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.

PHI-MVS		86.43 24	86.17 26	87.24 22	90.88 59	70.96 39	92.27 16	94.07 1	72.45 113	85.22 18	91.90 49	69.47 43	96.42 18	83.28 26	95.94 3	94.35 11

APDe-MVS		89.15 1	89.63 1	87.73 15	94.49 8	71.69 33	93.83 2	93.96 2	75.70 64	91.06 1	96.03 1	76.84 2	97.03 3	89.09 1	95.65 11	94.47 9

DeepC-MVS	79.81 2	87.08 18	86.88 18	87.69 18	91.16 54	72.32 29	90.31 41	93.94 3	77.12 38	82.82 51	94.23 14	72.13 28	97.09 2	84.83 14	95.37 14	93.65 35
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
MCST-MVS		87.37 13	87.25 11	87.73 15	94.53 7	72.46 26	89.82 48	93.82 4	73.07 101	84.86 26	92.89 37	76.22 4	96.33 19	84.89 13	95.13 18	94.40 10

3Dnovator+	77.84 4	85.48 35	84.47 42	88.51 2	91.08 55	73.49 11	93.18 4	93.78 5	80.79 9	76.66 125	93.37 27	60.40 146	96.75 8	77.20 64	93.73 38	95.29 1

SteuartSystems-ACMMP		88.72 4	88.86 4	88.32 4	92.14 44	72.96 15	93.73 3	93.67 6	80.19 12	88.10 5	94.80 2	73.76 17	97.11 1	87.51 4	95.82 6	94.90 3
Skip Steuart: Steuart Systems R&D Blog.
CSCG		86.41 25	86.19 25	87.07 26	92.91 33	72.48 25	90.81 31	93.56 7	73.95 84	83.16 47	91.07 65	75.94 5	95.19 45	79.94 49	94.38 32	93.55 39

FIs		82.07 66	82.42 54	81.04 166	88.80 98	58.34 214	88.26 90	93.49 8	76.93 43	78.47 85	91.04 66	69.92 40	92.34 148	69.87 129	84.97 111	92.44 68

DELS-MVS		85.41 38	85.30 36	85.77 45	88.49 106	67.93 88	85.52 166	93.44 9	78.70 23	83.63 44	89.03 107	74.57 9	95.71 29	80.26 47	94.04 36	93.66 30
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
FC-MVSNet-test		81.52 75	82.02 62	80.03 179	88.42 110	55.97 243	87.95 97	93.42 10	77.10 39	77.38 112	90.98 71	69.96 39	91.79 155	68.46 139	84.50 116	92.33 70

DeepPCF-MVS	80.84 1	88.10 5	88.56 5	86.73 30	92.24 43	69.03 64	89.57 54	93.39 11	77.53 33	89.79 3	94.12 16	78.98 1	96.58 16	85.66 6	95.72 7	94.58 6

HPM-MVS		87.11 16	86.98 15	87.50 20	93.88 19	72.16 30	92.19 17	93.33 12	76.07 61	83.81 40	93.95 20	69.77 42	96.01 26	85.15 8	94.66 26	94.32 13

HFP-MVS		87.58 9	87.47 10	87.94 10	94.58 5	73.54 9	93.04 5	93.24 13	76.78 46	84.91 21	94.44 8	70.78 33	96.61 12	84.53 16	94.89 22	93.66 30

#test#		87.33 14	87.13 14	87.94 10	94.58 5	73.54 9	92.34 14	93.24 13	75.23 71	84.91 21	94.44 8	70.78 33	96.61 12	83.75 23	94.89 22	93.66 30

ACMMPR		87.44 10	87.23 12	88.08 7	94.64 3	73.59 6	93.04 5	93.20 15	76.78 46	84.66 27	94.52 3	68.81 48	96.65 10	84.53 16	94.90 21	94.00 23

SD-MVS		88.06 6	88.50 6	86.71 31	92.60 41	72.71 19	91.81 22	93.19 16	77.87 27	90.32 2	94.00 19	74.83 8	93.78 98	87.63 3	94.27 35	93.65 35

ACMMP		85.89 31	85.39 33	87.38 21	93.59 21	72.63 21	92.74 10	93.18 17	76.78 46	80.73 66	93.82 21	64.33 76	96.29 20	82.67 34	90.69 58	93.23 47
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
region2R		87.42 12	87.20 13	88.09 6	94.63 4	73.55 7	93.03 7	93.12 18	76.73 49	84.45 30	94.52 3	69.09 46	96.70 9	84.37 19	94.83 24	94.03 20

PGM-MVS		86.68 21	86.27 24	87.90 13	94.22 16	73.38 12	90.22 43	93.04 19	75.53 66	83.86 38	94.42 10	67.87 52	96.64 11	82.70 33	94.57 27	93.66 30

DeepC-MVS_fast	79.65 3	86.91 19	86.62 20	87.76 14	93.52 22	72.37 28	91.26 25	93.04 19	76.62 51	84.22 35	93.36 28	71.44 31	96.76 7	80.82 42	95.33 15	94.16 15
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
UniMVSNet (Re)		81.60 74	81.11 72	83.09 104	88.38 111	64.41 153	87.60 104	93.02 21	78.42 26	78.56 83	88.16 125	69.78 41	93.26 119	69.58 131	76.49 193	91.60 85

canonicalmvs		85.91 30	85.87 29	86.04 43	89.84 72	69.44 62	90.45 39	93.00 22	76.70 50	88.01 7	91.23 62	73.28 19	93.91 89	81.50 38	88.80 77	94.77 4

CNVR-MVS		88.93 3	89.13 3	88.33 3	94.77 2	73.82 4	90.51 35	93.00 22	80.90 8	88.06 6	94.06 18	76.43 3	96.84 5	88.48 2	95.99 2	94.34 12

XVS		87.18 15	86.91 17	88.00 8	94.42 10	73.33 13	92.78 8	92.99 24	79.14 16	83.67 42	94.17 15	67.45 54	96.60 14	83.06 28	94.50 28	94.07 18

X-MVStestdata		80.37 98	77.83 131	88.00 8	94.42 10	73.33 13	92.78 8	92.99 24	79.14 16	83.67 42	12.47 294	67.45 54	96.60 14	83.06 28	94.50 28	94.07 18

APD-MVS_3200maxsize		85.97 29	85.88 28	86.22 40	92.69 37	69.53 58	91.93 20	92.99 24	73.54 91	85.94 13	94.51 6	65.80 68	95.61 30	83.04 30	92.51 45	93.53 41

test_prior3		86.73 20	86.86 19	86.33 37	92.61 39	69.59 56	88.85 68	92.97 27	75.41 67	84.91 21	93.54 22	74.28 14	95.48 34	83.31 24	95.86 4	93.91 25

test_prior				86.33 37	92.61 39	69.59 56		92.97 27					95.48 34			93.91 25

MSLP-MVS		85.43 37	85.76 31	84.45 68	91.93 47	70.24 47	90.71 32	92.86 29	77.46 35	84.22 35	92.81 42	67.16 57	92.94 134	80.36 45	94.35 33	90.16 132

HPM-MVS++		89.02 2	89.15 2	88.63 1	95.01 1	76.03 1	92.38 13	92.85 30	80.26 11	87.78 8	94.27 12	75.89 6	96.81 6	87.45 5	96.44 1	93.05 53

APD-MVS		87.44 10	87.52 9	87.19 23	94.24 15	72.39 27	91.86 21	92.83 31	73.01 102	88.58 4	94.52 3	73.36 18	96.49 17	84.26 20	95.01 19	92.70 60
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
NCCC		88.06 6	88.01 7	88.24 5	94.41 12	73.62 5	91.22 28	92.83 31	81.50 5	85.79 15	93.47 26	73.02 21	97.00 4	84.90 11	94.94 20	94.10 17

CP-MVS		87.11 16	86.92 16	87.68 19	94.20 17	73.86 3	93.98 1	92.82 33	76.62 51	83.68 41	94.46 7	67.93 50	95.95 27	84.20 21	94.39 31	93.23 47

nrg030		83.88 44	83.53 44	84.96 56	86.77 145	69.28 63	90.46 38	92.67 34	74.79 75	82.95 48	91.33 61	72.70 23	93.09 129	80.79 43	79.28 162	92.50 66

WR-MVS_H		78.51 133	78.49 116	78.56 196	88.02 119	56.38 238	88.43 79	92.67 34	77.14 37	73.89 159	87.55 133	66.25 62	89.24 202	58.92 202	73.55 223	90.06 141

MP-MVS		87.71 8	87.64 8	87.93 12	94.36 14	73.88 2	92.71 12	92.65 36	77.57 30	83.84 39	94.40 11	72.24 27	96.28 21	85.65 7	95.30 17	93.62 37

alignmvs		85.48 35	85.32 35	85.96 44	89.51 76	69.47 60	89.74 51	92.47 37	76.17 59	87.73 9	91.46 60	70.32 36	93.78 98	81.51 37	88.95 74	94.63 5

原ACMM1				84.35 71	93.01 32	68.79 70		92.44 38	63.96 205	81.09 63	91.57 56	66.06 65	95.45 36	67.19 147	94.82 25	88.81 166

HQP_MVS		83.64 48	83.14 48	85.14 52	90.08 68	68.71 75	91.25 26	92.44 38	79.12 18	78.92 78	91.00 69	60.42 144	95.38 40	78.71 53	86.32 103	91.33 92

plane_prior5								92.44 38					95.38 40	78.71 53	86.32 103	91.33 92

CDPH-MVS		85.76 32	85.29 37	87.17 24	93.49 23	71.08 37	88.58 77	92.42 41	68.32 167	84.61 28	93.48 24	72.32 26	96.15 24	79.00 50	95.43 13	94.28 14

UniMVSNet_NR-MVSNet		81.88 69	81.54 67	82.92 115	88.46 108	63.46 168	87.13 125	92.37 42	80.19 12	78.38 89	89.14 104	71.66 30	93.05 130	70.05 126	76.46 194	92.25 73

CLD-MVS		82.31 63	81.65 66	84.29 73	88.47 107	67.73 92	85.81 162	92.35 43	75.78 62	78.33 91	86.58 162	64.01 79	94.35 70	76.05 73	87.48 92	90.79 104
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
test11								92.23 44

mPP-MVS		86.67 22	86.32 23	87.72 17	94.41 12	73.55 7	92.74 10	92.22 45	76.87 44	82.81 52	94.25 13	66.44 61	96.24 22	82.88 32	94.28 34	93.38 43

DP-MVS Recon		83.11 55	82.09 60	86.15 41	94.44 9	70.92 41	88.79 70	92.20 46	70.53 136	79.17 75	91.03 68	64.12 78	96.03 25	68.39 140	90.14 64	91.50 89

HQP3-MVS								92.19 47							85.99 107

HQP-MVS		82.61 61	82.02 62	84.37 69	89.33 80	66.98 101	89.17 59	92.19 47	76.41 53	77.23 117	90.23 80	60.17 147	95.11 48	77.47 61	85.99 107	91.03 97

3Dnovator	76.31 5	83.38 52	82.31 58	86.59 34	87.94 121	72.94 17	90.64 33	92.14 49	77.21 36	75.47 137	92.83 39	58.56 152	94.72 65	73.24 100	92.71 43	92.13 77

Regformer-2		86.63 23	86.53 21	86.95 27	89.33 80	71.24 36	88.43 79	92.05 50	82.50 1	86.88 11	90.09 83	74.45 10	95.61 30	84.38 18	90.63 59	94.01 22

MVS_Test		83.15 53	83.06 49	83.41 93	86.86 142	63.21 175	86.11 154	92.00 51	74.31 79	82.87 50	89.44 101	70.03 38	93.21 120	77.39 63	88.50 85	93.81 29

PVSNet_BlendedMVS		80.60 89	80.02 84	82.36 134	88.85 94	65.40 122	86.16 152	92.00 51	69.34 153	78.11 100	86.09 174	66.02 66	94.27 73	71.52 119	82.06 136	87.39 200

PVSNet_Blended		80.98 81	80.34 80	82.90 116	88.85 94	65.40 122	84.43 183	92.00 51	67.62 172	78.11 100	85.05 192	66.02 66	94.27 73	71.52 119	89.50 69	89.01 161

QAPM		80.88 82	79.50 94	85.03 54	88.01 120	68.97 68	91.59 23	92.00 51	66.63 181	75.15 149	92.16 44	57.70 157	95.45 36	63.52 169	88.76 78	90.66 112

LPG-MVS_test		82.08 65	81.27 69	84.50 66	89.23 88	68.76 72	90.22 43	91.94 55	75.37 69	76.64 126	91.51 57	54.29 179	94.91 57	78.44 55	83.78 121	89.83 147

LGP-MVS_train				84.50 66	89.23 88	68.76 72		91.94 55	75.37 69	76.64 126	91.51 57	54.29 179	94.91 57	78.44 55	83.78 121	89.83 147

TEST9					93.26 27	72.96 15	88.75 72	91.89 57	68.44 166	85.00 20	93.10 32	74.36 13	95.41 39

DU-MVS		81.12 80	80.52 79	82.90 116	87.80 125	63.46 168	87.02 130	91.87 58	79.01 21	78.38 89	89.07 105	65.02 73	93.05 130	70.05 126	76.46 194	92.20 74

test_8					93.13 28	72.57 23	88.68 74	91.84 59	68.69 163	84.87 25	93.10 32	74.43 11	95.16 46

PAPM_NR		83.02 56	82.41 55	84.82 61	92.47 42	66.37 107	87.93 99	91.80 60	73.82 86	77.32 114	90.66 74	67.90 51	94.90 59	70.37 125	89.48 70	93.19 49

test12				86.80 29	92.63 38	70.70 45		91.79 61		82.71 53		71.67 29	96.16 23		94.50 28	93.54 40

agg_prior1		86.22 27	86.09 27	86.62 33	92.85 34	71.94 31	88.59 76	91.78 62	68.96 160	84.41 31	93.18 31	74.94 7	94.93 55	84.75 15	95.33 15	93.01 56

agg_prior					92.85 34	71.94 31		91.78 62		84.41 31			94.93 55

PAPR		81.66 73	80.89 74	83.99 82	90.27 65	64.00 161	86.76 140	91.77 64	68.84 161	77.13 121	89.50 93	67.63 53	94.88 60	67.55 142	88.52 84	93.09 51

PVSNet_Blended_VisFu		82.62 60	81.83 65	84.96 56	90.80 61	69.76 54	88.74 73	91.70 65	69.39 150	78.96 77	88.46 120	65.47 69	94.87 61	74.42 89	88.57 82	90.24 130

agg_prior3		86.16 28	85.85 30	87.10 25	93.31 24	72.86 18	88.77 71	91.68 66	68.29 168	84.26 34	92.83 39	72.83 22	95.42 38	84.97 9	95.71 8	93.02 54

HPM-MVS_fast		85.35 39	84.95 40	86.57 35	93.69 20	70.58 46	92.15 18	91.62 67	73.89 85	82.67 54	94.09 17	62.60 107	95.54 33	80.93 40	92.93 40	93.57 38

ACMM	73.20 8	80.78 87	79.84 86	83.58 89	89.31 85	68.37 81	89.99 45	91.60 68	70.28 139	77.25 115	89.66 90	53.37 186	93.53 111	74.24 92	82.85 128	88.85 165
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
pcd1.5k->3k		34.07 271	35.26 271	30.50 283	86.92 141	0.00 304	0.00 295	91.58 69	0.00 299	0.00 300	0.00 298	56.23 166	0.00 297	0.00 297	82.60 132	91.49 90

VPA-MVSNet		80.60 89	80.55 78	80.76 170	88.07 117	60.80 197	86.86 134	91.58 69	75.67 65	80.24 68	89.45 100	63.34 84	90.25 190	70.51 124	79.22 163	91.23 95

PS-MVSNAJss		82.07 66	81.31 68	84.34 72	86.51 147	67.27 98	89.27 57	91.51 71	71.75 121	79.37 73	90.22 81	63.15 89	94.27 73	77.69 59	82.36 135	91.49 90

TAPA-MVS	73.13 9	79.15 124	77.94 129	82.79 125	89.59 75	62.99 181	88.16 93	91.51 71	65.77 188	77.14 120	91.09 64	60.91 137	93.21 120	50.26 242	87.05 96	92.17 76
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
ACMP	74.13 6	81.51 77	80.57 77	84.36 70	89.42 78	68.69 78	89.97 46	91.50 73	74.46 77	75.04 152	90.41 77	53.82 184	94.54 67	77.56 60	82.91 127	89.86 146
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
PCF-MVS	73.52 7	80.38 97	78.84 110	85.01 55	87.71 128	68.99 67	83.65 195	91.46 74	63.00 210	77.77 107	90.28 78	66.10 63	95.09 52	61.40 186	88.22 88	90.94 101
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
Regformer-1		86.41 25	86.33 22	86.64 32	89.33 80	70.93 40	88.43 79	91.39 75	82.14 3	86.65 12	90.09 83	74.39 12	95.01 54	83.97 22	90.63 59	93.97 24

TranMVSNet+NR-MVSNet		80.84 83	80.31 81	82.42 132	87.85 123	62.33 186	87.74 102	91.33 76	80.55 10	77.99 103	89.86 87	65.23 71	92.62 139	67.05 149	75.24 210	92.30 71

Regformer-4		85.68 34	85.45 32	86.35 36	88.95 92	69.67 55	88.29 88	91.29 77	81.73 4	85.36 17	90.01 85	72.62 24	95.35 43	83.28 26	87.57 90	94.03 20

PS-CasMVS		78.01 144	78.09 126	77.77 206	87.71 128	54.39 250	88.02 94	91.22 78	77.50 34	73.26 162	88.64 114	60.73 138	88.41 213	61.88 182	73.88 221	90.53 122

v7n		78.97 129	77.58 137	83.14 102	83.45 200	65.51 120	88.32 86	91.21 79	73.69 87	72.41 172	86.32 170	57.93 156	93.81 96	69.18 133	75.65 202	90.11 135

PEN-MVS		77.73 151	77.69 136	77.84 204	87.07 140	53.91 252	87.91 100	91.18 80	77.56 32	73.14 164	88.82 110	61.23 132	89.17 203	59.95 194	72.37 228	90.43 126

CP-MVSNet		78.22 137	78.34 122	77.84 204	87.83 124	54.54 248	87.94 98	91.17 81	77.65 28	73.48 161	88.49 119	62.24 118	88.43 212	62.19 179	74.07 217	90.55 121

114514_t		80.68 88	79.51 93	84.20 75	94.09 18	67.27 98	89.64 53	91.11 82	58.75 242	74.08 158	90.72 73	58.10 155	95.04 53	69.70 130	89.42 71	90.30 129

NR-MVSNet		80.23 101	79.38 96	82.78 126	87.80 125	63.34 171	86.31 149	91.09 83	79.01 21	72.17 174	89.07 105	67.20 56	92.81 137	66.08 155	75.65 202	92.20 74

OpenMVS	72.83 10	79.77 111	78.33 123	84.09 78	85.17 159	69.91 51	90.57 34	90.97 84	66.70 177	72.17 174	91.91 48	54.70 176	93.96 84	61.81 184	90.95 56	88.41 185

MAR-MVS		81.84 70	80.70 75	85.27 50	91.32 53	71.53 35	89.82 48	90.92 85	69.77 145	78.50 84	86.21 171	62.36 115	94.52 69	65.36 160	92.05 46	89.77 150
Zhenyu Xu, Yiguang Liu, Xuelei Shi, Ying Wang, Yunan Zheng: MARMVS: Matching Ambiguity Reduced Multiple View Stereo for Efficient Large Scale Scene Reconstruction. CVPR 2020
OMC-MVS		82.69 59	81.97 64	84.85 60	88.75 100	67.42 94	87.98 95	90.87 86	74.92 74	79.72 70	91.65 52	62.19 119	93.96 84	75.26 85	86.42 102	93.16 50

UA-Net		85.08 42	84.96 39	85.45 47	92.07 45	68.07 87	89.78 50	90.86 87	82.48 2	84.60 29	93.20 30	69.35 44	95.22 44	71.39 121	90.88 57	93.07 52

EPP-MVSNet		83.40 51	83.02 50	84.57 64	90.13 67	64.47 151	92.32 15	90.73 88	74.45 78	79.35 74	91.10 63	69.05 47	95.12 47	72.78 103	87.22 94	94.13 16

v7		80.24 100	79.26 103	83.15 101	84.07 186	64.94 136	87.56 109	90.67 89	72.26 116	78.28 92	86.51 165	61.45 127	94.03 83	75.14 86	77.41 174	90.49 123

Test4		77.83 150	75.90 167	83.62 88	80.24 242	65.25 128	85.27 167	90.67 89	69.03 158	66.48 235	83.75 201	43.07 234	93.00 133	75.93 75	88.66 80	92.62 64

DTE-MVSNet		76.99 167	76.80 148	77.54 209	86.24 149	53.06 255	87.52 111	90.66 91	77.08 40	72.50 170	88.67 113	60.48 143	89.52 198	57.33 216	70.74 238	90.05 142

v10		79.74 112	78.67 111	82.97 114	84.06 187	64.95 135	87.88 101	90.62 92	73.11 100	75.11 150	86.56 163	61.46 126	94.05 82	73.68 95	75.55 204	89.90 144

v1192		79.59 114	78.43 120	83.07 106	83.55 198	64.52 144	86.93 132	90.58 93	70.83 132	77.78 106	85.90 175	59.15 150	93.94 86	73.96 94	77.19 178	90.76 105

v1144		80.03 107	79.03 107	83.01 109	83.78 194	64.51 146	87.11 127	90.57 94	71.96 120	78.08 102	86.20 172	61.41 128	93.94 86	74.93 87	77.23 176	90.60 115

testing_2		75.73 181	73.34 187	82.89 118	77.37 259	65.22 129	84.10 191	90.54 95	69.09 157	60.46 255	81.15 233	40.48 246	92.84 136	76.36 71	80.54 153	90.60 115

XVG-OURS-SEG-HR		80.81 86	79.76 88	83.96 84	85.60 156	68.78 71	83.54 198	90.50 96	70.66 135	76.71 124	91.66 51	60.69 139	91.26 170	76.94 67	81.58 141	91.83 81

HC-MVS		78.19 140	76.99 145	81.78 143	85.66 154	66.99 100	84.66 175	90.47 97	55.08 259	72.02 178	85.27 190	63.83 81	94.11 81	66.10 154	89.80 68	84.24 240

XVG-OURS		80.41 93	79.23 104	83.97 83	85.64 155	69.02 65	83.03 201	90.39 98	71.09 130	77.63 109	91.49 59	54.62 178	91.35 168	75.71 79	83.47 123	91.54 87

MVSFormer		82.85 58	82.05 61	85.24 51	87.35 135	70.21 48	90.50 36	90.38 99	68.55 164	81.32 60	89.47 96	61.68 122	93.46 113	78.98 51	90.26 62	92.05 79

test_djsdf		80.30 99	79.32 98	83.27 97	83.98 188	65.37 125	90.50 36	90.38 99	68.55 164	76.19 131	88.70 111	56.44 165	93.46 113	78.98 51	80.14 157	90.97 100

CPTT-MVS		83.73 46	83.33 47	84.92 59	93.28 26	70.86 42	92.09 19	90.38 99	68.75 162	79.57 71	92.83 39	60.60 142	93.04 132	80.92 41	91.56 51	90.86 103

v144192		79.47 118	78.37 121	82.78 126	83.35 201	63.96 162	86.96 131	90.36 102	69.99 142	77.50 110	85.67 181	60.66 140	93.77 100	74.27 91	76.58 192	90.62 113

v748		77.97 145	76.65 150	81.92 142	82.29 224	63.28 173	87.53 110	90.35 103	73.50 92	70.76 187	85.55 184	58.28 154	92.81 137	68.81 136	72.76 227	89.67 152

v1921920		79.22 123	78.03 127	82.80 123	83.30 204	63.94 163	86.80 136	90.33 104	69.91 143	77.48 111	85.53 185	58.44 153	93.75 102	73.60 98	76.85 186	90.71 108

MVS_111021_HR		85.14 41	84.75 41	86.32 39	91.65 50	72.70 20	85.98 156	90.33 104	76.11 60	82.08 56	91.61 55	71.36 32	94.17 78	81.02 39	92.58 44	92.08 78

v1240		78.99 128	77.78 133	82.64 130	83.21 205	63.54 166	86.62 142	90.30 106	69.74 148	77.33 113	85.68 180	57.04 163	93.76 101	73.13 101	76.92 183	90.62 113

v8		79.97 108	79.02 108	82.80 123	84.09 182	64.50 149	87.96 96	90.29 107	74.13 83	75.24 148	86.81 148	62.88 94	93.89 91	74.39 90	75.40 207	90.00 143

v1neww		80.40 94	79.54 90	82.98 111	84.10 180	64.51 146	87.57 106	90.22 108	73.25 94	78.47 85	86.65 158	62.83 97	93.86 92	75.72 77	77.02 180	90.58 118

v7new		80.40 94	79.54 90	82.98 111	84.10 180	64.51 146	87.57 106	90.22 108	73.25 94	78.47 85	86.65 158	62.83 97	93.86 92	75.72 77	77.02 180	90.58 118

v6		80.40 94	79.54 90	82.98 111	84.09 182	64.50 149	87.57 106	90.22 108	73.25 94	78.47 85	86.63 160	62.84 96	93.86 92	75.73 76	77.02 180	90.58 118

mvs_tets		79.13 125	77.77 134	83.22 99	84.70 165	66.37 107	89.17 59	90.19 111	69.38 152	75.40 141	89.46 98	44.17 230	93.15 125	76.78 70	80.70 149	90.14 133

jajsoiax		79.29 122	77.96 128	83.27 97	84.68 166	66.57 105	89.25 58	90.16 112	69.20 156	75.46 138	89.49 94	45.75 224	93.13 127	76.84 69	80.80 147	90.11 135

Vis-MVSNet		83.46 49	82.80 53	85.43 48	90.25 66	68.74 74	90.30 42	90.13 113	76.33 58	80.87 65	92.89 37	61.00 136	94.20 76	72.45 109	90.97 55	93.35 45
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
PS-MVSNAJ		81.69 72	81.02 73	83.70 87	89.51 76	68.21 85	84.28 187	90.09 114	70.79 133	81.26 62	85.62 183	63.15 89	94.29 71	75.62 81	88.87 76	88.59 178

v1141		80.19 103	79.31 99	82.85 119	83.84 191	64.12 158	87.14 122	90.08 115	73.13 97	78.27 93	86.39 167	62.67 105	93.75 102	75.40 83	76.83 188	90.68 109

divwei89l23v2f112		80.19 103	79.31 99	82.85 119	83.84 191	64.11 160	87.13 125	90.08 115	73.13 97	78.27 93	86.39 167	62.69 103	93.75 102	75.40 83	76.82 189	90.68 109

v1		80.19 103	79.31 99	82.85 119	83.83 193	64.12 158	87.14 122	90.07 117	73.13 97	78.27 93	86.38 169	62.72 102	93.75 102	75.41 82	76.82 189	90.68 109

v2v482		80.23 101	79.29 102	83.05 107	83.62 196	64.14 156	87.04 129	89.97 118	73.61 88	78.18 99	87.22 143	61.10 134	93.82 95	76.11 72	76.78 191	91.18 96

v52		77.94 148	76.37 153	82.67 128	79.39 251	65.52 118	86.43 145	89.94 119	72.28 114	72.15 176	84.94 194	55.70 169	93.44 115	73.64 96	72.84 226	89.06 158

V4		77.95 146	76.37 153	82.67 128	79.40 250	65.52 118	86.43 145	89.94 119	72.28 114	72.14 177	84.95 193	55.72 168	93.44 115	73.64 96	72.86 225	89.05 159

Regformer-3		85.23 40	85.07 38	85.70 46	88.95 92	69.01 66	88.29 88	89.91 121	80.95 7	85.01 19	90.01 85	72.45 25	94.19 77	82.50 35	87.57 90	93.90 27

V42		79.38 121	78.24 125	82.83 122	81.10 235	65.50 121	85.55 164	89.82 122	71.57 126	78.21 97	86.12 173	60.66 140	93.18 124	75.64 80	75.46 206	89.81 149

VNet		82.21 64	82.41 55	81.62 153	90.82 60	60.93 194	84.47 180	89.78 123	76.36 57	84.07 37	91.88 50	64.71 75	90.26 189	70.68 122	88.89 75	93.66 30

XVG-ACMP-BASELINE		76.11 177	74.27 181	81.62 153	83.20 206	64.67 140	83.60 197	89.75 124	69.75 146	71.85 179	87.09 146	32.78 265	92.11 152	69.99 128	80.43 154	88.09 189

EI-MVSNet-Vis-set		84.19 43	83.81 43	85.31 49	88.18 114	67.85 89	87.66 103	89.73 125	80.05 14	82.95 48	89.59 92	70.74 35	94.82 62	80.66 44	84.72 115	93.28 46

EI-MVSNet-UG-set		83.81 45	83.38 46	85.09 53	87.87 122	67.53 93	87.44 114	89.66 126	79.74 15	82.23 55	89.41 102	70.24 37	94.74 64	79.95 48	83.92 120	92.99 57

PAPM		77.68 153	76.40 152	81.51 156	87.29 137	61.85 191	83.78 194	89.59 127	64.74 195	71.23 184	88.70 111	62.59 108	93.66 108	52.66 234	87.03 97	89.01 161

DI_MVS_test_dynamic		79.89 109	78.58 115	83.85 86	82.89 215	65.32 126	86.12 153	89.55 128	69.64 149	70.55 188	85.82 179	57.24 160	93.81 96	76.85 68	88.55 83	92.41 69

MG-MVS		83.41 50	83.45 45	83.28 96	92.74 36	62.28 188	88.17 92	89.50 129	75.22 72	81.49 59	92.74 43	66.75 58	95.11 48	72.85 102	91.58 50	92.45 67

DI_MVS_test_normal		79.81 110	78.45 117	83.89 85	82.70 219	65.40 122	85.82 161	89.48 130	69.39 150	70.12 195	85.66 182	57.15 162	93.71 107	77.08 65	88.62 81	92.56 65

PLC	70.83 11	78.05 143	76.37 153	83.08 105	91.88 49	67.80 90	88.19 91	89.46 131	64.33 200	69.87 202	88.38 121	53.66 185	93.58 109	58.86 203	82.73 130	87.86 192
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
IterMVS-LS		80.06 106	79.38 96	82.11 137	85.89 151	63.20 176	86.79 137	89.34 132	74.19 80	75.45 139	86.72 151	66.62 59	92.39 146	72.58 107	76.86 185	90.75 106
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
API-MVS		81.99 68	81.23 70	84.26 74	90.94 57	70.18 50	91.10 29	89.32 133	71.51 127	78.66 82	88.28 124	65.26 70	95.10 51	64.74 166	91.23 54	87.51 198

GBi-Net		78.40 134	77.40 139	81.40 159	87.60 130	63.01 178	88.39 83	89.28 134	71.63 123	75.34 143	87.28 139	54.80 173	91.11 174	62.72 173	79.57 158	90.09 137

test1		78.40 134	77.40 139	81.40 159	87.60 130	63.01 178	88.39 83	89.28 134	71.63 123	75.34 143	87.28 139	54.80 173	91.11 174	62.72 173	79.57 158	90.09 137

FMVSNet1		77.44 163	76.12 160	81.40 159	86.81 144	63.01 178	88.39 83	89.28 134	70.49 137	74.39 157	87.28 139	49.06 209	91.11 174	60.91 189	78.52 165	90.09 137

cdsmvs_eth3d_5k		19.96 275	26.61 275	0.00 289	0.00 303	0.00 304	0.00 295	89.26 137	0.00 299	0.00 300	88.61 115	61.62 124	0.00 297	0.00 297	0.00 301	0.00 299

ab-mvs		79.51 115	78.97 109	81.14 164	88.46 108	60.91 195	83.84 193	89.24 138	70.36 138	79.03 76	88.87 109	63.23 88	90.21 191	65.12 161	82.57 133	92.28 72

cascas		76.72 171	74.64 175	82.99 110	85.78 153	65.88 114	82.33 204	89.21 139	60.85 227	72.74 167	81.02 235	47.28 215	93.75 102	67.48 143	85.02 110	89.34 156

LTVRE_ROB	69.57 13	76.25 175	74.54 178	81.41 158	88.60 103	64.38 154	79.24 225	89.12 140	70.76 134	69.79 203	87.86 126	49.09 208	93.20 122	56.21 221	80.16 155	86.65 217
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
F-COLMAP		76.38 174	74.33 180	82.50 131	89.28 86	66.95 103	88.41 82	89.03 141	64.05 202	66.83 231	88.61 115	46.78 218	92.89 135	57.48 213	78.55 164	87.67 194

v11		77.45 162	76.06 166	81.59 155	84.22 173	64.52 144	87.11 127	89.02 142	72.76 108	68.76 213	81.90 227	62.09 120	91.71 162	71.98 115	66.73 251	88.56 180

FMVSNet2		78.20 139	77.21 142	81.20 162	87.60 130	62.89 182	87.47 113	89.02 142	71.63 123	75.29 147	87.28 139	54.80 173	91.10 177	62.38 177	79.38 161	89.61 153

v18		77.67 155	76.35 157	81.64 152	84.09 182	64.47 151	87.27 119	89.01 144	72.59 112	69.39 206	82.04 219	62.85 95	91.80 154	72.72 104	67.20 249	88.63 172

ACMH	67.68 16	75.89 179	73.93 183	81.77 144	88.71 101	66.61 104	88.62 75	89.01 144	69.81 144	66.78 232	86.70 155	41.95 242	91.51 167	55.64 222	78.14 170	87.17 206
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
v17		77.68 153	76.35 157	81.69 149	84.15 177	64.65 141	87.33 116	88.99 146	72.70 110	69.25 210	82.07 218	62.82 99	91.79 155	72.69 106	67.15 250	88.63 172

v16		77.69 152	76.36 156	81.68 150	84.15 177	64.63 143	87.33 116	88.99 146	72.69 111	69.31 209	82.08 217	62.80 100	91.79 155	72.70 105	67.23 248	88.63 172

无先验							87.48 112	88.98 148	60.00 234				94.12 79	67.28 145		88.97 164

AdaColmap		80.58 91	79.42 95	84.06 79	93.09 31	68.91 69	89.36 56	88.97 149	69.27 154	75.70 136	89.69 89	57.20 161	95.77 28	63.06 172	88.41 86	87.50 199

v15		77.51 158	76.12 160	81.66 151	84.09 182	64.65 141	87.14 122	88.96 150	72.76 108	68.90 211	81.91 226	62.74 101	91.73 158	72.32 110	66.29 255	88.61 175

V14		77.52 156	76.12 160	81.70 148	84.15 177	64.77 139	87.21 121	88.95 151	72.80 107	68.79 212	81.94 225	62.69 103	91.72 160	72.31 111	66.27 256	88.60 176

EI-MVSNet		80.52 92	79.98 85	82.12 136	84.28 170	63.19 177	86.41 147	88.95 151	74.18 81	78.69 80	87.54 134	66.62 59	92.43 144	72.57 108	80.57 151	90.74 107

MVSTER		79.01 127	77.88 130	82.38 133	83.07 209	64.80 138	84.08 192	88.95 151	69.01 159	78.69 80	87.17 145	54.70 176	92.43 144	74.69 88	80.57 151	89.89 145

V9		77.52 156	76.11 163	81.73 147	84.19 176	64.89 137	87.26 120	88.94 154	72.87 106	68.65 215	81.96 224	62.65 106	91.72 160	72.27 112	66.24 257	88.60 176

v12		77.51 158	76.09 164	81.76 146	84.22 173	64.99 134	87.30 118	88.93 155	72.92 103	68.48 219	81.97 222	62.54 110	91.70 163	72.24 113	66.21 259	88.58 179

v13		77.50 160	76.07 165	81.77 144	84.23 172	65.07 133	87.34 115	88.91 156	72.92 103	68.35 220	81.97 222	62.53 111	91.69 164	72.20 114	66.22 258	88.56 180

liao1		76.53 172	75.30 171	80.21 178	83.93 189	62.32 187	84.66 175	88.81 157	60.23 232	70.16 194	84.07 198	55.30 172	90.73 185	67.37 144	83.21 126	87.59 197

FMVSNet3		77.88 149	76.85 147	80.97 167	86.84 143	62.36 185	86.52 144	88.77 158	71.13 128	75.34 143	86.66 157	54.07 182	91.10 177	62.72 173	79.57 158	89.45 155

WR-MVS		79.49 117	79.22 105	80.27 177	88.79 99	58.35 213	85.06 169	88.61 159	78.56 24	77.65 108	88.34 122	63.81 82	90.66 186	64.98 164	77.22 177	91.80 84

BH-untuned		79.47 118	78.60 113	82.05 138	89.19 90	65.91 113	86.07 155	88.52 160	72.18 117	75.42 140	87.69 129	61.15 133	93.54 110	60.38 192	86.83 98	86.70 216

IS-MVSNet		83.15 53	82.81 52	84.18 76	89.94 70	63.30 172	91.59 23	88.46 161	79.04 20	79.49 72	92.16 44	65.10 72	94.28 72	67.71 141	91.86 48	94.95 2

UGNet		80.83 85	79.59 89	84.54 65	88.04 118	68.09 86	89.42 55	88.16 162	76.95 42	76.22 130	89.46 98	49.30 207	93.94 86	68.48 138	90.31 61	91.60 85
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
VDD-MVS		83.01 57	82.36 57	84.96 56	91.02 56	66.40 106	88.91 66	88.11 163	77.57 30	84.39 33	93.29 29	52.19 193	93.91 89	77.05 66	88.70 79	94.57 7

diffmvs		79.51 115	78.59 114	82.25 135	83.31 203	62.66 183	84.17 188	88.11 163	67.64 170	76.09 135	87.47 136	64.01 79	91.15 173	71.71 118	84.82 114	92.94 58

v148		78.72 131	77.80 132	81.47 157	82.73 218	61.96 190	86.30 150	88.08 165	73.26 93	76.18 132	85.47 187	62.46 113	92.36 147	71.92 117	73.82 222	90.09 137

EG-PatchMatch MVS		74.04 190	71.82 196	80.71 171	84.92 163	67.42 94	85.86 160	88.08 165	66.04 186	64.22 246	83.85 199	35.10 263	92.56 142	57.44 214	80.83 146	82.16 254

MV-PatchMatch		76.12 176	74.46 179	81.13 165	85.37 158	69.79 53	84.42 184	87.95 167	65.03 194	67.46 226	85.33 189	53.28 187	91.73 158	58.01 211	83.27 125	81.85 255

BH-w/o		78.21 138	77.33 141	80.84 168	88.81 97	65.13 132	84.87 172	87.85 168	69.75 146	74.52 156	84.74 197	61.34 129	93.11 128	58.24 209	85.84 109	84.27 239

HY-MVS	69.67 12	77.95 146	77.15 144	80.36 173	87.57 134	60.21 201	83.37 200	87.78 169	66.11 184	75.37 142	87.06 147	63.27 86	90.48 188	61.38 187	82.43 134	90.40 128

1112_ss		77.40 165	76.43 151	80.32 175	89.11 91	60.41 200	83.65 195	87.72 170	62.13 220	73.05 165	86.72 151	62.58 109	89.97 193	62.11 181	80.80 147	90.59 117

mvs_anonymous		79.42 120	79.11 106	80.34 174	84.45 169	57.97 220	82.59 202	87.62 171	67.40 176	76.17 134	88.56 118	68.47 49	89.59 197	70.65 123	86.05 106	93.47 42

ACMH+	68.96 14	76.01 178	74.01 182	82.03 139	88.60 103	65.31 127	88.86 67	87.55 172	70.25 140	67.75 223	87.47 136	41.27 243	93.19 123	58.37 207	75.94 199	87.60 196

TSAR-MVS		85.71 33	85.33 34	86.84 28	91.34 52	72.50 24	89.07 64	87.28 173	76.41 53	85.80 14	90.22 81	74.15 16	95.37 42	81.82 36	91.88 47	92.65 63

BH-RMVSNet		79.61 113	78.44 119	83.14 102	89.38 79	65.93 112	84.95 171	87.15 174	73.56 90	78.19 98	89.79 88	56.67 164	93.36 118	59.53 199	86.74 99	90.13 134

Test_1112_low_res		76.40 173	75.44 170	79.27 188	89.28 86	58.09 216	81.69 208	87.07 175	59.53 238	72.48 171	86.67 156	61.30 130	89.33 200	60.81 191	80.15 156	90.41 127

LS3D		76.95 168	74.82 174	83.37 94	90.45 62	67.36 97	89.15 63	86.94 176	61.87 222	69.52 204	90.61 75	51.71 197	94.53 68	46.38 256	86.71 100	88.21 187

jason		81.39 78	80.29 82	84.70 63	86.63 146	69.90 52	85.95 157	86.77 177	63.24 207	81.07 64	89.47 96	61.08 135	92.15 151	78.33 58	90.07 66	92.05 79
jason: jason.
OurMVSNet-221017-0		74.26 189	72.42 193	79.80 182	83.76 195	59.59 203	85.92 159	86.64 178	66.39 183	66.96 230	87.58 131	39.46 249	91.60 166	65.76 158	69.27 241	88.22 186

VPNet		78.69 132	78.66 112	78.76 194	88.31 112	55.72 244	84.45 182	86.63 179	76.79 45	78.26 96	90.55 76	59.30 149	89.70 196	66.63 150	77.05 179	90.88 102

USDC		70.33 219	68.37 219	76.21 220	80.60 239	56.23 240	79.19 227	86.49 180	60.89 226	61.29 252	85.47 187	31.78 268	89.47 199	53.37 230	76.21 197	82.94 251

lupinMVS		81.39 78	80.27 83	84.76 62	87.35 135	70.21 48	85.55 164	86.41 181	62.85 213	81.32 60	88.61 115	61.68 122	92.24 150	78.41 57	90.26 62	91.83 81

TR-MVS		77.44 163	76.18 159	81.20 162	88.24 113	63.24 174	84.61 178	86.40 182	67.55 173	77.81 105	86.48 166	54.10 181	93.15 125	57.75 212	82.72 131	87.20 205

旧先验1					91.96 46	65.79 116		86.37 183			93.08 35	69.31 45			92.74 42	88.74 169

GA-MVS		76.87 169	75.17 172	81.97 140	82.75 217	62.58 184	81.44 211	86.35 184	72.16 119	74.74 154	82.89 208	46.20 219	92.02 153	68.85 135	81.09 144	91.30 94

CDS-MVSNet		79.07 126	77.70 135	83.17 100	87.60 130	68.23 84	84.40 185	86.20 185	67.49 174	76.36 128	86.54 164	61.54 125	90.79 184	61.86 183	87.33 93	90.49 123
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
MVS_111021_LR		82.61 61	82.11 59	84.11 77	88.82 96	71.58 34	85.15 168	86.16 186	74.69 76	80.47 67	91.04 66	62.29 116	90.55 187	80.33 46	90.08 65	90.20 131

MSDG		73.36 197	70.99 204	80.49 172	84.51 168	65.80 115	80.71 214	86.13 187	65.70 189	65.46 239	83.74 202	44.60 227	90.91 182	51.13 238	76.89 184	84.74 237

TransMVSNet (Re)		75.39 185	74.56 177	77.86 203	85.50 157	57.10 228	86.78 138	86.09 188	72.17 118	71.53 183	87.34 138	63.01 93	89.31 201	56.84 218	61.83 268	87.17 206

VDDNet		81.52 75	80.67 76	84.05 80	90.44 63	64.13 157	89.73 52	85.91 189	71.11 129	83.18 46	93.48 24	50.54 202	93.49 112	73.40 99	88.25 87	94.54 8

Baseline_NR-MVSNet		78.15 141	78.33 123	77.61 208	85.79 152	56.21 241	86.78 138	85.76 190	73.60 89	77.93 104	87.57 132	65.02 73	88.99 205	67.14 148	75.33 208	87.63 195

新几何1				83.42 91	93.13 28	70.71 44		85.48 191	57.43 250	81.80 57	91.98 47	63.28 85	92.27 149	64.60 167	92.99 39	87.27 203

liao		80.84 83	79.77 87	84.05 80	93.11 30	70.78 43	84.66 175	85.42 192	57.37 251	81.76 58	92.02 46	63.41 83	94.12 79	67.28 145	92.93 40	87.26 204

EPNet		83.72 47	82.92 51	86.14 42	84.22 173	69.48 59	91.05 30	85.27 193	81.30 6	76.83 122	91.65 52	66.09 64	95.56 32	76.00 74	93.85 37	93.38 43
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
UnsupCasMVSNet_eth		67.33 233	65.99 234	71.37 244	73.48 271	51.47 261	75.16 248	85.19 194	65.20 192	60.78 254	80.93 237	42.35 237	77.20 261	57.12 217	53.69 279	85.44 229

IB-MVS	68.01 15	75.85 180	73.36 186	83.31 95	84.76 164	66.03 109	83.38 199	85.06 195	70.21 141	69.40 205	81.05 234	45.76 223	94.66 66	65.10 162	75.49 205	89.25 157
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
TAMVS		78.89 130	77.51 138	83.03 108	87.80 125	67.79 91	84.72 174	85.05 196	67.63 171	76.75 123	87.70 128	62.25 117	90.82 183	58.53 206	87.13 95	90.49 123

testdata				79.97 180	90.90 58	64.21 155		84.71 197	59.27 239	85.40 16	92.91 36	62.02 121	89.08 204	68.95 134	91.37 53	86.63 218

MS-PatchMatch		73.83 191	72.67 190	77.30 212	83.87 190	66.02 110	81.82 207	84.66 198	61.37 225	68.61 217	82.82 210	47.29 214	88.21 214	59.27 200	84.32 118	77.68 268

CNLPA		78.08 142	76.79 149	81.97 140	90.40 64	71.07 38	87.59 105	84.55 199	66.03 187	72.38 173	89.64 91	57.56 158	86.04 227	59.61 197	83.35 124	88.79 167

MIMVSNet1		68.58 227	66.78 231	73.98 236	80.07 243	51.82 257	80.77 213	84.37 200	64.40 199	59.75 258	82.16 216	36.47 259	83.63 237	42.73 263	70.33 239	86.48 219

test_0402		72.79 204	70.44 207	79.84 181	88.13 116	65.99 111	85.93 158	84.29 201	65.57 190	67.40 227	85.49 186	46.92 217	92.61 140	35.88 272	74.38 216	80.94 258

EU-MVSNet		68.53 228	67.61 229	71.31 247	78.51 256	47.01 272	84.47 180	84.27 202	42.27 278	66.44 236	84.79 196	40.44 247	83.76 235	58.76 205	68.54 247	83.17 246

COLMAP_ROB	66.92 17	73.01 201	70.41 208	80.81 169	87.13 139	65.63 117	88.30 87	84.19 203	62.96 211	63.80 248	87.69 129	38.04 254	92.56 142	46.66 253	74.91 211	84.24 240
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
HyFIR		77.49 161	77.16 143	78.47 199	88.18 114	58.04 217	75.75 245	84.17 204	60.53 228	76.25 129	89.49 94	62.42 114	87.81 218	68.64 137	89.00 73	89.56 154

CMPMVS	51.72 21	70.19 220	68.16 222	76.28 219	73.15 273	57.55 225	79.47 224	83.92 205	48.02 275	56.48 267	84.81 195	43.13 233	86.42 226	62.67 176	81.81 140	84.89 235
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
XXY-MVS		75.41 184	75.56 169	74.96 227	83.59 197	57.82 223	80.59 217	83.87 206	66.54 182	74.93 153	88.31 123	63.24 87	80.09 250	62.16 180	76.85 186	86.97 211

DP-MVS		76.78 170	74.57 176	83.42 91	93.29 25	69.46 61	88.55 78	83.70 207	63.98 204	70.20 192	88.89 108	54.01 183	94.80 63	46.66 253	81.88 139	86.01 226

SixPastTwentyTwo		73.37 195	71.26 202	79.70 183	85.08 162	57.89 222	85.57 163	83.56 208	71.03 131	65.66 238	85.88 176	42.10 240	92.57 141	59.11 201	63.34 265	88.65 171

CVMVSNet		72.99 202	72.58 191	74.25 234	84.28 170	50.85 264	86.41 147	83.45 209	44.56 277	73.23 163	87.54 134	49.38 205	85.70 229	65.90 156	78.44 167	86.19 221

ITE_SJBPF				78.22 201	81.77 228	60.57 198		83.30 210	69.25 155	67.54 225	87.20 144	36.33 260	87.28 220	54.34 225	74.62 213	86.80 213

test222					91.50 51	68.26 83	84.16 189	83.20 211	54.63 260	79.74 69	91.63 54	58.97 151			91.42 52	86.77 214

EPNet_dtu		75.46 183	74.86 173	77.23 213	82.57 221	54.60 247	86.89 133	83.09 212	71.64 122	66.25 237	85.86 177	55.99 167	88.04 215	54.92 223	86.55 101	89.05 159
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
TDRefinement		67.49 231	64.34 237	76.92 214	73.47 272	61.07 193	84.86 173	82.98 213	59.77 236	58.30 261	85.13 191	26.06 273	87.89 216	47.92 251	60.59 272	81.81 256

OpenMVS_ROB	64.09 19	70.56 218	68.19 221	77.65 207	80.26 241	59.41 207	85.01 170	82.96 214	58.76 241	65.43 240	82.33 213	37.63 257	91.23 172	45.34 260	76.03 198	82.32 252

RPSCF		73.23 199	71.46 198	78.54 197	82.50 222	59.85 202	82.18 205	82.84 215	58.96 240	71.15 185	89.41 102	45.48 226	84.77 234	58.82 204	71.83 232	91.02 99

CostFormer		75.24 186	73.90 184	79.27 188	82.65 220	58.27 215	80.80 212	82.73 216	61.57 223	75.33 146	83.13 207	55.52 170	91.07 180	64.98 164	78.34 169	88.45 183

IterMVS		74.29 188	72.94 189	78.35 200	81.53 230	63.49 167	81.58 210	82.49 217	68.06 169	69.99 199	83.69 203	51.66 198	85.54 230	65.85 157	71.64 233	86.01 226
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
WTY-MVS		75.65 182	75.68 168	75.57 224	86.40 148	56.82 230	77.92 234	82.40 218	65.10 193	76.18 132	87.72 127	63.13 92	80.90 246	60.31 193	81.96 137	89.00 163

tpmp4_e23		73.45 194	71.17 203	80.31 176	83.55 198	59.56 205	81.88 206	82.33 219	57.94 247	70.51 190	81.62 229	51.19 201	91.63 165	53.96 227	77.51 173	89.75 151

PatchFormer-LS_test		74.50 187	73.05 188	78.86 193	82.95 213	59.55 206	81.65 209	82.30 220	67.44 175	71.62 182	78.15 251	52.34 190	88.92 210	65.05 163	75.90 200	88.12 188

DWT-MVSNet_test		73.70 192	71.86 195	79.21 190	82.91 214	58.94 208	82.34 203	82.17 221	65.21 191	71.05 186	78.31 249	44.21 229	90.17 192	63.29 171	77.28 175	88.53 182

FMVSNet5		69.50 223	67.96 225	74.15 235	82.97 212	55.35 245	80.01 220	82.12 222	62.56 216	63.02 249	81.53 230	36.92 258	81.92 243	48.42 248	74.06 218	85.17 233

UnsupCasMVSNet_bld		63.70 245	61.53 246	70.21 250	73.69 270	51.39 262	72.82 255	81.89 223	55.63 257	57.81 262	71.80 267	38.67 252	78.61 254	49.26 246	52.21 281	80.63 259

LFMVS		81.82 71	81.23 70	83.57 90	91.89 48	63.43 170	89.84 47	81.85 224	77.04 41	83.21 45	93.10 32	52.26 192	93.43 117	71.98 115	89.95 67	93.85 28

sss		73.60 193	73.64 185	73.51 237	82.80 216	55.01 246	76.12 240	81.69 225	62.47 217	74.68 155	85.85 178	57.32 159	78.11 257	60.86 190	80.93 145	87.39 200

TinyColmap		67.30 234	64.81 236	74.76 230	81.92 227	56.68 234	80.29 218	81.49 226	60.33 230	56.27 268	83.22 206	24.77 275	87.66 219	45.52 258	69.47 240	79.95 262

tpmvs		71.09 212	69.29 212	76.49 217	82.04 226	56.04 242	78.92 229	81.37 227	64.05 202	67.18 228	78.28 250	49.74 204	89.77 194	49.67 245	72.37 228	83.67 243

ANet_high		50.57 263	46.10 265	63.99 261	48.67 295	39.13 283	70.99 261	80.85 228	61.39 224	31.18 287	57.70 283	17.02 285	73.65 273	31.22 281	15.89 295	79.18 264

LCM-MVSNet		54.25 256	49.68 263	67.97 257	53.73 292	45.28 273	66.85 277	80.78 229	35.96 284	39.45 284	62.23 280	8.70 293	78.06 258	48.24 249	51.20 282	80.57 260

PVSNet	64.34 18	72.08 208	70.87 206	75.69 223	86.21 150	56.44 237	74.37 253	80.73 230	62.06 221	70.17 193	82.23 215	42.86 236	83.31 239	54.77 224	84.45 117	87.32 202

Gipuma		45.18 266	41.86 267	55.16 272	77.03 261	51.52 260	32.50 293	80.52 231	32.46 287	27.12 288	35.02 289	9.52 292	75.50 266	22.31 288	60.21 273	38.45 290
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
LCM-MVSNet-Re		77.05 166	76.94 146	77.36 210	87.20 138	51.60 259	80.06 219	80.46 232	75.20 73	67.69 224	86.72 151	62.48 112	88.98 206	63.44 170	89.25 72	91.51 88

tpm2		73.26 198	71.46 198	78.63 195	83.34 202	56.71 233	80.65 215	80.40 233	56.63 254	73.55 160	82.02 220	51.80 196	91.24 171	56.35 220	78.42 168	87.95 190

CR-MVSNet		73.37 195	71.27 201	79.67 184	81.32 233	65.19 130	75.92 242	80.30 234	59.92 235	72.73 168	81.19 231	52.50 188	86.69 222	59.84 195	77.71 171	87.11 209

Patchmtry		70.74 215	69.16 213	75.49 225	80.72 237	54.07 251	74.94 252	80.30 234	58.34 243	70.01 196	81.19 231	52.50 188	86.54 224	53.37 230	71.09 236	85.87 228

tpm cat1		70.57 217	68.31 220	77.35 211	82.41 223	57.95 221	78.08 233	80.22 236	52.04 269	68.54 218	77.66 254	52.00 195	87.84 217	51.77 235	72.07 231	86.25 220

MDTV_nov1_ep13			69.97 210		83.18 207	53.48 253	77.10 238	80.18 237	60.45 229	69.33 208	80.44 238	48.89 210	86.90 221	51.60 236	78.51 166

AllTest		70.96 213	68.09 224	79.58 186	85.15 160	63.62 164	84.58 179	79.83 238	62.31 218	60.32 256	86.73 149	32.02 266	88.96 208	50.28 240	71.57 234	86.15 222

TestCases				79.58 186	85.15 160	63.62 164		79.83 238	62.31 218	60.32 256	86.73 149	32.02 266	88.96 208	50.28 240	71.57 234	86.15 222

Vis-MVSNet (Re-imp)		78.36 136	78.45 117	78.07 202	88.64 102	51.78 258	86.70 141	79.63 240	74.14 82	75.11 150	90.83 72	61.29 131	89.75 195	58.10 210	91.60 49	92.69 62

MIMVSNet		70.69 216	69.30 211	74.88 228	84.52 167	56.35 239	75.87 244	79.42 241	64.59 196	67.76 222	82.41 212	41.10 244	81.54 245	46.64 255	81.34 142	86.75 215

tpmrst		72.39 205	72.13 194	73.18 239	80.54 240	49.91 266	79.91 221	79.08 242	63.11 208	71.69 181	79.95 241	55.32 171	82.77 241	65.66 159	73.89 220	86.87 212

pm-mvs1		70.87 214	69.03 214	76.39 218	79.41 249	58.92 209	80.64 216	78.89 243	55.71 256	67.14 229	83.58 205	48.48 211	85.25 232	52.94 233	74.48 215	84.99 234

PatchmatchNet		73.12 200	71.33 200	78.49 198	83.18 207	60.85 196	79.63 222	78.57 244	64.13 201	71.73 180	79.81 244	51.20 200	85.97 228	57.40 215	76.36 196	88.66 170
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
MDA-MVSNet-bldmvs		66.68 236	63.66 239	75.75 222	79.28 252	60.56 199	73.92 254	78.35 245	64.43 198	50.13 278	79.87 243	44.02 231	83.67 236	46.10 257	56.86 274	83.03 249

RPMNet		71.62 209	68.94 216	79.67 184	81.32 233	65.19 130	75.92 242	78.30 246	57.60 249	72.73 168	76.45 258	52.30 191	86.69 222	48.14 250	77.71 171	87.11 209

test20.03		67.45 232	66.95 230	68.94 252	75.48 268	44.84 274	77.50 235	77.67 247	66.66 178	63.01 250	83.80 200	47.02 216	78.40 255	42.53 264	68.86 245	83.58 244

test-LLR		72.94 203	72.43 192	74.48 231	81.35 231	58.04 217	78.38 230	77.46 248	66.66 178	69.95 200	79.00 247	48.06 212	79.24 252	66.13 152	84.83 112	86.15 222

test-mter		71.41 210	70.39 209	74.48 231	81.35 231	58.04 217	78.38 230	77.46 248	60.32 231	69.95 200	79.00 247	36.08 261	79.24 252	66.13 152	84.83 112	86.15 222

tpm		72.37 207	71.71 197	74.35 233	82.19 225	52.00 256	79.22 226	77.29 250	64.56 197	72.95 166	83.68 204	51.35 199	83.26 240	58.33 208	75.80 201	87.81 193

LF4IMVS		64.02 244	62.19 245	69.50 251	70.90 277	53.29 254	76.13 239	77.18 251	52.65 268	58.59 259	80.98 236	23.55 276	76.52 263	53.06 232	66.66 252	78.68 265

K. test v3		71.19 211	68.51 218	79.21 190	83.04 211	57.78 224	84.35 186	76.91 252	72.90 105	62.99 251	82.86 209	39.27 250	91.09 179	61.65 185	52.66 280	88.75 168

testgi		66.67 237	66.53 233	67.08 258	75.62 266	41.69 281	75.93 241	76.50 253	66.11 184	65.20 243	86.59 161	35.72 262	74.71 268	43.71 262	73.38 224	84.84 236

PatchMatch-RL		72.38 206	70.90 205	76.80 216	88.60 103	67.38 96	79.53 223	76.17 254	62.75 215	69.36 207	82.00 221	45.51 225	84.89 233	53.62 229	80.58 150	78.12 266

no-one		51.08 261	45.79 266	66.95 259	57.92 290	50.49 265	59.63 285	76.04 255	48.04 274	31.85 285	56.10 285	19.12 283	80.08 251	36.89 271	26.52 288	70.29 278

JIA-IIPM		66.32 239	62.82 244	76.82 215	77.09 260	61.72 192	65.34 278	75.38 256	58.04 246	64.51 245	62.32 279	42.05 241	86.51 225	51.45 237	69.22 242	82.21 253

LP		61.36 247	57.78 250	72.09 241	75.54 267	58.53 212	67.16 276	75.22 257	51.90 270	54.13 269	69.97 271	37.73 256	80.45 249	32.74 276	55.63 276	77.29 270

ADS-MVSNet2		66.20 240	63.33 240	74.82 229	79.92 244	58.75 210	67.55 274	75.19 258	53.37 265	65.25 241	75.86 259	42.32 238	80.53 248	41.57 265	68.91 243	85.18 231

testus		59.00 250	57.91 249	62.25 265	72.25 274	39.09 284	69.74 263	75.02 259	53.04 267	57.21 265	73.72 264	18.76 284	70.33 279	32.86 275	68.57 246	77.35 269

PatchT		68.46 229	67.85 227	70.29 249	80.70 238	43.93 276	72.47 256	74.88 260	60.15 233	70.55 188	76.57 257	49.94 203	81.59 244	50.58 239	74.83 212	85.34 230

test1235678		58.74 251	56.89 254	64.30 260	69.70 278	41.87 280	71.05 259	74.87 261	54.06 261	50.63 277	71.53 268	25.30 274	74.10 271	31.80 280	63.10 266	76.93 272

dp		66.80 235	65.43 235	70.90 248	79.74 246	48.82 269	75.12 250	74.77 262	59.61 237	64.08 247	77.23 255	42.89 235	80.72 247	48.86 247	66.58 253	83.16 247

MDA-MVSNet_test_wron		65.03 241	62.92 241	71.37 244	75.93 262	56.73 231	69.09 269	74.73 263	57.28 252	54.03 271	77.89 252	45.88 220	74.39 270	49.89 244	61.55 269	82.99 250

TESTMET0.1,1		69.89 222	69.00 215	72.55 240	79.27 253	56.85 229	78.38 230	74.71 264	57.64 248	68.09 221	77.19 256	37.75 255	76.70 262	63.92 168	84.09 119	84.10 242

testmv		53.85 257	51.03 259	62.31 264	61.46 287	38.88 285	70.95 262	74.69 265	51.11 272	41.26 281	66.85 274	14.28 287	72.13 276	29.19 282	49.51 283	75.93 274

YYNet1		65.03 241	62.91 242	71.38 243	75.85 263	56.60 235	69.12 268	74.66 266	57.28 252	54.12 270	77.87 253	45.85 221	74.48 269	49.95 243	61.52 270	83.05 248

1111		57.11 254	56.82 255	57.97 269	69.10 279	28.28 293	68.90 270	74.54 267	54.01 262	53.71 272	74.51 262	23.09 277	67.90 283	32.28 277	61.26 271	77.73 267

.test1245		45.55 265	50.02 261	32.14 282	69.10 279	28.28 293	68.90 270	74.54 267	54.01 262	53.71 272	74.51 262	23.09 277	67.90 283	32.28 277	0.02 298	0.25 297

Patchmatch-test1					78.56 255	58.71 211	64.97 280	73.71 269		70.00 198	80.13 240	33.36 264			65.60 260

PMMVS		69.34 224	68.67 217	71.35 246	75.67 265	62.03 189	75.17 247	73.46 270	50.00 273	68.68 214	79.05 245	52.07 194	78.13 256	61.16 188	82.77 129	73.90 275

PVSNet_0	57.27 20	61.67 246	59.27 247	68.85 254	79.61 247	57.44 226	68.01 272	73.44 271	55.93 255	58.54 260	70.41 270	44.58 228	77.55 260	47.01 252	35.91 286	71.55 277

test0.0.03 1		68.00 230	67.69 228	68.90 253	77.55 257	47.43 270	75.70 246	72.95 272	66.66 178	66.56 233	82.29 214	48.06 212	75.87 265	44.97 261	74.51 214	83.41 245

test12356		49.28 264	48.51 264	51.59 275	62.06 286	19.11 298	60.40 283	72.45 273	47.60 276	40.64 283	65.68 275	13.84 288	68.72 281	27.29 284	46.67 285	66.94 280

test2356		59.50 248	58.08 248	63.74 262	71.23 276	41.88 279	67.59 273	72.42 274	53.72 264	57.65 263	70.74 269	26.31 272	72.40 275	32.03 279	71.06 237	76.93 272

ADS-MVSNet		64.36 243	62.88 243	68.78 255	79.92 244	47.17 271	67.55 274	71.18 275	53.37 265	65.25 241	75.86 259	42.32 238	73.99 272	41.57 265	68.91 243	85.18 231

Patchmatch-RL test		68.82 226	66.60 232	75.47 226	77.43 258	59.57 204	71.16 258	70.33 276	62.94 212	68.65 215	81.66 228	31.06 269	85.49 231	69.58 131	66.58 253	81.34 257

gg-mvs-nofinetune		69.95 221	67.96 225	75.94 221	83.07 209	54.51 249	77.23 237	70.29 277	63.11 208	70.32 191	62.33 278	43.62 232	88.69 211	53.88 228	87.76 89	84.62 238

door-mid								69.98 278

FPMVS		53.68 258	51.64 258	59.81 267	65.08 284	51.03 263	69.48 266	69.58 279	41.46 279	40.67 282	72.32 266	16.46 286	70.00 280	24.24 287	65.42 261	58.40 284

door								69.44 280

testpf		56.51 255	57.58 252	53.30 273	71.99 275	41.19 282	46.89 290	69.32 281	58.06 245	52.87 275	69.45 273	27.99 271	72.73 274	59.59 198	62.07 267	45.98 288

wuykxyi23d		39.76 269	33.18 272	59.51 268	46.98 296	44.01 275	57.70 286	67.74 282	24.13 291	13.98 296	34.33 290	1.27 298	71.33 278	34.23 274	18.23 291	63.18 283

EPMVS		69.02 225	68.16 222	71.59 242	79.61 247	49.80 267	77.40 236	66.93 283	62.82 214	70.01 196	79.05 245	45.79 222	77.86 259	56.58 219	75.26 209	87.13 208

lessismore_v0				78.97 192	81.01 236	57.15 227		65.99 284		61.16 253	82.82 210	39.12 251	91.34 169	59.67 196	46.92 284	88.43 184

pmmvs3		57.79 252	54.26 256	68.37 256	64.02 285	56.72 232	75.12 250	65.17 285	40.20 280	52.93 274	69.86 272	20.36 280	75.48 267	45.45 259	55.25 278	72.90 276

MVS-HIRNet		59.14 249	57.67 251	63.57 263	81.65 229	43.50 277	71.73 257	65.06 286	39.59 282	51.43 276	57.73 282	38.34 253	82.58 242	39.53 268	73.95 219	64.62 282

PM-MVS		66.41 238	64.14 238	73.20 238	73.92 269	56.45 236	78.97 228	64.96 287	63.88 206	64.72 244	80.24 239	19.84 282	83.44 238	66.24 151	64.52 263	79.71 263

PMVS	37.38 22	44.16 267	40.28 268	55.82 271	40.82 298	42.54 278	65.12 279	63.99 288	34.43 286	24.48 289	57.12 284	3.92 295	76.17 264	17.10 290	55.52 277	48.75 286
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
new_pmnet		50.91 262	50.29 260	52.78 274	68.58 282	34.94 291	63.71 281	56.63 289	39.73 281	44.95 279	65.47 276	21.93 279	58.48 287	34.98 273	56.62 275	64.92 281

DSMNet-mixed		57.77 253	56.90 253	60.38 266	67.70 283	35.61 288	69.18 267	53.97 290	32.30 289	57.49 264	79.88 242	40.39 248	68.57 282	38.78 269	72.37 228	76.97 271

HyFIR lowres test		51.79 260	50.01 262	57.11 270	68.82 281	49.21 268	60.50 282	53.26 291	34.52 285	43.77 280	64.94 277	20.34 281	71.75 277	39.87 267	64.06 264	50.39 285

PNet_i23d		38.26 270	35.42 270	46.79 277	58.74 288	35.48 289	59.65 284	51.25 292	32.45 288	23.44 292	47.53 287	2.04 297	58.96 286	25.60 286	18.09 293	45.92 289

PMMVS2		40.82 268	38.86 269	46.69 278	53.84 291	16.45 299	48.61 289	49.92 293	37.49 283	31.67 286	60.97 281	8.14 294	56.42 288	28.42 283	30.72 287	67.19 279

E-PMN		31.77 272	30.64 273	35.15 280	52.87 293	27.67 295	57.09 287	47.86 294	24.64 290	16.40 294	33.05 291	11.23 290	54.90 289	14.46 292	18.15 292	22.87 292

EMVS		30.81 273	29.65 274	34.27 281	50.96 294	25.95 297	56.58 288	46.80 295	24.01 292	15.53 295	30.68 292	12.47 289	54.43 290	12.81 293	17.05 294	22.43 293

MVE	26.22 23	30.37 274	25.89 276	43.81 279	44.55 297	35.46 290	28.87 294	39.07 296	18.20 293	18.58 293	40.18 288	2.68 296	47.37 291	17.07 291	23.78 290	48.60 287
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
DUST3R		18.61 276	21.40 277	10.23 286	4.82 300	10.11 300	34.70 292	30.74 297	1.48 296	23.91 291	26.07 293	28.42 270	13.41 294	27.12 285	15.35 296	7.17 294

DeepMVS_CX				27.40 284	40.17 299	26.90 296		24.59 298	17.44 294	23.95 290	48.61 286	9.77 291	26.48 292	18.06 289	24.47 289	28.83 291

N_pmnet		52.79 259	53.26 257	51.40 276	78.99 254	7.68 301	69.52 265	3.89 299	51.63 271	57.01 266	74.98 261	40.83 245	65.96 285	37.78 270	64.67 262	80.56 261

wuyk23d		16.82 277	15.94 278	19.46 285	58.74 288	31.45 292	39.22 291	3.74 300	6.84 295	6.04 297	2.70 295	1.27 298	24.29 293	10.54 294	14.40 297	2.63 295

testmvs		6.04 280	8.02 281	0.10 288	0.08 301	0.03 303	69.74 263	0.04 301	0.05 297	0.31 298	1.68 296	0.02 301	0.04 295	0.24 295	0.02 298	0.25 297

test123		6.12 279	8.11 280	0.14 287	0.06 302	0.09 302	71.05 259	0.03 302	0.04 298	0.25 299	1.30 297	0.05 300	0.03 296	0.21 296	0.01 300	0.29 296

pcd_1.5k_mvsjas		5.26 281	7.02 282	0.00 289	0.00 303	0.00 304	0.00 295	0.00 303	0.00 299	0.00 300	0.00 298	63.15 89	0.00 297	0.00 297	0.00 301	0.00 299

sosnet-low-res		0.00 282	0.00 283	0.00 289	0.00 303	0.00 304	0.00 295	0.00 303	0.00 299	0.00 300	0.00 298	0.00 302	0.00 297	0.00 297	0.00 301	0.00 299

sosnet		0.00 282	0.00 283	0.00 289	0.00 303	0.00 304	0.00 295	0.00 303	0.00 299	0.00 300	0.00 298	0.00 302	0.00 297	0.00 297	0.00 301	0.00 299

uncertanet		0.00 282	0.00 283	0.00 289	0.00 303	0.00 304	0.00 295	0.00 303	0.00 299	0.00 300	0.00 298	0.00 302	0.00 297	0.00 297	0.00 301	0.00 299

Regformer		0.00 282	0.00 283	0.00 289	0.00 303	0.00 304	0.00 295	0.00 303	0.00 299	0.00 300	0.00 298	0.00 302	0.00 297	0.00 297	0.00 301	0.00 299

n2								0.00 303

nn								0.00 303

ab-mvs-re		7.23 278	9.64 279	0.00 289	0.00 303	0.00 304	0.00 295	0.00 303	0.00 299	0.00 300	86.72 151	0.00 302	0.00 297	0.00 297	0.00 301	0.00 299

uanet		0.00 282	0.00 283	0.00 289	0.00 303	0.00 304	0.00 295	0.00 303	0.00 299	0.00 300	0.00 298	0.00 302	0.00 297	0.00 297	0.00 301	0.00 299

test9_res														84.90 11	95.70 10	92.87 59

train_agg														84.97 9	95.71 8	93.02 54

Patchmatch-test					75.77 264	36.09 287					72.86 265

agg_prior2														82.91 31	95.45 12	92.70 60

test_prior4						72.60 22	89.01 65

test_prior2							88.85 68		75.41 67	84.91 21	93.54 22	74.28 14		83.31 24	95.86 4

旧先验2							86.56 143		58.10 244	87.04 10			88.98 206	74.07 93

新几何2							86.29 151

原ACMM2							86.86 134

testdata2													91.01 181	62.37 178

segment_acmp												73.08 20

testdata1							84.14 190		75.71 63

plane_prior7					90.08 68	68.51 80

plane_prior6					89.84 72	68.70 77						60.42 144

plane_prior4											91.00 69

plane_prior3						68.60 79			78.44 25	78.92 78

plane_prior2							91.25 26		79.12 18

plane_prior1					89.90 71

plane_prior						68.71 75	90.38 40		77.62 29						86.16 105

abl_6																91.81 83

HQP5-MVS						66.98 101

HQP-NCC					89.33 80		89.17 59		76.41 53	77.23 117

ACMP_Plane					89.33 80		89.17 59		76.41 53	77.23 117

BP-MVS

High-resolution multi-view benchmark