何2							98.26 169

旧先验1					98.97 60	92.90 69		97.74 75			99.15 27	91.05 20			99.33 53	99.60 59

无先验							98.52 136	97.82 63	87.20 174				99.90 31	87.64 151		99.85 21

原ACMM2							98.69 113

testdata2													99.88 35	84.16 181

segment_acmp												90.56 35

testdata1							97.89 193		92.43 50

plane_prior7					93.84 209	85.73 233

plane_prior6					93.92 206	86.02 226						72.92 226

plane_prior5								96.30 191					97.75 171	93.46 93	86.17 207	92.67 211

plane_prior4											96.52 156

plane_prior2							99.02 78		93.38 35

plane_prior1					93.90 208

plane_prior						86.07 223	99.14 66		93.81 28						86.26 206

n2								0.00 362

nn								0.00 362

door-mid								84.90 347

test11								97.68 81

door								85.30 346

HQP5-MVS						86.39 210

BP-MVS														93.82 88

HQP3-MVS								96.37 186							86.29 204

HQP2-MVS												73.34 221

NP-MVS					93.94 205	86.22 217					96.67 150

ACMMP++_ref															82.64 234

ACMMP++															83.83 223

Test By Simon												83.62 123

High-resolution multi-view benchmark