High-res multi-view results

This table lists the benchmark results for the high-res multi-view scenario. The following metrics are evaluated:

Accuracy [%]: The fraction of the reconstruction which is closer to the ground truth than the evaluation threshold distance (*). Larger is better.
Completeness [%]: The fraction of the ground truth which is closer to the reconstruction than the evaluation threshold distance (*). Larger is better.
F₁ score [%]: The harmonic mean of accuracy and completeness, used to rank methods based on both metrics. Larger is better.
Time [s]: The runtime of the method. Smaller is better.

(*) For exact definitions, detailing how potentially incomplete ground truth is taken into account, see our paper.

The datasets are grouped into different categories, and result averages are computed for a category and method if results of the method are available for all datasets within the category. Note that the category "all" includes both the high-res multi-view and the low-res many-view scenarios.

Methods with suffix _ROB may participate in the Robust Vision Challenge.

Click a dataset result cell to show a visualization of the reconstruction. For training datasets, ground truth and accuracy / completeness visualizations are also available. The visualizations may not work with mobile browsers.

Method	all	high-res multi-view	indoor	outdoor	botani.	boulde.	bridge	door	exhibi.	lectur.	living.	lounge	observ.	old co.	statue	terrac.

APDe-MVS		97.82 1	97.73 1	98.08 9	99.15 25	94.82 13	98.81 2	98.30 22	94.76 24	98.30 5	98.90 2	93.77 8	99.68 38	97.93 1	99.69 1	99.75 1

MVS_0304		96.05 51	95.45 53	97.85 15	97.75 106	94.50 16	96.87 155	97.95 82	95.46 6	95.60 74	98.01 49	80.96 195	99.83 15	97.23 2	99.25 47	99.23 50

TSAR-MVS + MP.		97.42 6	97.33 6	97.69 29	99.25 20	94.24 24	98.07 34	97.85 90	93.72 47	98.57 3	98.35 25	93.69 9	99.40 90	97.06 3	99.46 26	99.44 33

CNVR-MVS		97.68 2	97.44 5	98.37 3	98.90 33	95.86 2	97.27 119	98.08 51	95.81 3	97.87 12	98.31 34	94.26 4	99.68 38	97.02 4	99.49 24	99.57 14

SD-MVS		97.41 7	97.53 2	97.06 57	98.57 52	94.46 17	97.92 46	98.14 41	94.82 21	99.01 1	98.55 10	94.18 5	97.41 284	96.94 5	99.64 4	99.32 44

Regformer-4		96.97 23	96.87 17	97.25 49	98.34 63	92.66 67	96.96 145	98.01 70	95.12 13	97.14 24	98.42 19	91.82 35	99.61 48	96.90 6	99.13 57	99.50 25

CANet		96.39 43	96.02 45	97.50 39	97.62 113	93.38 50	97.02 140	97.96 80	95.42 8	94.86 84	97.81 62	87.38 91	99.82 19	96.88 7	99.20 52	99.29 46

TSAR-MVS + GP.		96.69 34	96.49 33	97.27 48	98.31 68	93.39 49	96.79 166	96.72 202	94.17 36	97.44 16	97.66 72	92.76 14	99.33 95	96.86 8	97.76 96	99.08 63

Regformer-3		96.85 28	96.80 23	97.01 58	98.34 63	92.02 85	96.96 145	97.76 93	95.01 16	97.08 29	98.42 19	91.71 36	99.54 69	96.80 9	99.13 57	99.48 29

Regformer-2		97.16 14	96.99 12	97.67 30	98.32 66	93.84 36	96.83 159	98.10 48	95.24 10	97.49 14	98.25 40	92.57 20	99.61 48	96.80 9	99.29 44	99.56 16

Regformer-1		97.10 16	96.96 14	97.54 38	98.32 66	93.48 47	96.83 159	97.99 77	95.20 12	97.46 15	98.25 40	92.48 23	99.58 56	96.79 11	99.29 44	99.55 18

DeepPCF-MVS	93.97 1	96.61 37	97.09 8	95.15 140	98.09 84	86.63 261	96.00 235	98.15 39	95.43 7	97.95 10	98.56 8	93.40 10	99.36 94	96.77 12	99.48 25	99.45 31

SMA-MVS		97.34 8	97.03 10	98.28 4	99.02 30	95.42 5	97.94 44	98.18 35	90.57 147	98.85 2	98.93 1	93.33 11	99.83 15	96.76 13	99.68 2	99.60 10

HSP-MVS		97.53 5	97.49 4	97.63 35	99.40 5	93.77 41	98.53 9	97.85 90	95.55 5	98.56 4	97.81 62	93.90 6	99.65 42	96.62 14	99.21 51	99.48 29

MSLP-MVS++		96.94 25	97.06 9	96.59 69	98.72 38	91.86 89	97.67 71	98.49 12	94.66 27	97.24 19	98.41 22	92.31 27	98.94 131	96.61 15	99.46 26	98.96 73

MP-MVS-pluss		96.70 33	96.27 40	97.98 11	99.23 23	94.71 14	96.96 145	98.06 58	90.67 138	95.55 76	98.78 3	91.07 45	99.86 8	96.58 16	99.55 15	99.38 40
MP-MVS-pluss: MP-MVS-pluss. MP-MVS-pluss
SteuartSystems-ACMMP		97.62 3	97.53 2	97.87 14	98.39 60	94.25 23	98.43 16	98.27 24	95.34 9	98.11 6	98.56 8	94.53 3	99.71 30	96.57 17	99.62 8	99.65 3
Skip Steuart: Steuart Systems R&D Blog.
MCST-MVS		97.18 12	96.84 19	98.20 6	99.30 16	95.35 6	97.12 135	98.07 56	93.54 53	96.08 54	97.69 69	93.86 7	99.71 30	96.50 18	99.39 35	99.55 18

EI-MVSNet-Vis-set		96.51 39	96.47 34	96.63 66	98.24 72	91.20 110	96.89 154	97.73 96	94.74 25	96.49 42	98.49 14	90.88 50	99.58 56	96.44 19	98.32 81	99.13 58

VDD-MVS		93.82 106	93.08 110	96.02 100	97.88 100	89.96 145	97.72 65	95.85 244	92.43 86	95.86 63	98.44 17	68.42 317	99.39 91	96.31 20	94.85 150	98.71 93

ACMMP_Plus		97.20 11	96.86 18	98.23 5	99.09 26	95.16 9	97.60 87	98.19 33	92.82 78	97.93 11	98.74 4	91.60 39	99.86 8	96.26 21	99.52 18	99.67 2

EI-MVSNet-UG-set		96.34 44	96.30 39	96.47 77	98.20 77	90.93 121	96.86 156	97.72 99	94.67 26	96.16 51	98.46 15	90.43 54	99.58 56	96.23 22	97.96 90	98.90 80

xiu_mvs_v1_base_debu		95.01 72	94.76 68	95.75 111	96.58 152	91.71 90	96.25 220	97.35 146	92.99 69	96.70 32	96.63 125	82.67 162	99.44 85	96.22 23	97.46 100	96.11 199

xiu_mvs_v1_base		95.01 72	94.76 68	95.75 111	96.58 152	91.71 90	96.25 220	97.35 146	92.99 69	96.70 32	96.63 125	82.67 162	99.44 85	96.22 23	97.46 100	96.11 199

xiu_mvs_v1_base_debi		95.01 72	94.76 68	95.75 111	96.58 152	91.71 90	96.25 220	97.35 146	92.99 69	96.70 32	96.63 125	82.67 162	99.44 85	96.22 23	97.46 100	96.11 199

alignmvs		95.87 57	95.23 60	97.78 21	97.56 118	95.19 8	97.86 50	97.17 157	94.39 32	96.47 43	96.40 137	85.89 107	99.20 102	96.21 26	95.11 148	98.95 75

canonicalmvs		96.02 53	95.45 53	97.75 25	97.59 116	95.15 10	98.28 22	97.60 111	94.52 29	96.27 48	96.12 148	87.65 85	99.18 105	96.20 27	94.82 152	98.91 79

zzz-MVS		97.07 18	96.77 25	97.97 12	99.37 10	94.42 19	97.15 133	98.08 51	95.07 14	96.11 52	98.59 6	90.88 50	99.90 1	96.18 28	99.50 22	99.58 12

MTAPA		97.08 17	96.78 24	97.97 12	99.37 10	94.42 19	97.24 121	98.08 51	95.07 14	96.11 52	98.59 6	90.88 50	99.90 1	96.18 28	99.50 22	99.58 12

APD-MVS_3200maxsize		96.81 29	96.71 27	97.12 56	99.01 31	92.31 74	97.98 43	98.06 58	93.11 66	97.44 16	98.55 10	90.93 48	99.55 67	96.06 30	99.25 47	99.51 24

MVS_111021_HR		96.68 36	96.58 31	96.99 59	98.46 54	92.31 74	96.20 225	98.90 2	94.30 35	95.86 63	97.74 67	92.33 24	99.38 93	96.04 31	99.42 31	99.28 49

PHI-MVS		96.77 31	96.46 35	97.71 28	98.40 58	94.07 30	98.21 28	98.45 15	89.86 157	97.11 27	98.01 49	92.52 22	99.69 36	96.03 32	99.53 17	99.36 42

HPM-MVS++		97.34 8	96.97 13	98.47 1	99.08 27	96.16 1	97.55 93	97.97 79	95.59 4	96.61 36	97.89 53	92.57 20	99.84 14	95.95 33	99.51 20	99.40 36

DELS-MVS		96.61 37	96.38 38	97.30 45	97.79 103	93.19 54	95.96 236	98.18 35	95.23 11	95.87 62	97.65 73	91.45 41	99.70 35	95.87 34	99.44 30	99.00 71
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_LR		96.24 47	96.19 44	96.39 83	98.23 76	91.35 104	96.24 223	98.79 4	93.99 39	95.80 66	97.65 73	89.92 61	99.24 101	95.87 34	99.20 52	98.58 99

NCCC		97.30 10	97.03 10	98.11 8	98.77 36	95.06 11	97.34 113	98.04 65	95.96 2	97.09 28	97.88 55	93.18 12	99.71 30	95.84 36	99.17 54	99.56 16

VNet		95.89 56	95.45 53	97.21 53	98.07 85	92.94 61	97.50 96	98.15 39	93.87 41	97.52 13	97.61 79	85.29 113	99.53 72	95.81 37	95.27 145	99.16 54

XVS		97.18 12	96.96 14	97.81 18	99.38 8	94.03 32	98.59 7	98.20 31	94.85 17	96.59 38	98.29 37	91.70 37	99.80 21	95.66 38	99.40 33	99.62 7

X-MVStestdata		91.71 181	89.67 239	97.81 18	99.38 8	94.03 32	98.59 7	98.20 31	94.85 17	96.59 38	32.69 359	91.70 37	99.80 21	95.66 38	99.40 33	99.62 7

casdiffmvs		95.23 66	94.84 66	96.40 81	96.90 140	91.71 90	97.36 111	97.30 150	91.02 131	94.81 85	96.18 144	87.74 82	98.77 146	95.65 40	96.55 127	98.71 93

HFP-MVS		97.14 15	96.92 16	97.83 16	99.42 3	94.12 28	98.52 10	98.32 19	93.21 60	97.18 21	98.29 37	92.08 29	99.83 15	95.63 41	99.59 10	99.54 20

ACMMPR		97.07 18	96.84 19	97.79 20	99.44 2	93.88 34	98.52 10	98.31 21	93.21 60	97.15 23	98.33 31	91.35 42	99.86 8	95.63 41	99.59 10	99.62 7

HPM-MVS		96.69 34	96.45 36	97.40 41	99.36 12	93.11 56	98.87 1	98.06 58	91.17 126	96.40 46	97.99 51	90.99 47	99.58 56	95.61 43	99.61 9	99.49 27
Chunlin Ren, Qingshan Xu, Shikun Zhang, Jiaqi Yang: Hierarchical Prior Mining for Non-local Multi-View Stereo. ICCV 2023
CP-MVS		97.02 21	96.81 22	97.64 33	99.33 14	93.54 45	98.80 3	98.28 23	92.99 69	96.45 45	98.30 36	91.90 34	99.85 11	95.61 43	99.68 2	99.54 20

DeepC-MVS	93.07 3	96.06 50	95.66 51	97.29 46	97.96 91	93.17 55	97.30 118	98.06 58	93.92 40	93.38 111	98.66 5	86.83 96	99.73 26	95.60 45	99.22 50	98.96 73
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
region2R		97.07 18	96.84 19	97.77 23	99.46 1	93.79 38	98.52 10	98.24 28	93.19 63	97.14 24	98.34 28	91.59 40	99.87 5	95.46 46	99.59 10	99.64 4

lupinMVS		94.99 76	94.56 74	96.29 91	96.34 166	91.21 108	95.83 242	96.27 221	88.93 189	96.22 49	96.88 109	86.20 104	98.85 139	95.27 47	99.05 63	98.82 87

mPP-MVS		96.86 27	96.60 29	97.64 33	99.40 5	93.44 48	98.50 13	98.09 50	93.27 59	95.95 61	98.33 31	91.04 46	99.88 3	95.20 48	99.57 14	99.60 10

test_part3							97.50 96		93.81 45		98.53 12		99.87 5	95.19 49

ESAPD		97.57 4	97.29 7	98.41 2	99.28 17	95.74 3	97.50 96	98.26 25	93.81 45	98.10 7	98.53 12	95.31 1	99.87 5	95.19 49	99.63 5	99.63 5

DeepC-MVS_fast	93.89 2	96.93 26	96.64 28	97.78 21	98.64 47	94.30 21	97.41 104	98.04 65	94.81 22	96.59 38	98.37 24	91.24 43	99.64 47	95.16 51	99.52 18	99.42 35
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
jason		94.84 81	94.39 82	96.18 96	95.52 196	90.93 121	96.09 229	96.52 214	89.28 170	96.01 59	97.32 92	84.70 120	98.77 146	95.15 52	98.91 69	98.85 84
jason: jason.
#test#		97.02 21	96.75 26	97.83 16	99.42 3	94.12 28	98.15 29	98.32 19	92.57 83	97.18 21	98.29 37	92.08 29	99.83 15	95.12 53	99.59 10	99.54 20

abl_6		96.40 42	96.21 42	96.98 60	98.89 34	92.20 79	97.89 48	98.03 67	93.34 58	97.22 20	98.42 19	87.93 80	99.72 29	95.10 54	99.07 62	99.02 65

train_agg		96.30 45	95.83 48	97.72 26	98.70 39	94.19 25	96.41 201	98.02 68	88.58 201	96.03 55	97.56 84	92.73 16	99.59 53	95.04 55	99.37 40	99.39 37

agg_prior3		96.16 49	95.67 50	97.62 36	98.67 41	93.88 34	96.41 201	98.00 72	87.93 226	95.81 65	97.47 88	92.33 24	99.59 53	95.04 55	99.37 40	99.39 37

agg_prior1		96.22 48	95.77 49	97.56 37	98.67 41	93.79 38	96.28 217	98.00 72	88.76 198	95.68 70	97.55 86	92.70 18	99.57 64	95.01 57	99.32 42	99.32 44

test_prior3		96.46 41	96.20 43	97.23 50	98.67 41	92.99 58	96.35 209	98.00 72	92.80 79	96.03 55	97.59 80	92.01 31	99.41 88	95.01 57	99.38 36	99.29 46

test_prior2							96.35 209		92.80 79	96.03 55	97.59 80	92.01 31		95.01 57	99.38 36

nrg030		94.05 99	93.31 107	96.27 92	95.22 214	94.59 15	98.34 19	97.46 128	92.93 76	91.21 172	96.64 121	87.23 93	98.22 191	94.99 60	85.80 266	95.98 207

VDDNet		93.05 129	92.07 137	96.02 100	96.84 143	90.39 135	98.08 33	95.85 244	86.22 264	95.79 67	98.46 15	67.59 320	99.19 103	94.92 61	94.85 150	98.47 112

APD-MVS		96.95 24	96.60 29	98.01 10	99.03 29	94.93 12	97.72 65	98.10 48	91.50 114	98.01 9	98.32 33	92.33 24	99.58 56	94.85 62	99.51 20	99.53 23
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
MP-MVS		96.77 31	96.45 36	97.72 26	99.39 7	93.80 37	98.41 17	98.06 58	93.37 55	95.54 77	98.34 28	90.59 53	99.88 3	94.83 63	99.54 16	99.49 27
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
test9_res														94.81 64	99.38 36	99.45 31

PS-MVSNAJ		95.37 62	95.33 58	95.49 125	97.35 126	90.66 129	95.31 264	97.48 123	93.85 42	96.51 41	95.70 173	88.65 71	99.65 42	94.80 65	98.27 82	96.17 194

HPM-MVS_fast		96.51 39	96.27 40	97.22 52	99.32 15	92.74 64	98.74 4	98.06 58	90.57 147	96.77 31	98.35 25	90.21 57	99.53 72	94.80 65	99.63 5	99.38 40

xiu_mvs_v2_base		95.32 64	95.29 59	95.40 130	97.22 128	90.50 132	95.44 259	97.44 135	93.70 49	96.46 44	96.18 144	88.59 74	99.53 72	94.79 67	97.81 93	96.17 194

CSCG		96.05 51	95.91 47	96.46 79	99.24 21	90.47 133	98.30 21	98.57 11	89.01 183	93.97 100	97.57 82	92.62 19	99.76 24	94.66 68	99.27 46	99.15 56

ACMMP		96.27 46	95.93 46	97.28 47	99.24 21	92.62 68	98.25 25	98.81 3	92.99 69	94.56 89	98.39 23	88.96 66	99.85 11	94.57 69	97.63 97	99.36 42
Qingshan Xu, Weihang Kong, Wenbing Tao, Marc Pollefeys: Multi-Scale Geometric Consistency Guided and Planar Prior Assisted Multi-View Stereo. IEEE Transactions on Pattern Analysis and Machine Intelligence
PGM-MVS		96.81 29	96.53 32	97.65 31	99.35 13	93.53 46	97.65 74	98.98 1	92.22 89	97.14 24	98.44 17	91.17 44	99.85 11	94.35 70	99.46 26	99.57 14

LFMVS		93.60 113	92.63 123	96.52 71	98.13 83	91.27 106	97.94 44	93.39 323	90.57 147	96.29 47	98.31 34	69.00 313	99.16 107	94.18 71	95.87 137	99.12 60

MVSFormer		95.37 62	95.16 62	95.99 102	96.34 166	91.21 108	98.22 26	97.57 114	91.42 118	96.22 49	97.32 92	86.20 104	97.92 245	94.07 72	99.05 63	98.85 84

test_djsdf		93.07 128	92.76 117	94.00 195	93.49 293	88.70 198	98.22 26	97.57 114	91.42 118	90.08 196	95.55 180	82.85 159	97.92 245	94.07 72	91.58 209	95.40 237

mvs_anonymous		93.82 106	93.74 90	94.06 192	96.44 163	85.41 273	95.81 243	97.05 173	89.85 159	90.09 195	96.36 139	87.44 90	97.75 261	93.97 74	96.69 123	99.02 65

VPA-MVSNet		93.24 123	92.48 132	95.51 123	95.70 192	92.39 73	97.86 50	98.66 9	92.30 88	92.09 144	95.37 189	80.49 207	98.40 179	93.95 75	85.86 265	95.75 220

agg_prior2														93.94 76	99.38 36	99.50 25

mvs_tets		92.31 159	91.76 146	93.94 204	93.41 295	88.29 205	97.63 85	97.53 118	92.04 103	88.76 234	96.45 135	74.62 284	98.09 205	93.91 77	91.48 211	95.45 231

Effi-MVS+		94.93 77	94.45 80	96.36 86	96.61 150	91.47 100	96.41 201	97.41 139	91.02 131	94.50 90	95.92 156	87.53 88	98.78 144	93.89 78	96.81 118	98.84 86

jajsoiax		92.42 154	91.89 144	94.03 194	93.33 299	88.50 202	97.73 63	97.53 118	92.00 105	88.85 233	96.50 133	75.62 277	98.11 202	93.88 79	91.56 210	95.48 227

XVG-OURS-SEG-HR		93.86 105	93.55 96	94.81 160	97.06 136	88.53 201	95.28 265	97.45 132	91.68 111	94.08 97	97.68 70	82.41 171	98.90 134	93.84 80	92.47 193	96.98 164

PS-MVSNAJss		93.74 109	93.51 99	94.44 178	93.91 280	89.28 185	97.75 59	97.56 117	92.50 85	89.94 198	96.54 131	88.65 71	98.18 195	93.83 81	90.90 221	95.86 209

EPNet		95.20 69	94.56 74	97.14 55	92.80 312	92.68 66	97.85 53	94.87 291	96.64 1	92.46 133	97.80 64	86.23 102	99.65 42	93.72 82	98.62 75	99.10 62
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
test_normal		92.01 170	90.75 193	95.80 108	93.24 301	89.97 143	95.93 238	96.24 224	90.62 142	81.63 308	93.45 279	74.98 281	98.89 136	93.61 83	97.04 114	98.55 100

PVSNet_Blended_VisFu		95.27 65	94.91 65	96.38 84	98.20 77	90.86 123	97.27 119	98.25 27	90.21 151	94.18 96	97.27 94	87.48 89	99.73 26	93.53 84	97.77 95	98.55 100

DI_MVS_plusplus_test		92.01 170	90.77 191	95.73 114	93.34 297	89.78 150	96.14 227	96.18 227	90.58 146	81.80 307	93.50 276	74.95 282	98.90 134	93.51 85	96.94 115	98.51 105

CPTT-MVS		95.57 60	95.19 61	96.70 63	99.27 19	91.48 99	98.33 20	98.11 46	87.79 229	95.17 81	98.03 47	87.09 94	99.61 48	93.51 85	99.42 31	99.02 65

MVSTER		93.20 124	92.81 116	94.37 181	96.56 155	89.59 161	97.06 137	97.12 163	91.24 124	91.30 161	95.96 153	82.02 179	98.05 218	93.48 87	90.55 226	95.47 229

PVSNet_BlendedMVS		94.06 98	93.92 86	94.47 177	98.27 69	89.46 169	96.73 171	98.36 16	90.17 152	94.36 92	95.24 194	88.02 77	99.58 56	93.44 88	90.72 224	94.36 292

PVSNet_Blended		94.87 80	94.56 74	95.81 107	98.27 69	89.46 169	95.47 258	98.36 16	88.84 192	94.36 92	96.09 151	88.02 77	99.58 56	93.44 88	98.18 84	98.40 119

3Dnovator	91.36 5	95.19 70	94.44 81	97.44 40	96.56 155	93.36 52	98.65 6	98.36 16	94.12 37	89.25 229	98.06 46	82.20 176	99.77 23	93.41 90	99.32 42	99.18 53

EPP-MVSNet		95.22 68	95.04 64	95.76 109	97.49 125	89.56 162	98.67 5	97.00 181	90.69 137	94.24 95	97.62 78	89.79 62	98.81 142	93.39 91	96.49 129	98.92 78

CHOSEN 280x420		93.12 126	92.72 121	94.34 183	96.71 149	87.27 244	90.29 334	97.72 99	86.61 260	91.34 158	95.29 191	84.29 126	98.41 178	93.25 92	98.94 68	97.35 160

3Dnovator+	91.43 4	95.40 61	94.48 79	98.16 7	96.90 140	95.34 7	98.48 14	97.87 87	94.65 28	88.53 239	98.02 48	83.69 130	99.71 30	93.18 93	98.96 67	99.44 33

0601test		94.78 83	94.23 83	96.43 80	97.74 107	91.22 107	96.85 157	97.10 166	91.23 125	95.71 69	96.93 106	84.30 125	99.31 97	93.10 94	95.12 147	98.75 88

HQP_MVS		93.78 108	93.43 103	94.82 158	96.21 171	89.99 140	97.74 61	97.51 121	94.85 17	91.34 158	96.64 121	81.32 191	98.60 159	93.02 95	92.23 196	95.86 209

plane_prior5								97.51 121					98.60 159	93.02 95	92.23 196	95.86 209

MVS_Test		94.89 79	94.62 72	95.68 116	96.83 145	89.55 163	96.70 179	97.17 157	91.17 126	95.60 74	96.11 150	87.87 81	98.76 148	93.01 97	97.17 111	98.72 91

diffmvs		94.47 87	94.23 83	95.18 134	96.32 168	88.22 211	96.27 218	97.04 176	92.55 84	93.60 104	95.94 155	86.79 97	98.70 153	92.98 98	96.61 125	98.63 98

CLD-MVS		92.98 131	92.53 129	94.32 184	96.12 180	89.20 188	95.28 265	97.47 126	92.66 81	89.90 199	95.62 176	80.58 205	98.40 179	92.73 99	92.40 194	95.38 239
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
XVG-OURS		93.72 110	93.35 106	94.80 161	97.07 134	88.61 199	94.79 274	97.46 128	91.97 106	93.99 98	97.86 58	81.74 185	98.88 138	92.64 100	92.67 192	96.92 172

旧先验2							95.94 237		81.66 310	97.34 18			98.82 141	92.26 101

CDPH-MVS		95.97 54	95.38 56	97.77 23	98.93 32	94.44 18	96.35 209	97.88 85	86.98 250	96.65 35	97.89 53	91.99 33	99.47 81	92.26 101	99.46 26	99.39 37

FIs		94.09 97	93.70 91	95.27 132	95.70 192	92.03 84	98.10 31	98.68 7	93.36 57	90.39 182	96.70 116	87.63 86	97.94 241	92.25 103	90.50 228	95.84 212

LPG-MVS_test		92.94 133	92.56 126	94.10 190	96.16 176	88.26 207	97.65 74	97.46 128	91.29 121	90.12 192	97.16 99	79.05 229	98.73 150	92.25 103	91.89 204	95.31 243

LGP-MVS_train				94.10 190	96.16 176	88.26 207		97.46 128	91.29 121	90.12 192	97.16 99	79.05 229	98.73 150	92.25 103	91.89 204	95.31 243

cascas		91.20 214	90.08 222	94.58 175	94.97 226	89.16 191	93.65 299	97.59 113	79.90 323	89.40 221	92.92 286	75.36 278	98.36 182	92.14 106	94.75 154	96.23 191

OPM-MVS		93.28 122	92.76 117	94.82 158	94.63 243	90.77 127	96.65 184	97.18 155	93.72 47	91.68 151	97.26 95	79.33 226	98.63 156	92.13 107	92.28 195	95.07 256

BP-MVS														92.13 107

HQP-MVS		93.19 125	92.74 120	94.54 176	95.86 185	89.33 180	96.65 184	97.39 140	93.55 50	90.14 186	95.87 158	80.95 196	98.50 167	92.13 107	92.10 201	95.78 216

DP-MVS Recon		95.68 58	95.12 63	97.37 42	99.19 24	94.19 25	97.03 138	98.08 51	88.35 215	95.09 82	97.65 73	89.97 60	99.48 80	92.08 110	98.59 76	98.44 116

testing_2		87.33 289	85.03 296	94.22 186	87.77 340	89.32 182	94.97 272	97.11 165	89.22 172	71.64 341	88.73 328	55.16 347	97.94 241	91.95 111	88.73 245	95.41 233

Test4		89.48 257	87.50 268	95.44 129	90.76 328	89.72 151	95.78 246	97.09 167	90.28 150	77.67 333	91.74 307	55.42 346	98.08 206	91.92 112	96.83 117	98.52 103

VPNet		92.23 164	91.31 169	94.99 148	95.56 195	90.96 119	97.22 126	97.86 89	92.96 75	90.96 174	96.62 128	75.06 280	98.20 192	91.90 113	83.65 302	95.80 215

sss		94.51 86	93.80 89	96.64 64	97.07 134	91.97 87	96.32 213	98.06 58	88.94 188	94.50 90	96.78 111	84.60 121	99.27 99	91.90 113	96.02 133	98.68 96

anonymousdsp		92.16 166	91.55 160	93.97 199	92.58 316	89.55 163	97.51 95	97.42 138	89.42 168	88.40 240	94.84 209	80.66 204	97.88 250	91.87 115	91.28 215	94.48 288

ACMP	89.59 10	92.62 143	92.14 136	94.05 193	96.40 164	88.20 214	97.36 111	97.25 154	91.52 113	88.30 243	96.64 121	78.46 248	98.72 152	91.86 116	91.48 211	95.23 250
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
HyFIR lowres test		93.66 111	92.92 114	95.87 105	98.24 72	89.88 147	94.58 277	98.49 12	85.06 278	93.78 101	95.78 167	82.86 158	98.67 154	91.77 117	95.71 141	99.07 64

UGNet		94.04 100	93.28 108	96.31 88	96.85 142	91.19 111	97.88 49	97.68 105	94.40 31	93.00 125	96.18 144	73.39 294	99.61 48	91.72 118	98.46 78	98.13 128
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
UniMVSNet_NR-MVSNet		93.37 119	92.67 122	95.47 127	95.34 204	92.83 62	97.17 131	98.58 10	92.98 74	90.13 190	95.80 163	88.37 76	97.85 251	91.71 119	83.93 296	95.73 222

DU-MVS		92.90 135	92.04 138	95.49 125	94.95 228	92.83 62	97.16 132	98.24 28	93.02 68	90.13 190	95.71 171	83.47 132	97.85 251	91.71 119	83.93 296	95.78 216

Effi-MVS+-dtu		93.08 127	93.21 109	92.68 259	96.02 182	83.25 296	97.14 134	96.72 202	93.85 42	91.20 173	93.44 280	83.08 143	98.30 188	91.69 121	95.73 140	96.50 186

mvs-test1		93.63 112	93.69 92	93.46 233	96.02 182	84.61 283	97.24 121	96.72 202	93.85 42	92.30 139	95.76 168	83.08 143	98.89 136	91.69 121	96.54 128	96.87 174

UniMVSNet (Re)		93.31 121	92.55 127	95.61 118	95.39 201	93.34 53	97.39 108	98.71 5	93.14 65	90.10 194	94.83 211	87.71 83	98.03 224	91.67 123	83.99 295	95.46 230

LCM-MVSNet-Re		92.50 148	92.52 130	92.44 262	96.82 146	81.89 305	96.92 152	93.71 318	92.41 87	84.30 292	94.60 222	85.08 116	97.03 297	91.51 124	97.36 106	98.40 119

FC-MVSNet-test		93.94 103	93.57 95	95.04 146	95.48 198	91.45 102	98.12 30	98.71 5	93.37 55	90.23 185	96.70 116	87.66 84	97.85 251	91.49 125	90.39 229	95.83 213

PMMVS		92.86 137	92.34 134	94.42 180	94.92 231	86.73 257	94.53 279	96.38 217	84.78 283	94.27 94	95.12 199	83.13 139	98.40 179	91.47 126	96.49 129	98.12 129

Vis-MVSNet		95.23 66	94.81 67	96.51 74	97.18 130	91.58 98	98.26 24	98.12 43	94.38 33	94.90 83	98.15 42	82.28 173	98.92 132	91.45 127	98.58 77	99.01 69
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
CHOSEN 1792x2688		94.15 93	93.51 99	96.06 98	98.27 69	89.38 176	95.18 270	98.48 14	85.60 271	93.76 102	97.11 102	83.15 137	99.61 48	91.33 128	98.72 73	99.19 52

OMC-MVS		95.09 71	94.70 71	96.25 94	98.46 54	91.28 105	96.43 199	97.57 114	92.04 103	94.77 87	97.96 52	87.01 95	99.09 121	91.31 129	96.77 119	98.36 123

MG-MVS		95.61 59	95.38 56	96.31 88	98.42 57	90.53 131	96.04 231	97.48 123	93.47 54	95.67 73	98.10 43	89.17 64	99.25 100	91.27 130	98.77 71	99.13 58

ACMM	89.79 8	92.96 132	92.50 131	94.35 182	96.30 169	88.71 197	97.58 90	97.36 145	91.40 120	90.53 178	96.65 120	79.77 219	98.75 149	91.24 131	91.64 207	95.59 225
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
WTY-MVS		94.71 84	94.02 85	96.79 62	97.71 109	92.05 83	96.59 192	97.35 146	90.61 144	94.64 88	96.93 106	86.41 101	99.39 91	91.20 132	94.71 156	98.94 76

Anonymous20240529		91.98 174	90.73 194	95.73 114	98.14 82	89.40 175	97.99 42	97.72 99	79.63 324	93.54 107	97.41 91	69.94 311	99.56 66	91.04 133	91.11 217	98.22 125

CANet_DTU		94.37 88	93.65 94	96.55 70	96.46 162	92.13 81	96.21 224	96.67 209	94.38 33	93.53 108	97.03 105	79.34 225	99.71 30	90.76 134	98.45 79	97.82 143

ab-mvs		93.57 115	92.55 127	96.64 64	97.28 127	91.96 88	95.40 260	97.45 132	89.81 161	93.22 119	96.28 141	79.62 222	99.46 82	90.74 135	93.11 186	98.50 107

CostFormer		91.18 217	90.70 196	92.62 260	94.84 235	81.76 306	94.09 290	94.43 303	84.15 289	92.72 132	93.77 266	79.43 224	98.20 192	90.70 136	92.18 199	97.90 137

Anonymous202405211		92.07 169	90.83 190	95.76 109	98.19 79	88.75 196	97.58 90	95.00 281	86.00 267	93.64 103	97.45 89	66.24 326	99.53 72	90.68 137	92.71 190	99.01 69

tpmrst		91.44 203	91.32 168	91.79 284	95.15 218	79.20 327	93.42 303	95.37 262	88.55 203	93.49 109	93.67 270	82.49 168	98.27 189	90.41 138	89.34 238	97.90 137

UA-Net		95.95 55	95.53 52	97.20 54	97.67 110	92.98 60	97.65 74	98.13 42	94.81 22	96.61 36	98.35 25	88.87 67	99.51 77	90.36 139	97.35 107	99.11 61

IS-MVSNet		94.90 78	94.52 77	96.05 99	97.67 110	90.56 130	98.44 15	96.22 225	93.21 60	93.99 98	97.74 67	85.55 111	98.45 171	89.98 140	97.86 91	99.14 57

EI-MVSNet		93.03 130	92.88 115	93.48 231	95.77 190	86.98 253	96.44 197	97.12 163	90.66 140	91.30 161	97.64 76	86.56 99	98.05 218	89.91 141	90.55 226	95.41 233

IterMVS-LS		92.29 161	91.94 143	93.34 238	96.25 170	86.97 254	96.57 195	97.05 173	90.67 138	89.50 220	94.80 214	86.59 98	97.64 269	89.91 141	86.11 264	95.40 237
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
CDS-MVSNet		94.14 95	93.54 97	95.93 103	96.18 174	91.46 101	96.33 212	97.04 176	88.97 187	93.56 105	96.51 132	87.55 87	97.89 249	89.80 143	95.95 135	98.44 116
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
WR-MVS		92.34 157	91.53 161	94.77 164	95.13 220	90.83 124	96.40 205	97.98 78	91.88 107	89.29 226	95.54 181	82.50 167	97.80 256	89.79 144	85.27 273	95.69 223

NR-MVSNet		92.34 157	91.27 171	95.53 122	94.95 228	93.05 57	97.39 108	98.07 56	92.65 82	84.46 290	95.71 171	85.00 117	97.77 260	89.71 145	83.52 303	95.78 216

Anonymous20231211		90.63 236	89.42 244	94.27 185	98.24 72	89.19 190	98.05 35	97.89 83	79.95 322	88.25 246	94.96 200	72.56 296	98.13 198	89.70 146	85.14 275	95.49 226

testdata				95.46 128	98.18 81	88.90 195		97.66 106	82.73 303	97.03 30	98.07 45	90.06 58	98.85 139	89.67 147	98.98 66	98.64 97

Baseline_NR-MVSNet		91.20 214	90.62 204	92.95 250	93.83 283	88.03 227	97.01 142	95.12 276	88.42 212	89.70 211	95.13 198	83.47 132	97.44 281	89.66 148	83.24 305	93.37 309

PatchFormer-LS_test		91.68 191	91.18 176	93.19 245	95.24 213	83.63 293	95.53 255	95.44 259	89.82 160	91.37 156	92.58 292	80.85 203	98.52 165	89.65 149	90.16 231	97.42 159

XXY-MVS		92.16 166	91.23 173	94.95 153	94.75 239	90.94 120	97.47 102	97.43 137	89.14 180	88.90 231	96.43 136	79.71 220	98.24 190	89.56 150	87.68 252	95.67 224

XVG-ACMP-BASELINE		90.93 223	90.21 220	93.09 246	94.31 254	85.89 266	95.33 262	97.26 152	91.06 130	89.38 222	95.44 188	68.61 315	98.60 159	89.46 151	91.05 219	94.79 278

AdaColmap		94.34 89	93.68 93	96.31 88	98.59 49	91.68 94	96.59 192	97.81 92	89.87 156	92.15 142	97.06 104	83.62 131	99.54 69	89.34 152	98.07 87	97.70 147

TranMVSNet+NR-MVSNet		92.50 148	91.63 155	95.14 141	94.76 238	92.07 82	97.53 94	98.11 46	92.90 77	89.56 217	96.12 148	83.16 136	97.60 272	89.30 153	83.20 306	95.75 220

1314		92.81 140	92.03 139	95.14 141	95.33 207	89.52 166	96.04 231	97.44 135	87.72 232	86.25 277	95.33 190	83.84 128	98.79 143	89.26 154	97.05 113	97.11 162

v2v482		91.59 195	90.85 187	93.80 208	93.87 282	88.17 216	96.94 151	96.88 196	89.54 163	89.53 218	94.90 204	81.70 187	98.02 227	89.25 155	85.04 282	95.20 251

114514_t		93.95 102	93.06 111	96.63 66	99.07 28	91.61 95	97.46 103	97.96 80	77.99 332	93.00 125	97.57 82	86.14 106	99.33 95	89.22 156	99.15 55	98.94 76

PAPM_NR		95.01 72	94.59 73	96.26 93	98.89 34	90.68 128	97.24 121	97.73 96	91.80 108	92.93 130	96.62 128	89.13 65	99.14 110	89.21 157	97.78 94	98.97 72

IB-MVS	87.33 17	89.91 250	88.28 261	94.79 163	95.26 212	87.70 239	95.12 271	93.95 317	89.35 169	87.03 269	92.49 293	70.74 306	99.19 103	89.18 158	81.37 315	97.49 157
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
HY-MVS	89.66 9	93.87 104	92.95 113	96.63 66	97.10 133	92.49 72	95.64 251	96.64 210	89.05 182	93.00 125	95.79 166	85.77 110	99.45 84	89.16 159	94.35 157	97.96 134

v6		91.69 186	91.00 180	93.75 213	94.14 262	88.12 221	97.20 127	96.98 182	89.19 173	89.90 199	94.42 232	83.04 147	98.07 210	89.07 160	85.10 277	95.07 256

v1neww		91.70 184	91.01 178	93.75 213	94.19 257	88.14 219	97.20 127	96.98 182	89.18 175	89.87 202	94.44 230	83.10 141	98.06 215	89.06 161	85.09 278	95.06 259

v7new		91.70 184	91.01 178	93.75 213	94.19 257	88.14 219	97.20 127	96.98 182	89.18 175	89.87 202	94.44 230	83.10 141	98.06 215	89.06 161	85.09 278	95.06 259

V42		91.58 196	90.87 185	93.73 216	94.05 274	88.50 202	97.32 116	96.97 185	88.80 197	89.71 210	94.33 237	82.54 166	98.05 218	89.01 163	85.07 280	94.64 284

OurMVSNet-221017-0		90.51 239	90.19 221	91.44 293	93.41 295	81.25 309	96.98 144	96.28 220	91.68 111	86.55 275	96.30 140	74.20 287	97.98 232	88.96 164	87.40 257	95.09 253

API-MVS		94.84 81	94.49 78	95.90 104	97.90 99	92.00 86	97.80 56	97.48 123	89.19 173	94.81 85	96.71 114	88.84 68	99.17 106	88.91 165	98.76 72	96.53 184

divwei89l23v2f112		91.61 192	90.89 182	93.78 210	94.01 275	88.22 211	96.96 145	96.96 186	89.17 177	89.75 208	94.28 248	83.02 149	98.03 224	88.86 166	84.98 286	95.08 254

v1141		91.61 192	90.89 182	93.78 210	94.01 275	88.24 209	96.96 145	96.96 186	89.17 177	89.75 208	94.29 246	82.99 151	98.03 224	88.85 167	85.00 283	95.07 256

v1		91.61 192	90.89 182	93.78 210	94.01 275	88.21 213	96.96 145	96.96 186	89.17 177	89.78 207	94.29 246	82.97 153	98.05 218	88.85 167	84.99 284	95.08 254

test-LLR		91.42 204	91.19 175	92.12 274	94.59 244	80.66 312	94.29 284	92.98 331	91.11 128	90.76 176	92.37 295	79.02 231	98.07 210	88.81 169	96.74 120	97.63 148

test-mter		90.19 246	89.54 242	92.12 274	94.59 244	80.66 312	94.29 284	92.98 331	87.68 233	90.76 176	92.37 295	67.67 319	98.07 210	88.81 169	96.74 120	97.63 148

TAMVS		94.01 101	93.46 101	95.64 117	96.16 176	90.45 134	96.71 176	96.89 195	89.27 171	93.46 110	96.92 108	87.29 92	97.94 241	88.70 171	95.74 139	98.53 102

Patchmatch-RL test		87.38 288	86.24 287	90.81 302	88.74 336	78.40 330	88.12 344	93.17 324	87.11 245	82.17 303	89.29 325	81.95 181	95.60 326	88.64 172	77.02 325	98.41 118

TESTMET0.1,1		90.06 248	89.42 244	91.97 278	94.41 251	80.62 314	94.29 284	91.97 340	87.28 242	90.44 181	92.47 294	68.79 314	97.67 266	88.50 173	96.60 126	97.61 152

Vis-MVSNet (Re-imp)		94.15 93	93.88 87	94.95 153	97.61 114	87.92 233	98.10 31	95.80 247	92.22 89	93.02 124	97.45 89	84.53 123	97.91 248	88.24 174	97.97 89	99.02 65

DWT-MVSNet_test		90.76 227	89.89 230	93.38 236	95.04 224	83.70 291	95.85 241	94.30 309	88.19 220	90.46 180	92.80 287	73.61 292	98.50 167	88.16 175	90.58 225	97.95 135

1112_ss		93.37 119	92.42 133	96.21 95	97.05 137	90.99 117	96.31 214	96.72 202	86.87 256	89.83 204	96.69 118	86.51 100	99.14 110	88.12 176	93.67 174	98.50 107

CVMVSNet		91.23 213	91.75 147	89.67 315	95.77 190	74.69 335	96.44 197	94.88 288	85.81 268	92.18 141	97.64 76	79.07 228	95.58 327	88.06 177	95.86 138	98.74 89

v7		91.47 202	90.73 194	93.68 221	94.13 263	88.16 217	97.09 136	97.05 173	88.38 213	89.80 205	94.52 223	82.21 175	98.01 228	88.00 178	85.42 269	94.87 268

MAR-MVS		94.22 91	93.46 101	96.51 74	98.00 86	92.19 80	97.67 71	97.47 126	88.13 224	93.00 125	95.84 160	84.86 119	99.51 77	87.99 179	98.17 85	97.83 142
Zhenyu Xu, Yiguang Liu, Xuelei Shi, Ying Wang, Yunan Zheng: MARMVS: Matching Ambiguity Reduced Multiple View Stereo for Efficient Large Scale Scene Reconstruction. CVPR 2020
原ACMM1				96.38 84	98.59 49	91.09 116		97.89 83	87.41 238	95.22 80	97.68 70	90.25 55	99.54 69	87.95 180	99.12 60	98.49 109

CP-MVSNet		91.89 176	91.24 172	93.82 207	95.05 223	88.57 200	97.82 55	98.19 33	91.70 110	88.21 247	95.76 168	81.96 180	97.52 276	87.86 181	84.65 289	95.37 240

v148		90.99 221	90.38 211	92.81 254	93.83 283	85.80 267	96.78 168	96.68 207	89.45 167	88.75 235	93.93 261	82.96 155	97.82 255	87.83 182	83.25 304	94.80 276

v1144		91.37 207	90.60 205	93.68 221	93.89 281	88.23 210	96.84 158	97.03 179	88.37 214	89.69 212	94.39 233	82.04 178	97.98 232	87.80 183	85.37 271	94.84 270

gm-plane-assit					93.22 303	78.89 329			84.82 282		93.52 275		98.64 155	87.72 184

pmmvs4		90.93 223	89.85 232	94.17 188	93.34 297	90.79 126	94.60 276	96.02 231	84.62 284	87.45 258	95.15 196	81.88 183	97.45 280	87.70 185	87.87 251	94.27 297

Test_1112_low_res		92.84 139	91.84 145	95.85 106	97.04 138	89.97 143	95.53 255	96.64 210	85.38 272	89.65 214	95.18 195	85.86 108	99.10 118	87.70 185	93.58 179	98.49 109

无先验							95.79 244	97.87 87	83.87 294				99.65 42	87.68 187		98.89 82

1121		94.71 84	93.83 88	97.34 43	98.57 52	93.64 43	96.04 231	97.73 96	81.56 314	95.68 70	97.85 59	90.23 56	99.65 42	87.68 187	99.12 60	98.73 90

Fast-Effi-MVS+		93.46 117	92.75 119	95.59 119	96.77 147	90.03 137	96.81 163	97.13 162	88.19 220	91.30 161	94.27 250	86.21 103	98.63 156	87.66 189	96.46 131	98.12 129

CNLPA		94.28 90	93.53 98	96.52 71	98.38 61	92.55 70	96.59 192	96.88 196	90.13 153	91.91 146	97.24 96	85.21 114	99.09 121	87.64 190	97.83 92	97.92 136

v8		91.29 212	90.53 207	93.57 228	94.15 261	88.12 221	97.34 113	97.06 172	88.99 184	88.32 242	94.26 252	83.08 143	98.01 228	87.62 191	83.92 298	94.57 286

pmmvs5		89.86 253	88.87 253	92.82 251	92.86 310	86.23 264	96.26 219	95.39 260	84.24 288	87.12 266	94.51 224	74.27 286	97.36 288	87.61 192	87.57 253	94.86 269

Fast-Effi-MVS+-dtu		92.29 161	91.99 141	93.21 244	95.27 209	85.52 272	97.03 138	96.63 212	92.09 97	89.11 230	95.14 197	80.33 211	98.08 206	87.54 193	94.74 155	96.03 206

OpenMVS	89.19 12	92.86 137	91.68 150	96.40 81	95.34 204	92.73 65	98.27 23	98.12 43	84.86 281	85.78 280	97.75 66	78.89 244	99.74 25	87.50 194	98.65 74	96.73 177

v52		90.70 233	90.00 226	92.82 251	93.24 301	87.03 251	97.60 87	97.14 161	88.21 218	87.69 254	93.94 260	80.91 199	98.07 210	87.39 195	83.87 300	93.36 310

V4		90.71 232	90.00 226	92.82 251	93.21 304	87.03 251	97.59 89	97.16 160	88.21 218	87.69 254	93.92 262	80.93 198	98.06 215	87.39 195	83.90 299	93.39 308

semantic-postprocess				91.82 282	95.52 196	84.20 286		96.15 228	90.61 144	87.39 261	94.27 250	75.63 276	96.44 306	87.34 197	86.88 260	94.82 274

PLC	91.00 6	94.11 96	93.43 103	96.13 97	98.58 51	91.15 115	96.69 181	97.39 140	87.29 241	91.37 156	96.71 114	88.39 75	99.52 76	87.33 198	97.13 112	97.73 145
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
tpm		90.25 243	89.74 238	91.76 287	93.92 279	79.73 323	93.98 291	93.54 322	88.28 216	91.99 145	93.25 283	77.51 267	97.44 281	87.30 199	87.94 250	98.12 129

GA-MVS		91.38 206	90.31 212	94.59 171	94.65 242	87.62 240	94.34 282	96.19 226	90.73 136	90.35 183	93.83 263	71.84 298	97.96 239	87.22 200	93.61 177	98.21 126

BH-untuned		92.94 133	92.62 124	93.92 205	97.22 128	86.16 265	96.40 205	96.25 223	90.06 154	89.79 206	96.17 147	83.19 135	98.35 183	87.19 201	97.27 109	97.24 161

v144192		91.06 219	90.28 214	93.39 235	93.66 288	87.23 247	96.83 159	97.07 170	87.43 237	89.69 212	94.28 248	81.48 188	98.00 231	87.18 202	84.92 287	94.93 266

RPSCF		90.75 229	90.86 186	90.42 309	96.84 143	76.29 333	95.61 252	96.34 218	83.89 292	91.38 155	97.87 56	76.45 270	98.78 144	87.16 203	92.23 196	96.20 192

PS-CasMVS		91.55 198	90.84 189	93.69 220	94.96 227	88.28 206	97.84 54	98.24 28	91.46 116	88.04 249	95.80 163	79.67 221	97.48 278	87.02 204	84.54 291	95.31 243

pm-mvs1		90.72 231	89.65 241	93.96 200	94.29 255	89.63 159	97.79 57	96.82 199	89.07 181	86.12 279	95.48 187	78.61 246	97.78 258	86.97 205	81.67 313	94.46 289

IterMVS		90.15 247	89.67 239	91.61 289	95.48 198	83.72 289	94.33 283	96.12 229	89.99 155	87.31 264	94.15 255	75.78 275	96.27 309	86.97 205	86.89 259	94.83 272
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
F-COLMAP		93.58 114	92.98 112	95.37 131	98.40 58	88.98 193	97.18 130	97.29 151	87.75 231	90.49 179	97.10 103	85.21 114	99.50 79	86.70 207	96.72 122	97.63 148

PVSNet	86.66 18	92.24 163	91.74 149	93.73 216	97.77 105	83.69 292	92.88 313	96.72 202	87.91 227	93.00 125	94.86 208	78.51 247	99.05 127	86.53 208	97.45 104	98.47 112

v1192		91.07 218	90.23 218	93.58 227	93.70 286	87.82 236	96.73 171	97.07 170	87.77 230	89.58 215	94.32 238	80.90 202	97.97 235	86.52 209	85.48 267	94.95 262

新几何1				97.32 44	98.60 48	93.59 44		97.75 94	81.58 312	95.75 68	97.85 59	90.04 59	99.67 40	86.50 210	99.13 57	98.69 95

v10		91.04 220	90.23 218	93.49 230	94.12 265	88.16 217	97.32 116	97.08 169	88.26 217	88.29 244	94.22 253	82.17 177	97.97 235	86.45 211	84.12 294	94.33 293

v1921920		90.85 225	90.03 225	93.29 240	93.55 289	86.96 255	96.74 170	97.04 176	87.36 239	89.52 219	94.34 236	80.23 213	97.97 235	86.27 212	85.21 274	94.94 264

MDTV_nov1_ep13_2view						70.35 342	93.10 311		83.88 293	93.55 106		82.47 170		86.25 213		98.38 122

test_post1							92.81 315				16.58 363	80.53 206	97.68 265	86.20 214

PAPR		94.18 92	93.42 105	96.48 76	97.64 112	91.42 103	95.55 253	97.71 103	88.99 184	92.34 138	95.82 162	89.19 63	99.11 112	86.14 215	97.38 105	98.90 80

GBi-Net		91.35 208	90.27 215	94.59 171	96.51 158	91.18 112	97.50 96	96.93 191	88.82 194	89.35 223	94.51 224	73.87 288	97.29 291	86.12 216	88.82 241	95.31 243

test1		91.35 208	90.27 215	94.59 171	96.51 158	91.18 112	97.50 96	96.93 191	88.82 194	89.35 223	94.51 224	73.87 288	97.29 291	86.12 216	88.82 241	95.31 243

FMVSNet3		91.78 178	90.69 197	95.03 147	96.53 157	92.27 76	97.02 140	96.93 191	89.79 162	89.35 223	94.65 220	77.01 268	97.47 279	86.12 216	88.82 241	95.35 241

EPMVS		90.70 233	89.81 234	93.37 237	94.73 240	84.21 285	93.67 298	88.02 350	89.50 165	92.38 136	93.49 277	77.82 265	97.78 258	86.03 219	92.68 191	98.11 132

MVS		91.71 181	90.44 208	95.51 123	95.20 216	91.59 97	96.04 231	97.45 132	73.44 344	87.36 262	95.60 177	85.42 112	99.10 118	85.97 220	97.46 100	95.83 213

testdata2													99.67 40	85.96 221

K. test v3		87.64 287	86.75 285	90.32 310	93.02 309	79.48 325	96.61 189	92.08 339	90.66 140	80.25 327	94.09 256	67.21 323	96.65 305	85.96 221	80.83 318	94.83 272

WR-MVS_H		92.00 172	91.35 166	93.95 201	95.09 222	89.47 167	98.04 36	98.68 7	91.46 116	88.34 241	94.68 218	85.86 108	97.56 273	85.77 223	84.24 293	94.82 274

gg-mvs-nofinetune		87.82 285	85.61 292	94.44 178	94.46 248	89.27 187	91.21 329	84.61 356	80.88 317	89.89 201	74.98 348	71.50 300	97.53 275	85.75 224	97.21 110	96.51 185

v748		90.34 241	89.54 242	92.75 256	93.25 300	85.71 269	97.61 86	97.17 157	88.54 204	87.20 265	93.54 274	81.02 194	98.01 228	85.73 225	81.80 311	94.52 287

tpm2		89.96 249	89.21 248	92.23 268	94.91 233	81.25 309	93.78 294	94.42 304	80.62 320	91.56 152	93.44 280	76.44 271	97.94 241	85.60 226	92.08 203	97.49 157

v1240		90.70 233	89.85 232	93.23 242	93.51 292	86.80 256	96.61 189	97.02 180	87.16 244	89.58 215	94.31 239	79.55 223	97.98 232	85.52 227	85.44 268	94.90 267

PEN-MVS		91.20 214	90.44 208	93.48 231	94.49 247	87.91 235	97.76 58	98.18 35	91.29 121	87.78 252	95.74 170	80.35 210	97.33 289	85.46 228	82.96 307	95.19 252

QAPM		93.45 118	92.27 135	96.98 60	96.77 147	92.62 68	98.39 18	98.12 43	84.50 286	88.27 245	97.77 65	82.39 172	99.81 20	85.40 229	98.81 70	98.51 105

Anonymous20240521		91.32 210	90.43 210	93.98 196	94.93 230	89.28 185	98.04 36	97.53 118	89.49 166	86.68 274	94.82 212	81.72 186	98.05 218	85.31 230	85.39 270	94.61 285

EU-MVSNet		88.72 267	88.90 252	88.20 318	93.15 307	74.21 336	96.63 188	94.22 312	85.18 275	87.32 263	95.97 152	76.16 272	94.98 331	85.27 231	86.17 262	95.41 233

BH-w/o		92.14 168	91.75 147	93.31 239	96.99 139	85.73 268	95.67 248	95.69 249	88.73 199	89.26 228	94.82 212	82.97 153	98.07 210	85.26 232	96.32 132	96.13 198

FMVSNet2		91.31 211	90.08 222	94.99 148	96.51 158	92.21 77	97.41 104	96.95 189	88.82 194	88.62 236	94.75 216	73.87 288	97.42 283	85.20 233	88.55 247	95.35 241

PM-MVS		83.48 308	81.86 311	88.31 317	87.83 339	77.59 331	93.43 302	91.75 341	86.91 253	80.63 317	89.91 313	44.42 353	95.84 322	85.17 234	76.73 327	91.50 337

LF4IMVS		87.94 284	87.25 276	89.98 313	92.38 320	80.05 322	94.38 281	95.25 270	87.59 235	84.34 291	94.74 217	64.31 331	97.66 268	84.83 235	87.45 254	92.23 331

PatchmatchNet		91.91 175	91.35 166	93.59 225	95.38 202	84.11 287	93.15 309	95.39 260	89.54 163	92.10 143	93.68 269	82.82 160	98.13 198	84.81 236	95.32 144	98.52 103
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
pmmvs6		87.81 286	86.19 288	92.69 258	91.32 325	86.30 263	97.34 113	96.41 216	80.59 321	84.05 297	94.37 235	67.37 322	97.67 266	84.75 237	79.51 321	94.09 300

v18		88.71 268	87.52 267	92.27 264	94.16 260	88.11 223	96.82 162	95.96 233	87.03 246	80.76 314	89.81 315	83.15 137	96.22 310	84.69 238	75.31 332	92.49 320

v7n		90.76 227	89.86 231	93.45 234	93.54 290	87.60 241	97.70 70	97.37 143	88.85 191	87.65 256	94.08 257	81.08 193	98.10 203	84.68 239	83.79 301	94.66 283

SixPastTwentyTwo		89.15 261	88.54 258	90.98 298	93.49 293	80.28 319	96.70 179	94.70 292	90.78 134	84.15 295	95.57 178	71.78 299	97.71 264	84.63 240	85.07 280	94.94 264

v17		88.67 270	87.47 270	92.26 266	94.13 263	88.09 225	96.81 163	95.95 234	87.02 247	80.72 315	89.75 317	83.11 140	96.20 311	84.61 241	75.15 334	92.49 320

v16		88.69 269	87.50 268	92.26 266	94.19 257	88.11 223	96.81 163	95.95 234	87.01 248	80.71 316	89.80 316	83.08 143	96.20 311	84.61 241	75.34 331	92.48 322

TDRefinement		86.53 294	84.76 299	91.85 281	82.23 349	84.25 284	96.38 207	95.35 263	84.97 280	84.09 296	94.94 201	65.76 329	98.34 185	84.60 243	74.52 339	92.97 311

ACMH	87.59 16	90.53 238	89.42 244	93.87 206	96.21 171	87.92 233	97.24 121	96.94 190	88.45 205	83.91 298	96.27 142	71.92 297	98.62 158	84.43 244	89.43 237	95.05 261
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
ACMH+	87.92 14	90.20 245	89.18 249	93.25 241	96.48 161	86.45 262	96.99 143	96.68 207	88.83 193	84.79 289	96.22 143	70.16 310	98.53 164	84.42 245	88.04 249	94.77 280

v15		88.53 272	87.31 272	92.20 269	94.09 269	88.05 226	96.72 174	95.90 238	87.01 248	80.53 319	89.60 321	83.02 149	96.13 313	84.29 246	74.64 335	92.41 326

V14		88.52 273	87.30 273	92.17 271	94.12 265	87.99 228	96.72 174	95.91 237	86.98 250	80.50 320	89.63 318	83.03 148	96.12 315	84.23 247	74.60 337	92.40 327

V9		88.49 276	87.26 275	92.18 270	94.12 265	87.97 231	96.73 171	95.90 238	86.95 252	80.40 322	89.61 319	82.98 152	96.13 313	84.14 248	74.55 338	92.44 324

MS-PatchMatch		90.27 242	89.77 235	91.78 285	94.33 253	84.72 282	95.55 253	96.73 201	86.17 265	86.36 276	95.28 193	71.28 302	97.80 256	84.09 249	98.14 86	92.81 315

v12		88.46 277	87.23 278	92.17 271	94.10 268	87.99 228	96.71 176	95.90 238	86.91 253	80.34 324	89.58 322	82.92 156	96.11 317	84.09 249	74.50 340	92.42 325

v13		88.45 278	87.22 279	92.16 273	94.08 271	87.95 232	96.71 176	95.90 238	86.86 257	80.27 326	89.55 323	82.92 156	96.12 315	84.02 251	74.63 336	92.40 327

PatchMatch-RL		92.90 135	92.02 140	95.56 120	98.19 79	90.80 125	95.27 267	97.18 155	87.96 225	91.86 148	95.68 174	80.44 208	98.99 129	84.01 252	97.54 99	96.89 173

lessismore_v0				90.45 308	91.96 323	79.09 328		87.19 353		80.32 325	94.39 233	66.31 325	97.55 274	84.00 253	76.84 326	94.70 281

CMPMVS	62.92 21	85.62 302	84.92 297	87.74 320	89.14 335	73.12 338	94.17 287	96.80 200	73.98 342	73.65 337	94.93 202	66.36 324	97.61 271	83.95 254	91.28 215	92.48 322
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
MVP-Stereo		90.74 230	90.08 222	92.71 257	93.19 306	88.20 214	95.86 240	96.27 221	86.07 266	84.86 288	94.76 215	77.84 264	97.75 261	83.88 255	98.01 88	92.17 333
Qingsong Yan: MVP-Stereo: A Parallel Multi-View Patchmatch Stereo Method with Dilation Matching for Photogrammetric Application.
LS3D		93.57 115	92.61 125	96.47 77	97.59 116	91.61 95	97.67 71	97.72 99	85.17 276	90.29 184	98.34 28	84.60 121	99.73 26	83.85 256	98.27 82	98.06 133

v11		88.41 279	87.19 282	92.08 276	94.08 271	87.77 237	96.75 169	95.85 244	86.74 258	80.50 320	89.50 324	82.49 168	96.08 318	83.55 257	75.20 333	92.38 329

Patchmatch-test1		91.54 199	90.85 187	93.59 225	95.59 194	84.95 279	94.72 275	95.58 255	90.82 133	92.25 140	93.58 273	75.80 274	97.41 284	83.35 258	95.98 134	98.40 119

testpf		80.97 315	81.40 313	79.65 333	91.53 324	72.43 339	73.47 356	89.55 348	78.63 329	80.81 312	89.06 326	61.36 336	91.36 345	83.34 259	84.89 288	75.15 352

DTE-MVSNet		90.56 237	89.75 237	93.01 248	93.95 278	87.25 245	97.64 78	97.65 108	90.74 135	87.12 266	95.68 174	79.97 217	97.00 300	83.33 260	81.66 314	94.78 279

BH-RMVSNet		92.72 142	91.97 142	94.97 151	97.16 131	87.99 228	96.15 226	95.60 253	90.62 142	91.87 147	97.15 101	78.41 249	98.57 162	83.16 261	97.60 98	98.36 123

pmmvs-eth3d		86.22 297	84.45 300	91.53 290	88.34 337	87.25 245	94.47 280	95.01 280	83.47 298	79.51 330	89.61 319	69.75 312	95.71 324	83.13 262	76.73 327	91.64 334

FMVSNet1		89.88 252	88.31 260	94.59 171	95.41 200	91.18 112	97.50 96	96.93 191	86.62 259	87.41 260	94.51 224	65.94 328	97.29 291	83.04 263	87.43 255	95.31 243

tfpn_ndepth		91.88 177	90.96 181	94.62 170	97.73 108	89.93 146	97.75 59	92.92 333	88.93 189	91.73 149	93.80 265	78.91 237	98.49 170	83.02 264	93.86 173	95.45 231

MDTV_nov1_ep13			90.76 192		95.22 214	80.33 317	93.03 312	95.28 267	88.14 223	92.84 131	93.83 263	81.34 190	98.08 206	82.86 265	94.34 158

TR-MVS		91.48 201	90.59 206	94.16 189	96.40 164	87.33 242	95.67 248	95.34 266	87.68 233	91.46 154	95.52 182	76.77 269	98.35 183	82.85 266	93.61 177	96.79 176

JIA-IIPM		88.26 282	87.04 283	91.91 279	93.52 291	81.42 308	89.38 340	94.38 305	80.84 318	90.93 175	80.74 345	79.22 227	97.92 245	82.76 267	91.62 208	96.38 190

PVSNet_0	82.17 19	85.46 303	83.64 304	90.92 300	95.27 209	79.49 324	90.55 333	95.60 253	83.76 295	83.00 301	89.95 312	71.09 303	97.97 235	82.75 268	60.79 350	95.31 243

ambc				86.56 324	83.60 346	70.00 344	85.69 348	94.97 284		80.60 318	88.45 330	37.42 354	96.84 303	82.69 269	75.44 330	92.86 312

USDC		88.94 262	87.83 265	92.27 264	94.66 241	84.96 278	93.86 293	95.90 238	87.34 240	83.40 300	95.56 179	67.43 321	98.19 194	82.64 270	89.67 236	93.66 304

tpmp4_e23		89.58 256	88.59 256	92.54 261	95.16 217	81.53 307	94.11 289	95.09 277	81.66 310	88.60 237	93.44 280	75.11 279	98.33 186	82.45 271	91.72 206	97.75 144

tfpn1000		91.99 173	91.05 177	94.80 161	97.78 104	89.66 158	97.91 47	92.90 334	88.99 184	91.73 149	94.84 209	78.99 236	98.33 186	82.41 272	93.91 172	96.40 189

ITE_SJBPF				92.43 263	95.34 204	85.37 274		95.92 236	91.47 115	87.75 253	96.39 138	71.00 304	97.96 239	82.36 273	89.86 235	93.97 301

UnsupCasMVSNet_eth		85.99 299	84.45 300	90.62 306	89.97 331	82.40 302	93.62 300	97.37 143	89.86 157	78.59 332	92.37 295	65.25 330	95.35 330	82.27 274	70.75 344	94.10 298

GG-mvs-BLEND				93.62 223	93.69 287	89.20 188	92.39 321	83.33 357		87.98 251	89.84 314	71.00 304	96.87 302	82.08 275	95.40 143	94.80 276

view600		92.55 144	91.68 150	95.18 134	97.98 87	89.44 171	98.00 38	94.57 298	92.09 97	93.17 120	95.52 182	78.14 255	99.11 112	81.61 276	94.04 166	96.98 164

view800		92.55 144	91.68 150	95.18 134	97.98 87	89.44 171	98.00 38	94.57 298	92.09 97	93.17 120	95.52 182	78.14 255	99.11 112	81.61 276	94.04 166	96.98 164

conf0.05thres1000		92.55 144	91.68 150	95.18 134	97.98 87	89.44 171	98.00 38	94.57 298	92.09 97	93.17 120	95.52 182	78.14 255	99.11 112	81.61 276	94.04 166	96.98 164

tfpn		92.55 144	91.68 150	95.18 134	97.98 87	89.44 171	98.00 38	94.57 298	92.09 97	93.17 120	95.52 182	78.14 255	99.11 112	81.61 276	94.04 166	96.98 164

thres600view7		92.49 150	91.60 156	95.18 134	97.91 98	89.47 167	97.65 74	94.66 293	92.18 96	93.33 112	94.91 203	78.06 259	99.10 118	81.61 276	94.06 164	96.98 164

LTVRE_ROB	88.41 13	90.99 221	89.92 229	94.19 187	96.18 174	89.55 163	96.31 214	97.09 167	87.88 228	85.67 281	95.91 157	78.79 245	98.57 162	81.50 281	89.98 232	94.44 290
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
tpmvs		89.83 254	89.15 250	91.89 280	94.92 231	80.30 318	93.11 310	95.46 258	86.28 262	88.08 248	92.65 289	80.44 208	98.52 165	81.47 282	89.92 234	96.84 175

conf200view11		92.45 151	91.58 157	95.05 145	97.92 95	89.37 177	97.71 67	94.66 293	92.20 91	93.31 113	94.90 204	78.06 259	99.08 123	81.40 283	94.08 160	96.70 179

thres100view900		92.43 153	91.58 157	94.98 150	97.92 95	89.37 177	97.71 67	94.66 293	92.20 91	93.31 113	94.90 204	78.06 259	99.08 123	81.40 283	94.08 160	96.48 187

tfpn200view9		92.38 156	91.52 162	94.95 153	97.85 101	89.29 183	97.41 104	94.88 288	92.19 94	93.27 117	94.46 228	78.17 252	99.08 123	81.40 283	94.08 160	96.48 187

thres400		92.42 154	91.52 162	95.12 143	97.85 101	89.29 183	97.41 104	94.88 288	92.19 94	93.27 117	94.46 228	78.17 252	99.08 123	81.40 283	94.08 160	96.98 164

tfpn111		92.45 151	91.58 157	95.06 144	97.92 95	89.37 177	97.71 67	94.66 293	92.20 91	93.31 113	94.90 204	78.06 259	99.11 112	81.37 287	94.06 164	96.70 179

DP-MVS		92.76 141	91.51 164	96.52 71	98.77 36	90.99 117	97.38 110	96.08 230	82.38 305	89.29 226	97.87 56	83.77 129	99.69 36	81.37 287	96.69 123	98.89 82

conf0.01		91.74 179	90.67 198	94.94 156	97.55 119	89.68 152	97.64 78	93.14 325	88.43 206	91.24 166	94.30 240	78.91 237	98.45 171	81.28 289	93.57 180	96.70 179

conf0.002		91.74 179	90.67 198	94.94 156	97.55 119	89.68 152	97.64 78	93.14 325	88.43 206	91.24 166	94.30 240	78.91 237	98.45 171	81.28 289	93.57 180	96.70 179

thresconf0.02		91.69 186	90.67 198	94.75 165	97.55 119	89.68 152	97.64 78	93.14 325	88.43 206	91.24 166	94.30 240	78.91 237	98.45 171	81.28 289	93.57 180	96.11 199

tfpn_n400		91.69 186	90.67 198	94.75 165	97.55 119	89.68 152	97.64 78	93.14 325	88.43 206	91.24 166	94.30 240	78.91 237	98.45 171	81.28 289	93.57 180	96.11 199

tfpnconf		91.69 186	90.67 198	94.75 165	97.55 119	89.68 152	97.64 78	93.14 325	88.43 206	91.24 166	94.30 240	78.91 237	98.45 171	81.28 289	93.57 180	96.11 199

tfpnview11		91.69 186	90.67 198	94.75 165	97.55 119	89.68 152	97.64 78	93.14 325	88.43 206	91.24 166	94.30 240	78.91 237	98.45 171	81.28 289	93.57 180	96.11 199

thres200		92.23 164	91.39 165	94.75 165	97.61 114	89.03 192	96.60 191	95.09 277	92.08 102	93.28 116	94.00 258	78.39 250	99.04 128	81.26 295	94.18 159	96.19 193

CR-MVSNet		90.82 226	89.77 235	93.95 201	94.45 249	87.19 248	90.23 335	95.68 251	86.89 255	92.40 134	92.36 298	80.91 199	97.05 295	81.09 296	93.95 170	97.60 153

MSDG		91.42 204	90.24 217	94.96 152	97.15 132	88.91 194	93.69 297	96.32 219	85.72 270	86.93 271	96.47 134	80.24 212	98.98 130	80.57 297	95.05 149	96.98 164

dp		88.90 264	88.26 262	90.81 302	94.58 246	76.62 332	92.85 314	94.93 286	85.12 277	90.07 197	93.07 284	75.81 273	98.12 201	80.53 298	87.42 256	97.71 146

tpm cat1		88.36 280	87.21 280	91.81 283	95.13 220	80.55 315	92.58 317	95.70 248	74.97 340	87.45 258	91.96 303	78.01 263	98.17 196	80.39 299	88.74 244	96.72 178

AllTest		90.23 244	88.98 251	93.98 196	97.94 93	86.64 258	96.51 196	95.54 256	85.38 272	85.49 283	96.77 112	70.28 308	99.15 108	80.02 300	92.87 187	96.15 196

TestCases				93.98 196	97.94 93	86.64 258		95.54 256	85.38 272	85.49 283	96.77 112	70.28 308	99.15 108	80.02 300	92.87 187	96.15 196

ADS-MVSNet2		89.45 258	88.59 256	92.03 277	95.86 185	82.26 303	90.93 330	94.32 308	83.23 300	91.28 164	91.81 305	79.01 233	95.99 319	79.52 302	91.39 213	97.84 140

ADS-MVSNet		89.89 251	88.68 255	93.53 229	95.86 185	84.89 280	90.93 330	95.07 279	83.23 300	91.28 164	91.81 305	79.01 233	97.85 251	79.52 302	91.39 213	97.84 140

our_test_3		88.78 266	87.98 264	91.20 296	92.45 318	82.53 299	93.61 301	95.69 249	85.77 269	84.88 287	93.71 267	79.99 216	96.78 304	79.47 304	86.24 261	94.28 296

EPNet_dtu		91.71 181	91.28 170	92.99 249	93.76 285	83.71 290	96.69 181	95.28 267	93.15 64	87.02 270	95.95 154	83.37 134	97.38 287	79.46 305	96.84 116	97.88 139
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
TransMVSNet (Re)		88.94 262	87.56 266	93.08 247	94.35 252	88.45 204	97.73 63	95.23 271	87.47 236	84.26 293	95.29 191	79.86 218	97.33 289	79.44 306	74.44 341	93.45 307

EG-PatchMatch MVS		87.02 292	85.44 293	91.76 287	92.67 314	85.00 277	96.08 230	96.45 215	83.41 299	79.52 329	93.49 277	57.10 342	97.72 263	79.34 307	90.87 222	92.56 318

Patchmtry		88.64 271	87.25 276	92.78 255	94.09 269	86.64 258	89.82 338	95.68 251	80.81 319	87.63 257	92.36 298	80.91 199	97.03 297	78.86 308	85.12 276	94.67 282

FMVSNet5		87.29 290	85.79 291	91.78 285	94.80 237	87.28 243	95.49 257	95.28 267	84.09 290	83.85 299	91.82 304	62.95 333	94.17 334	78.48 309	85.34 272	93.91 302

COLMAP_ROB	87.81 15	90.40 240	89.28 247	93.79 209	97.95 92	87.13 250	96.92 152	95.89 243	82.83 302	86.88 273	97.18 98	73.77 291	99.29 98	78.44 310	93.62 176	94.95 262
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
test0.0.03 1		89.37 260	88.70 254	91.41 294	92.47 317	85.63 270	95.22 269	92.70 336	91.11 128	86.91 272	93.65 271	79.02 231	93.19 339	78.00 311	89.18 239	95.41 233

MIMVSNet		88.50 275	86.76 284	93.72 218	94.84 235	87.77 237	91.39 325	94.05 314	86.41 261	87.99 250	92.59 291	63.27 332	95.82 323	77.44 312	92.84 189	97.57 155

MDA-MVSNet_test_wron		85.87 300	84.23 302	90.80 304	92.38 320	82.57 298	93.17 307	95.15 274	82.15 306	67.65 343	92.33 301	78.20 251	95.51 328	77.33 313	79.74 319	94.31 295

YYNet1		85.87 300	84.23 302	90.78 305	92.38 320	82.46 301	93.17 307	95.14 275	82.12 307	67.69 342	92.36 298	78.16 254	95.50 329	77.31 314	79.73 320	94.39 291

UnsupCasMVSNet_bld		82.13 314	79.46 316	90.14 312	88.00 338	82.47 300	90.89 332	96.62 213	78.94 328	75.61 335	84.40 343	56.63 343	96.31 308	77.30 315	66.77 349	91.63 335

PCF-MVS	89.48 11	91.56 197	89.95 228	96.36 86	96.60 151	92.52 71	92.51 318	97.26 152	79.41 325	88.90 231	96.56 130	84.04 127	99.55 67	77.01 316	97.30 108	97.01 163
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
testgi		87.97 283	87.21 280	90.24 311	92.86 310	80.76 311	96.67 183	94.97 284	91.74 109	85.52 282	95.83 161	62.66 334	94.47 333	76.25 317	88.36 248	95.48 227

TinyColmap		86.82 293	85.35 295	91.21 295	94.91 233	82.99 297	93.94 292	94.02 316	83.58 296	81.56 309	94.68 218	62.34 335	98.13 198	75.78 318	87.35 258	92.52 319

ppachtmachnet_test		88.35 281	87.29 274	91.53 290	92.45 318	83.57 294	93.75 295	95.97 232	84.28 287	85.32 286	94.18 254	79.00 235	96.93 301	75.71 319	84.99 284	94.10 298

PAPM		91.52 200	90.30 213	95.20 133	95.30 208	89.83 148	93.38 304	96.85 198	86.26 263	88.59 238	95.80 163	84.88 118	98.15 197	75.67 320	95.93 136	97.63 148

tfpnnormal		89.70 255	88.40 259	93.60 224	95.15 218	90.10 136	97.56 92	98.16 38	87.28 242	86.16 278	94.63 221	77.57 266	98.05 218	74.48 321	84.59 290	92.65 316

DSMNet-mixed		86.34 296	86.12 290	87.00 323	89.88 332	70.43 340	94.93 273	90.08 347	77.97 333	85.42 285	92.78 288	74.44 285	93.96 335	74.43 322	95.14 146	96.62 183

Patchmatch-test		89.42 259	87.99 263	93.70 219	95.27 209	85.11 275	88.98 341	94.37 306	81.11 315	87.10 268	93.69 268	82.28 173	97.50 277	74.37 323	94.76 153	98.48 111

LCM-MVSNet		72.55 321	69.39 324	82.03 329	70.81 359	65.42 350	90.12 337	94.36 307	55.02 351	65.88 346	81.72 344	24.16 363	89.96 348	74.32 324	68.10 347	90.71 340

new-patchmatchnet		83.18 309	81.87 310	87.11 322	86.88 342	75.99 334	93.70 296	95.18 273	85.02 279	77.30 334	88.40 331	65.99 327	93.88 336	74.19 325	70.18 345	91.47 338

MDA-MVSNet-bldmvs		85.00 304	82.95 306	91.17 297	93.13 308	83.33 295	94.56 278	95.00 281	84.57 285	65.13 347	92.65 289	70.45 307	95.85 321	73.57 326	77.49 324	94.33 293

pmmvs3		79.97 316	77.50 320	87.39 321	82.80 347	79.38 326	92.70 316	90.75 345	70.69 346	78.66 331	87.47 339	51.34 350	93.40 337	73.39 327	69.65 346	89.38 342

PatchT		88.87 265	87.42 271	93.22 243	94.08 271	85.10 276	89.51 339	94.64 297	81.92 308	92.36 137	88.15 334	80.05 215	97.01 299	72.43 328	93.65 175	97.54 156

Anonymous20231206		87.09 291	86.14 289	89.93 314	91.22 326	80.35 316	96.11 228	95.35 263	83.57 297	84.16 294	93.02 285	73.54 293	95.61 325	72.16 329	86.14 263	93.84 303

MVS-HIRNet		82.47 313	81.21 314	86.26 325	95.38 202	69.21 345	88.96 342	89.49 349	66.28 347	80.79 313	74.08 350	68.48 316	97.39 286	71.93 330	95.47 142	92.18 332

new_pmnet		82.89 310	81.12 315	88.18 319	89.63 333	80.18 320	91.77 324	92.57 337	76.79 336	75.56 336	88.23 333	61.22 337	94.48 332	71.43 331	82.92 308	89.87 341

TAPA-MVS	90.10 7	92.30 160	91.22 174	95.56 120	98.33 65	89.60 160	96.79 166	97.65 108	81.83 309	91.52 153	97.23 97	87.94 79	98.91 133	71.31 332	98.37 80	98.17 127
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
test20.03		86.14 298	85.40 294	88.35 316	90.12 329	80.06 321	95.90 239	95.20 272	88.59 200	81.29 310	93.62 272	71.43 301	92.65 340	71.26 333	81.17 316	92.34 330

tmp_tt		51.94 335	53.82 332	46.29 348	33.73 364	45.30 364	78.32 355	67.24 364	18.02 359	50.93 353	87.05 340	52.99 349	53.11 362	70.76 334	25.29 359	40.46 359

MIMVSNet1		84.93 305	83.05 305	90.56 307	89.56 334	84.84 281	95.40 260	95.35 263	83.91 291	80.38 323	92.21 302	57.23 341	93.34 338	70.69 335	82.75 310	93.50 305

RPMNet		88.52 273	86.72 286	93.95 201	94.45 249	87.19 248	90.23 335	94.99 283	77.87 334	92.40 134	87.55 338	80.17 214	97.05 295	68.84 336	93.95 170	97.60 153

N_pmnet		78.73 318	78.71 317	78.79 335	92.80 312	46.50 362	94.14 288	43.71 365	78.61 330	80.83 311	91.66 308	74.94 283	96.36 307	67.24 337	84.45 292	93.50 305

OpenMVS_ROB	81.14 20	84.42 306	82.28 307	90.83 301	90.06 330	84.05 288	95.73 247	94.04 315	73.89 343	80.17 328	91.53 309	59.15 339	97.64 269	66.92 338	89.05 240	90.80 339

testus		82.63 312	82.15 308	84.07 327	87.31 341	67.67 346	93.18 305	94.29 310	82.47 304	82.14 304	90.69 310	53.01 348	91.94 343	66.30 339	89.96 233	92.62 317

test2356		82.77 311	82.14 309	84.65 326	85.77 343	70.36 341	91.22 328	93.69 321	81.58 312	81.82 306	89.00 327	60.63 338	90.77 346	64.74 340	90.80 223	92.82 313

PMMVS2		70.19 324	66.92 326	80.01 332	76.35 351	65.67 349	86.22 347	87.58 352	64.83 349	62.38 348	80.29 347	26.78 361	88.49 352	63.79 341	54.07 351	85.88 346

test_0402		86.46 295	84.79 298	91.45 292	95.02 225	85.55 271	96.29 216	94.89 287	80.90 316	82.21 302	93.97 259	68.21 318	97.29 291	62.98 342	88.68 246	91.51 336

DeepMVS_CX				74.68 340	90.84 327	64.34 351		81.61 360	65.34 348	67.47 345	88.01 335	48.60 351	80.13 357	62.33 343	73.68 343	79.58 350

test1235678		79.82 317	78.53 318	83.69 328	82.55 348	67.55 347	92.50 319	94.13 313	79.28 326	72.10 340	86.45 341	57.27 340	90.68 347	61.60 344	80.90 317	92.82 313

no-one		68.12 325	63.78 328	81.13 330	74.01 354	70.22 343	87.61 346	90.71 346	72.63 345	53.13 352	71.89 351	30.29 357	91.45 344	61.53 345	32.21 355	81.72 349

LP		84.13 307	81.85 312	90.97 299	93.20 305	82.12 304	87.68 345	94.27 311	76.80 335	81.93 305	88.52 329	72.97 295	95.95 320	59.53 346	81.73 312	94.84 270

test12356		74.97 320	74.13 321	77.49 336	78.81 350	56.23 358	88.53 343	92.75 335	75.14 337	67.50 344	85.07 342	44.88 352	89.96 348	58.71 347	75.75 329	86.26 344

1111		78.29 319	77.55 319	80.50 331	83.89 344	59.98 354	91.89 322	93.71 318	75.06 338	73.60 338	87.67 336	55.66 344	92.60 341	58.54 348	77.92 323	88.93 343

.test1245		65.38 327	69.22 325	53.86 347	83.89 344	59.98 354	91.89 322	93.71 318	75.06 338	73.60 338	87.67 336	55.66 344	92.60 341	58.54 348	2.96 361	9.00 361

wuykxyi23d		56.92 331	51.11 335	74.38 341	62.30 361	61.47 353	80.09 353	84.87 355	49.62 354	30.80 360	57.20 358	7.03 366	82.94 355	55.69 350	32.36 354	78.72 351

testmv		72.22 322	70.02 322	78.82 334	73.06 357	61.75 352	91.24 327	92.31 338	74.45 341	61.06 349	80.51 346	34.21 355	88.63 351	55.31 351	68.07 348	86.06 345

FPMVS		71.27 323	69.85 323	75.50 338	74.64 352	59.03 356	91.30 326	91.50 342	58.80 350	57.92 350	88.28 332	29.98 359	85.53 354	53.43 352	82.84 309	81.95 348

ANet_high		63.94 328	59.58 329	77.02 337	61.24 362	66.06 348	85.66 349	87.93 351	78.53 331	42.94 354	71.04 352	25.42 362	80.71 356	52.60 353	30.83 357	84.28 347

PNet_i23d		59.01 329	55.87 330	68.44 342	73.98 355	51.37 359	81.36 352	82.41 358	52.37 353	42.49 356	70.39 353	11.39 364	79.99 358	49.77 354	38.71 353	73.97 353

Gipuma		67.86 326	65.41 327	75.18 339	92.66 315	73.45 337	66.50 358	94.52 302	53.33 352	57.80 351	66.07 354	30.81 356	89.20 350	48.15 355	78.88 322	62.90 356
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
PMVS	53.92 22	58.58 330	55.40 331	68.12 343	51.00 363	48.64 360	78.86 354	87.10 354	46.77 355	35.84 359	74.28 349	8.76 365	86.34 353	42.07 356	73.91 342	69.38 354
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
MVE	50.73 23	53.25 333	48.81 336	66.58 344	65.34 360	57.50 357	72.49 357	70.94 363	40.15 358	39.28 358	63.51 355	6.89 368	73.48 361	38.29 357	42.38 352	68.76 355
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
E-PMN		53.28 332	52.56 333	55.43 345	74.43 353	47.13 361	83.63 351	76.30 361	42.23 356	42.59 355	62.22 356	28.57 360	74.40 359	31.53 358	31.51 356	44.78 357

EMVS		52.08 334	51.31 334	54.39 346	72.62 358	45.39 363	83.84 350	75.51 362	41.13 357	40.77 357	59.65 357	30.08 358	73.60 360	28.31 359	29.90 358	44.18 358

wuyk23d		25.11 337	24.57 339	26.74 350	73.98 355	39.89 365	57.88 359	9.80 366	12.27 360	10.39 361	6.97 364	7.03 366	36.44 363	25.43 360	17.39 360	3.89 363

testmvs		13.36 339	16.33 340	4.48 352	5.04 365	2.26 367	93.18 305	3.28 367	2.70 361	8.24 362	21.66 360	2.29 370	2.19 364	7.58 361	2.96 361	9.00 361

test123		13.04 340	15.66 341	5.18 351	4.51 366	3.45 366	92.50 319	1.81 368	2.50 362	7.58 363	20.15 361	3.67 369	2.18 365	7.13 362	1.07 363	9.90 360

cdsmvs_eth3d_5k		23.24 338	30.99 338	0.00 353	0.00 367	0.00 368	0.00 360	97.63 110	0.00 363	0.00 364	96.88 109	84.38 124	0.00 366	0.00 363	0.00 364	0.00 364

pcd_1.5k_mvsjas		7.39 342	9.85 343	0.00 353	0.00 367	0.00 368	0.00 360	0.00 369	0.00 363	0.00 364	0.00 365	88.65 71	0.00 366	0.00 363	0.00 364	0.00 364

pcd1.5k->3k		38.37 336	40.51 337	31.96 349	94.29 255	0.00 368	0.00 360	97.69 104	0.00 363	0.00 364	0.00 365	81.45 189	0.00 366	0.00 363	91.11 217	95.89 208

sosnet-low-res		0.00 343	0.00 344	0.00 353	0.00 367	0.00 368	0.00 360	0.00 369	0.00 363	0.00 364	0.00 365	0.00 371	0.00 366	0.00 363	0.00 364	0.00 364

sosnet		0.00 343	0.00 344	0.00 353	0.00 367	0.00 368	0.00 360	0.00 369	0.00 363	0.00 364	0.00 365	0.00 371	0.00 366	0.00 363	0.00 364	0.00 364

uncertanet		0.00 343	0.00 344	0.00 353	0.00 367	0.00 368	0.00 360	0.00 369	0.00 363	0.00 364	0.00 365	0.00 371	0.00 366	0.00 363	0.00 364	0.00 364

Regformer		0.00 343	0.00 344	0.00 353	0.00 367	0.00 368	0.00 360	0.00 369	0.00 363	0.00 364	0.00 365	0.00 371	0.00 366	0.00 363	0.00 364	0.00 364

ab-mvs-re		8.06 341	10.74 342	0.00 353	0.00 367	0.00 368	0.00 360	0.00 369	0.00 363	0.00 364	96.69 118	0.00 371	0.00 366	0.00 363	0.00 364	0.00 364

uanet		0.00 343	0.00 344	0.00 353	0.00 367	0.00 368	0.00 360	0.00 369	0.00 363	0.00 364	0.00 365	0.00 371	0.00 366	0.00 363	0.00 364	0.00 364

GSMVS																98.45 114

test_part2					99.28 17	95.74 3				98.10 7

test_part1								98.26 25				95.31 1			99.63 5	99.63 5

sam_mvs1												82.76 161				98.45 114

sam_mvs												81.94 182

MTGPA								98.08 51

test_post											17.58 362	81.76 184	98.08 206

patchmatchnet-post											90.45 311	82.65 165	98.10 203

MTMP							97.86 50	82.03 359

TEST9					98.70 39	94.19 25	96.41 201	98.02 68	88.17 222	96.03 55	97.56 84	92.74 15	99.59 53

test_8					98.67 41	94.06 31	96.37 208	98.01 70	88.58 201	95.98 60	97.55 86	92.73 16	99.58 56

agg_prior					98.67 41	93.79 38		98.00 72		95.68 70			99.57 64

test_prior4						93.66 42	96.42 200

test_prior				97.23 50	98.67 41	92.99 58		98.00 72					99.41 88			99.29 46

新几何2							95.79 244

旧先验1					98.38 61	93.38 50		97.75 94			98.09 44	92.30 28			99.01 65	99.16 54

原ACMM2							95.67 248

test222					98.24 72	92.21 77	95.33 262	97.60 111	79.22 327	95.25 79	97.84 61	88.80 69			99.15 55	98.72 91

segment_acmp												92.89 13

testdata1							95.26 268		93.10 67

test12				97.65 31	98.46 54	94.26 22		97.66 106		95.52 78		90.89 49	99.46 82		99.25 47	99.22 51

plane_prior7					96.21 171	89.98 142

plane_prior6					96.10 181	90.00 138						81.32 191

plane_prior4											96.64 121

plane_prior3						90.00 138			94.46 30	91.34 158

plane_prior2							97.74 61		94.85 17

plane_prior1					96.14 179

plane_prior						89.99 140	97.24 121		94.06 38						92.16 200

n2								0.00 369

nn								0.00 369

door-mid								91.06 344

test11								97.88 85

door								91.13 343

HQP5-MVS						89.33 180

HQP-NCC					95.86 185		96.65 184		93.55 50	90.14 186

ACMP_Plane					95.86 185		96.65 184		93.55 50	90.14 186

HQP4-MVS										90.14 186			98.50 167			95.78 216

HQP3-MVS								97.39 140							92.10 201

HQP2-MVS												80.95 196

NP-MVS					95.99 184	89.81 149					95.87 158

ACMMP++_ref															90.30 230

ACMMP++															91.02 220

Test By Simon												88.73 70

High-resolution multi-view benchmark