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.

DeepPCF-MVS	80.84 1	88.10 6	88.56 6	86.73 39	92.24 50	69.03 79	89.57 62	93.39 14	77.53 39	89.79 4	94.12 22	78.98 1	96.58 20	85.66 12	95.72 9	94.58 7

HSP-MVS		89.28 1	89.76 1	87.85 18	94.28 15	73.46 14	92.90 8	92.73 38	80.27 13	91.35 1	94.16 20	78.35 2	96.77 9	89.59 1	94.22 42	93.33 52

APDe-MVS		89.15 2	89.63 2	87.73 20	94.49 8	71.69 41	93.83 2	93.96 3	75.70 72	91.06 2	96.03 1	76.84 3	97.03 5	89.09 2	95.65 13	94.47 11

CNVR-MVS		88.93 4	89.13 4	88.33 3	94.77 2	73.82 6	90.51 39	93.00 25	80.90 10	88.06 9	94.06 24	76.43 4	96.84 7	88.48 4	95.99 4	94.34 15

MCST-MVS		87.37 18	87.25 16	87.73 20	94.53 7	72.46 31	89.82 53	93.82 5	73.07 122	84.86 34	92.89 45	76.22 5	96.33 23	84.89 19	95.13 22	94.40 12

CSCG		86.41 33	86.19 32	87.07 35	92.91 40	72.48 30	90.81 34	93.56 9	73.95 97	83.16 55	91.07 76	75.94 6	95.19 55	79.94 58	94.38 37	93.55 45

HPM-MVS++		89.02 3	89.15 3	88.63 1	95.01 1	76.03 1	92.38 14	92.85 33	80.26 14	87.78 11	94.27 16	75.89 7	96.81 8	87.45 9	96.44 1	93.05 62

TSAR-MVS + MP.		88.02 10	88.11 8	87.72 22	93.68 25	72.13 37	91.41 26	92.35 49	74.62 89	88.90 5	93.85 27	75.75 8	96.00 33	87.80 5	94.63 31	95.04 2

agg_prior1		86.22 36	86.09 35	86.62 42	92.85 41	71.94 39	88.59 89	91.78 74	68.96 190	84.41 39	93.18 38	74.94 9	94.93 65	84.75 22	95.33 19	93.01 65

SD-MVS		88.06 7	88.50 7	86.71 40	92.60 48	72.71 24	91.81 23	93.19 19	77.87 32	90.32 3	94.00 25	74.83 10	93.78 112	87.63 7	94.27 40	93.65 41

DELS-MVS		85.41 47	85.30 45	85.77 55	88.49 128	67.93 107	85.52 194	93.44 12	78.70 28	83.63 52	89.03 117	74.57 11	95.71 38	80.26 56	94.04 43	93.66 36
Christian Sormann, Emanuele Santellani, Mattia Rossi, Andreas Kuhn, Friedrich Fraundorfer: DELS-MVS: Deep Epipolar Line Search for Multi-View Stereo. Winter Conference on Applications of Computer Vision (WACV), 2023
Regformer-2		86.63 29	86.53 27	86.95 36	89.33 97	71.24 44	88.43 92	92.05 58	82.50 1	86.88 14	90.09 94	74.45 12	95.61 39	84.38 25	90.63 68	94.01 27

train_agg		86.43 31	86.20 31	87.13 33	93.26 33	72.96 19	88.75 84	91.89 68	68.69 193	85.00 27	93.10 39	74.43 13	95.41 48	84.97 15	95.71 10	93.02 63

test_8					93.13 35	72.57 28	88.68 87	91.84 71	68.69 193	84.87 33	93.10 39	74.43 13	95.16 56

Regformer-1		86.41 33	86.33 28	86.64 41	89.33 97	70.93 50	88.43 92	91.39 88	82.14 3	86.65 15	90.09 94	74.39 15	95.01 64	83.97 30	90.63 68	93.97 29

TEST9					93.26 33	72.96 19	88.75 84	91.89 68	68.44 197	85.00 27	93.10 39	74.36 16	95.41 48

test_prior3		86.73 26	86.86 25	86.33 46	92.61 46	69.59 71	88.85 79	92.97 30	75.41 76	84.91 29	93.54 29	74.28 17	95.48 43	83.31 32	95.86 6	93.91 30

test_prior2							88.85 79		75.41 76	84.91 29	93.54 29	74.28 17		83.31 32	95.86 6

TSAR-MVS + GP.		85.71 42	85.33 43	86.84 37	91.34 60	72.50 29	89.07 74	87.28 202	76.41 59	85.80 20	90.22 92	74.15 19	95.37 52	81.82 45	91.88 55	92.65 72

SteuartSystems-ACMMP		88.72 5	88.86 5	88.32 4	92.14 52	72.96 19	93.73 3	93.67 7	80.19 15	88.10 8	94.80 4	73.76 20	97.11 3	87.51 8	95.82 8	94.90 4
Skip Steuart: Steuart Systems R&D Blog.
APD-MVS		87.44 15	87.52 13	87.19 31	94.24 16	72.39 32	91.86 22	92.83 34	73.01 123	88.58 6	94.52 7	73.36 21	96.49 21	84.26 27	95.01 23	92.70 69
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
canonicalmvs		85.91 39	85.87 37	86.04 53	89.84 83	69.44 77	90.45 43	93.00 25	76.70 56	88.01 10	91.23 73	73.28 22	93.91 103	81.50 47	88.80 86	94.77 5

segment_acmp												73.08 23

NCCC		88.06 7	88.01 10	88.24 5	94.41 12	73.62 7	91.22 30	92.83 34	81.50 7	85.79 21	93.47 33	73.02 24	97.00 6	84.90 17	94.94 24	94.10 21

agg_prior3		86.16 37	85.85 38	87.10 34	93.31 30	72.86 23	88.77 82	91.68 78	68.29 199	84.26 42	92.83 47	72.83 25	95.42 47	84.97 15	95.71 10	93.02 63

nrg030		83.88 54	83.53 54	84.96 69	86.77 171	69.28 78	90.46 42	92.67 39	74.79 87	82.95 56	91.33 72	72.70 26	93.09 146	80.79 52	79.28 194	92.50 75

Regformer-4		85.68 43	85.45 41	86.35 45	88.95 112	69.67 70	88.29 101	91.29 90	81.73 5	85.36 23	90.01 96	72.62 27	95.35 53	83.28 34	87.57 101	94.03 25

Regformer-3		85.23 49	85.07 47	85.70 56	88.95 112	69.01 81	88.29 101	89.91 135	80.95 9	85.01 26	90.01 96	72.45 28	94.19 90	82.50 43	87.57 101	93.90 32

CDPH-MVS		85.76 41	85.29 46	87.17 32	93.49 29	71.08 45	88.58 90	92.42 47	68.32 198	84.61 36	93.48 31	72.32 29	96.15 30	79.00 60	95.43 15	94.28 18

MP-MVS		87.71 11	87.64 12	87.93 14	94.36 14	73.88 4	92.71 13	92.65 41	77.57 35	83.84 47	94.40 15	72.24 30	96.28 25	85.65 13	95.30 21	93.62 43

DeepC-MVS	79.81 2	87.08 24	86.88 24	87.69 24	91.16 62	72.32 35	90.31 45	93.94 4	77.12 44	82.82 59	94.23 18	72.13 31	97.09 4	84.83 20	95.37 16	93.65 41
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
test12				86.80 38	92.63 45	70.70 55		91.79 73		82.71 61		71.67 32	96.16 29		94.50 33	93.54 46

UniMVSNet_NR-MVSNet		81.88 81	81.54 79	82.92 137	88.46 130	63.46 195	87.13 140	92.37 48	80.19 15	78.38 106	89.14 114	71.66 33	93.05 148	70.05 145	76.46 227	92.25 83

DeepC-MVS_fast	79.65 3	86.91 25	86.62 26	87.76 19	93.52 28	72.37 34	91.26 27	93.04 22	76.62 57	84.22 43	93.36 35	71.44 34	96.76 10	80.82 51	95.33 19	94.16 19
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
MVS_111021_HR		85.14 51	84.75 51	86.32 48	91.65 58	72.70 25	85.98 175	90.33 117	76.11 68	82.08 66	91.61 65	71.36 35	94.17 92	81.02 48	92.58 52	92.08 89

ACMMP_Plus		88.05 9	88.08 9	87.94 11	93.70 23	73.05 18	90.86 33	93.59 8	76.27 66	88.14 7	95.09 3	71.06 36	96.67 13	87.67 6	96.37 2	94.09 22

MP-MVS-pluss		87.67 12	87.72 11	87.54 26	93.64 26	72.04 38	89.80 55	93.50 10	75.17 83	86.34 16	95.29 2	70.86 37	96.00 33	88.78 3	96.04 3	94.58 7
MP-MVS-pluss: MP-MVS-pluss. MP-MVS-pluss
HFP-MVS		87.58 13	87.47 14	87.94 11	94.58 5	73.54 11	93.04 5	93.24 16	76.78 52	84.91 29	94.44 12	70.78 38	96.61 16	84.53 23	94.89 26	93.66 36

#test#		87.33 19	87.13 19	87.94 11	94.58 5	73.54 11	92.34 15	93.24 16	75.23 80	84.91 29	94.44 12	70.78 38	96.61 16	83.75 31	94.89 26	93.66 36

EI-MVSNet-Vis-set		84.19 53	83.81 53	85.31 59	88.18 137	67.85 108	87.66 116	89.73 139	80.05 17	82.95 56	89.59 103	70.74 40	94.82 72	80.66 53	84.72 132	93.28 53

CANet		86.45 30	86.10 34	87.51 27	90.09 77	70.94 49	89.70 59	92.59 42	81.78 4	81.32 73	91.43 71	70.34 41	97.23 2	84.26 27	93.36 46	94.37 13

alignmvs		85.48 44	85.32 44	85.96 54	89.51 92	69.47 75	89.74 57	92.47 43	76.17 67	87.73 12	91.46 70	70.32 42	93.78 112	81.51 46	88.95 83	94.63 6

EI-MVSNet-UG-set		83.81 55	83.38 56	85.09 66	87.87 144	67.53 112	87.44 127	89.66 140	79.74 18	82.23 65	89.41 112	70.24 43	94.74 74	79.95 57	83.92 138	92.99 66

MVS_Test		83.15 65	83.06 60	83.41 114	86.86 168	63.21 202	86.11 173	92.00 62	74.31 92	82.87 58	89.44 111	70.03 44	93.21 137	77.39 77	88.50 95	93.81 35

FC-MVSNet-test		81.52 88	82.02 74	80.03 205	88.42 132	55.97 276	87.95 110	93.42 13	77.10 45	77.38 129	90.98 82	69.96 45	91.79 181	68.46 158	84.50 133	92.33 79

FIs		82.07 78	82.42 66	81.04 190	88.80 119	58.34 242	88.26 103	93.49 11	76.93 49	78.47 102	91.04 77	69.92 46	92.34 170	69.87 148	84.97 128	92.44 77

UniMVSNet (Re)		81.60 87	81.11 84	83.09 125	88.38 133	64.41 179	87.60 117	93.02 24	78.42 31	78.56 99	88.16 137	69.78 47	93.26 136	69.58 150	76.49 226	91.60 98

HPM-MVS		87.11 22	86.98 21	87.50 28	93.88 22	72.16 36	92.19 18	93.33 15	76.07 69	83.81 48	93.95 26	69.77 48	96.01 32	85.15 14	94.66 30	94.32 17

Effi-MVS+		83.62 59	83.08 59	85.24 62	88.38 133	67.45 113	88.89 77	89.15 156	75.50 75	82.27 64	88.28 135	69.61 49	94.45 81	77.81 72	87.84 99	93.84 34

PHI-MVS		86.43 31	86.17 33	87.24 30	90.88 67	70.96 47	92.27 17	94.07 2	72.45 134	85.22 25	91.90 58	69.47 50	96.42 22	83.28 34	95.94 5	94.35 14

UA-Net		85.08 52	84.96 48	85.45 57	92.07 53	68.07 105	89.78 56	90.86 100	82.48 2	84.60 37	93.20 37	69.35 51	95.22 54	71.39 140	90.88 66	93.07 61

旧先验1					91.96 54	65.79 140		86.37 212			93.08 43	69.31 52			92.74 50	88.74 198

region2R		87.42 17	87.20 18	88.09 6	94.63 4	73.55 9	93.03 7	93.12 21	76.73 55	84.45 38	94.52 7	69.09 53	96.70 12	84.37 26	94.83 28	94.03 25

EPP-MVSNet		83.40 63	83.02 61	84.57 77	90.13 75	64.47 177	92.32 16	90.73 101	74.45 91	79.35 90	91.10 74	69.05 54	95.12 57	72.78 122	87.22 108	94.13 20

ACMMPR		87.44 15	87.23 17	88.08 7	94.64 3	73.59 8	93.04 5	93.20 18	76.78 52	84.66 35	94.52 7	68.81 55	96.65 14	84.53 23	94.90 25	94.00 28

mvs_anonymous		79.42 140	79.11 125	80.34 199	84.45 198	57.97 247	82.59 235	87.62 197	67.40 207	76.17 155	88.56 128	68.47 56	89.59 231	70.65 142	86.05 122	93.47 48

MPTG		87.53 14	87.41 15	87.90 15	94.18 19	74.25 2	90.23 47	92.02 59	79.45 19	85.88 18	94.80 4	68.07 57	96.21 27	86.69 10	95.34 17	93.23 54

MTAPA		87.23 20	87.00 20	87.90 15	94.18 19	74.25 2	86.58 160	92.02 59	79.45 19	85.88 18	94.80 4	68.07 57	96.21 27	86.69 10	95.34 17	93.23 54

CP-MVS		87.11 22	86.92 22	87.68 25	94.20 18	73.86 5	93.98 1	92.82 36	76.62 57	83.68 49	94.46 11	67.93 59	95.95 35	84.20 29	94.39 36	93.23 54

PAPM_NR		83.02 68	82.41 67	84.82 74	92.47 49	66.37 130	87.93 112	91.80 72	73.82 103	77.32 131	90.66 85	67.90 60	94.90 69	70.37 144	89.48 80	93.19 58

PGM-MVS		86.68 27	86.27 30	87.90 15	94.22 17	73.38 15	90.22 48	93.04 22	75.53 74	83.86 46	94.42 14	67.87 61	96.64 15	82.70 41	94.57 32	93.66 36

PAPR		81.66 86	80.89 87	83.99 99	90.27 73	64.00 187	86.76 156	91.77 76	68.84 191	77.13 138	89.50 104	67.63 62	94.88 70	67.55 162	88.52 94	93.09 60

Fast-Effi-MVS+		80.81 100	79.92 99	83.47 110	88.85 114	64.51 171	85.53 192	89.39 147	70.79 157	78.49 101	85.06 222	67.54 63	93.58 123	67.03 170	86.58 115	92.32 80

XVS		87.18 21	86.91 23	88.00 9	94.42 10	73.33 16	92.78 9	92.99 27	79.14 21	83.67 50	94.17 19	67.45 64	96.60 18	83.06 36	94.50 33	94.07 23

X-MVStestdata		80.37 117	77.83 151	88.00 9	94.42 10	73.33 16	92.78 9	92.99 27	79.14 21	83.67 50	12.47 337	67.45 64	96.60 18	83.06 36	94.50 33	94.07 23

NR-MVSNet		80.23 120	79.38 115	82.78 149	87.80 147	63.34 198	86.31 168	91.09 96	79.01 26	72.17 206	89.07 115	67.20 66	92.81 158	66.08 176	75.65 236	92.20 85

MSLP-MVS++		85.43 46	85.76 40	84.45 81	91.93 55	70.24 58	90.71 36	92.86 32	77.46 41	84.22 43	92.81 50	67.16 67	92.94 152	80.36 54	94.35 38	90.16 147

MG-MVS		83.41 62	83.45 55	83.28 117	92.74 43	62.28 216	88.17 105	89.50 144	75.22 81	81.49 72	92.74 51	66.75 68	95.11 58	72.85 121	91.58 58	92.45 76

EI-MVSNet		80.52 111	79.98 98	82.12 160	84.28 199	63.19 204	86.41 165	88.95 168	74.18 94	78.69 96	87.54 154	66.62 69	92.43 165	72.57 127	80.57 180	90.74 121

IterMVS-LS		80.06 125	79.38 115	82.11 161	85.89 177	63.20 203	86.79 153	89.34 148	74.19 93	75.45 167	86.72 174	66.62 69	92.39 167	72.58 126	76.86 218	90.75 120
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
mPP-MVS		86.67 28	86.32 29	87.72 22	94.41 12	73.55 9	92.74 11	92.22 52	76.87 50	82.81 60	94.25 17	66.44 71	96.24 26	82.88 40	94.28 39	93.38 49

WR-MVS_H		78.51 154	78.49 136	78.56 232	88.02 141	56.38 271	88.43 92	92.67 39	77.14 43	73.89 189	87.55 153	66.25 72	89.24 237	58.92 230	73.55 259	90.06 156

PCF-MVS	73.52 7	80.38 116	78.84 129	85.01 68	87.71 151	68.99 82	83.65 227	91.46 87	63.00 244	77.77 124	90.28 89	66.10 73	95.09 62	61.40 211	88.22 98	90.94 115
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
EPNet		83.72 57	82.92 63	86.14 51	84.22 202	69.48 74	91.05 32	85.27 222	81.30 8	76.83 139	91.65 62	66.09 74	95.56 41	76.00 89	93.85 44	93.38 49
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
原ACMM1				84.35 85	93.01 39	68.79 85		92.44 44	63.96 239	81.09 78	91.57 66	66.06 75	95.45 45	67.19 167	94.82 29	88.81 195

PVSNet_BlendedMVS		80.60 108	80.02 97	82.36 158	88.85 114	65.40 146	86.16 171	92.00 62	69.34 179	78.11 117	86.09 202	66.02 76	94.27 85	71.52 138	82.06 164	87.39 233

PVSNet_Blended		80.98 94	80.34 93	82.90 138	88.85 114	65.40 146	84.43 215	92.00 62	67.62 203	78.11 117	85.05 223	66.02 76	94.27 85	71.52 138	89.50 79	89.01 189

APD-MVS_3200maxsize		85.97 38	85.88 36	86.22 49	92.69 44	69.53 73	91.93 21	92.99 27	73.54 109	85.94 17	94.51 10	65.80 78	95.61 39	83.04 38	92.51 53	93.53 47

PVSNet_Blended_VisFu		82.62 72	81.83 77	84.96 69	90.80 69	69.76 68	88.74 86	91.70 77	69.39 176	78.96 93	88.46 130	65.47 79	94.87 71	74.42 105	88.57 91	90.24 145

API-MVS		81.99 80	81.23 82	84.26 88	90.94 65	70.18 64	91.10 31	89.32 149	71.51 150	78.66 98	88.28 135	65.26 80	95.10 61	64.74 187	91.23 63	87.51 231

TranMVSNet+NR-MVSNet		80.84 97	80.31 94	82.42 156	87.85 145	62.33 214	87.74 115	91.33 89	80.55 12	77.99 120	89.86 98	65.23 81	92.62 160	67.05 169	75.24 245	92.30 81

IS-MVSNet		83.15 65	82.81 64	84.18 90	89.94 81	63.30 199	91.59 24	88.46 183	79.04 25	79.49 88	92.16 52	65.10 82	94.28 84	67.71 160	91.86 56	94.95 3

DU-MVS		81.12 93	80.52 92	82.90 138	87.80 147	63.46 195	87.02 145	91.87 70	79.01 26	78.38 106	89.07 115	65.02 83	93.05 148	70.05 145	76.46 227	92.20 85

Baseline_NR-MVSNet		78.15 162	78.33 143	77.61 245	85.79 178	56.21 274	86.78 154	85.76 219	73.60 107	77.93 121	87.57 152	65.02 83	88.99 241	67.14 168	75.33 242	87.63 228

VNet		82.21 76	82.41 67	81.62 177	90.82 68	60.93 223	84.47 211	89.78 137	76.36 64	84.07 45	91.88 59	64.71 85	90.26 222	70.68 141	88.89 84	93.66 36

MVS_0304		86.37 35	85.81 39	88.02 8	90.13 75	72.39 32	89.66 60	92.75 37	81.64 6	82.66 63	92.04 54	64.44 86	97.35 1	84.76 21	94.25 41	94.33 16

Test By Simon												64.33 87

ACMMP		85.89 40	85.39 42	87.38 29	93.59 27	72.63 26	92.74 11	93.18 20	76.78 52	80.73 82	93.82 28	64.33 87	96.29 24	82.67 42	90.69 67	93.23 54
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
DP-MVS Recon		83.11 67	82.09 72	86.15 50	94.44 9	70.92 51	88.79 81	92.20 53	70.53 162	79.17 91	91.03 79	64.12 89	96.03 31	68.39 159	90.14 73	91.50 102

diffmvs		79.51 135	78.59 133	82.25 159	83.31 234	62.66 211	84.17 220	88.11 186	67.64 201	76.09 156	87.47 156	64.01 90	91.15 206	71.71 137	84.82 131	92.94 67

CLD-MVS		82.31 75	81.65 78	84.29 87	88.47 129	67.73 111	85.81 181	92.35 49	75.78 70	78.33 108	86.58 187	64.01 90	94.35 82	76.05 88	87.48 106	90.79 118
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
MVS		78.19 161	76.99 165	81.78 167	85.66 180	66.99 121	84.66 206	90.47 110	55.08 299	72.02 210	85.27 218	63.83 92	94.11 95	66.10 175	89.80 77	84.24 277

WR-MVS		79.49 137	79.22 124	80.27 202	88.79 120	58.35 241	85.06 200	88.61 181	78.56 29	77.65 125	88.34 133	63.81 93	90.66 219	64.98 185	77.22 210	91.80 97

1121		80.84 97	79.77 102	84.05 94	93.11 37	70.78 53	84.66 206	85.42 221	57.37 289	81.76 71	92.02 55	63.41 94	94.12 93	67.28 165	92.93 48	87.26 238

VPA-MVSNet		80.60 108	80.55 91	80.76 194	88.07 139	60.80 226	86.86 150	91.58 81	75.67 73	80.24 84	89.45 110	63.34 95	90.25 223	70.51 143	79.22 195	91.23 108

新几何1				83.42 112	93.13 35	70.71 54		85.48 220	57.43 288	81.80 70	91.98 56	63.28 96	92.27 171	64.60 188	92.99 47	87.27 237

HY-MVS	69.67 12	77.95 168	77.15 163	80.36 198	87.57 157	60.21 230	83.37 232	87.78 195	66.11 216	75.37 170	87.06 169	63.27 97	90.48 221	61.38 212	82.43 162	90.40 142

XXY-MVS		75.41 216	75.56 192	74.96 266	83.59 228	57.82 251	80.59 252	83.87 234	66.54 213	74.93 183	88.31 134	63.24 98	80.09 294	62.16 203	76.85 219	86.97 245

ab-mvs		79.51 135	78.97 128	81.14 188	88.46 130	60.91 224	83.84 225	89.24 154	70.36 164	79.03 92	88.87 119	63.23 99	90.21 224	65.12 182	82.57 161	92.28 82

xiu_mvs_v2_base		81.69 84	81.05 85	83.60 107	89.15 108	68.03 106	84.46 213	90.02 131	70.67 160	81.30 76	86.53 190	63.17 100	94.19 90	75.60 97	88.54 93	88.57 209

pcd_1.5k_mvsjas		5.26 321	7.02 322	0.00 332	0.00 345	0.00 346	0.00 337	0.00 348	0.00 341	0.00 342	0.00 343	63.15 101	0.00 344	0.00 341	0.00 342	0.00 340

PS-MVSNAJss		82.07 78	81.31 80	84.34 86	86.51 173	67.27 118	89.27 66	91.51 84	71.75 144	79.37 89	90.22 92	63.15 101	94.27 85	77.69 73	82.36 163	91.49 103

PS-MVSNAJ		81.69 84	81.02 86	83.70 105	89.51 92	68.21 103	84.28 219	90.09 127	70.79 157	81.26 77	85.62 211	63.15 101	94.29 83	75.62 96	88.87 85	88.59 207

WTY-MVS		75.65 213	75.68 191	75.57 262	86.40 174	56.82 262	77.92 275	82.40 251	65.10 226	76.18 153	87.72 147	63.13 104	80.90 290	60.31 219	81.96 165	89.00 191

TransMVSNet (Re)		75.39 217	74.56 208	77.86 240	85.50 184	57.10 260	86.78 154	86.09 217	72.17 141	71.53 215	87.34 158	63.01 105	89.31 236	56.84 249	61.83 306	87.17 240

v8		79.97 128	79.02 127	82.80 146	84.09 212	64.50 175	87.96 109	90.29 120	74.13 96	75.24 176	86.81 171	62.88 106	93.89 105	74.39 106	75.40 241	90.00 158

v18		77.67 177	76.35 179	81.64 176	84.09 212	64.47 177	87.27 133	89.01 161	72.59 133	69.39 241	82.04 255	62.85 107	91.80 180	72.72 123	67.20 288	88.63 201

v6		80.40 113	79.54 108	82.98 132	84.09 212	64.50 175	87.57 119	90.22 121	73.25 115	78.47 102	86.63 184	62.84 108	93.86 106	75.73 91	77.02 213	90.58 132

v1neww		80.40 113	79.54 108	82.98 132	84.10 210	64.51 171	87.57 119	90.22 121	73.25 115	78.47 102	86.65 182	62.83 109	93.86 106	75.72 92	77.02 213	90.58 132

v7new		80.40 113	79.54 108	82.98 132	84.10 210	64.51 171	87.57 119	90.22 121	73.25 115	78.47 102	86.65 182	62.83 109	93.86 106	75.72 92	77.02 213	90.58 132

v17		77.68 175	76.35 179	81.69 173	84.15 207	64.65 166	87.33 130	88.99 163	72.70 131	69.25 245	82.07 254	62.82 111	91.79 181	72.69 125	67.15 289	88.63 201

v16		77.69 174	76.36 178	81.68 174	84.15 207	64.63 168	87.33 130	88.99 163	72.69 132	69.31 244	82.08 253	62.80 112	91.79 181	72.70 124	67.23 287	88.63 201

abl_6		85.23 49	84.95 49	86.07 52	92.23 51	70.48 57	90.80 35	92.08 57	73.51 110	85.26 24	94.16 20	62.75 113	95.92 36	82.46 44	91.30 62	91.81 96

v15		77.51 182	76.12 182	81.66 175	84.09 212	64.65 166	87.14 137	88.96 167	72.76 129	68.90 246	81.91 262	62.74 114	91.73 185	72.32 129	66.29 294	88.61 204

v1		80.19 122	79.31 118	82.85 141	83.83 224	64.12 184	87.14 137	90.07 130	73.13 118	78.27 110	86.38 197	62.72 115	93.75 116	75.41 98	76.82 222	90.68 123

divwei89l23v2f112		80.19 122	79.31 118	82.85 141	83.84 222	64.11 186	87.13 140	90.08 128	73.13 118	78.27 110	86.39 193	62.69 116	93.75 116	75.40 99	76.82 222	90.68 123

V14		77.52 180	76.12 182	81.70 172	84.15 207	64.77 164	87.21 136	88.95 168	72.80 128	68.79 247	81.94 261	62.69 116	91.72 187	72.31 130	66.27 295	88.60 205

v1141		80.19 122	79.31 118	82.85 141	83.84 222	64.12 184	87.14 137	90.08 128	73.13 118	78.27 110	86.39 193	62.67 118	93.75 116	75.40 99	76.83 221	90.68 123

V9		77.52 180	76.11 185	81.73 171	84.19 206	64.89 161	87.26 134	88.94 171	72.87 127	68.65 250	81.96 260	62.65 119	91.72 187	72.27 131	66.24 296	88.60 205

HPM-MVS_fast		85.35 48	84.95 49	86.57 44	93.69 24	70.58 56	92.15 19	91.62 79	73.89 98	82.67 62	94.09 23	62.60 120	95.54 42	80.93 49	92.93 48	93.57 44

PAPM		77.68 175	76.40 174	81.51 180	87.29 163	61.85 220	83.78 226	89.59 141	64.74 229	71.23 217	88.70 121	62.59 121	93.66 122	52.66 266	87.03 111	89.01 189

1112_ss		77.40 188	76.43 173	80.32 200	89.11 111	60.41 229	83.65 227	87.72 196	62.13 255	73.05 196	86.72 174	62.58 122	89.97 226	62.11 205	80.80 176	90.59 131

v12		77.51 182	76.09 186	81.76 170	84.22 202	64.99 158	87.30 132	88.93 172	72.92 124	68.48 254	81.97 258	62.54 123	91.70 190	72.24 132	66.21 298	88.58 208

v13		77.50 184	76.07 187	81.77 168	84.23 201	65.07 157	87.34 129	88.91 173	72.92 124	68.35 255	81.97 258	62.53 124	91.69 191	72.20 133	66.22 297	88.56 210

LCM-MVSNet-Re		77.05 190	76.94 166	77.36 248	87.20 164	51.60 298	80.06 255	80.46 274	75.20 82	67.69 259	86.72 174	62.48 125	88.98 242	63.44 192	89.25 82	91.51 101

v148		78.72 151	77.80 152	81.47 181	82.73 250	61.96 219	86.30 169	88.08 190	73.26 114	76.18 153	85.47 215	62.46 126	92.36 169	71.92 136	73.82 257	90.09 152

MAR-MVS		81.84 82	80.70 88	85.27 61	91.32 61	71.53 43	89.82 53	90.92 98	69.77 171	78.50 100	86.21 199	62.36 127	94.52 79	65.36 181	92.05 54	89.77 172
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
MVS_111021_LR		82.61 73	82.11 71	84.11 91	88.82 117	71.58 42	85.15 199	86.16 215	74.69 88	80.47 83	91.04 77	62.29 128	90.55 220	80.33 55	90.08 74	90.20 146

TAMVS		78.89 150	77.51 158	83.03 129	87.80 147	67.79 110	84.72 205	85.05 225	67.63 202	76.75 140	87.70 148	62.25 129	90.82 216	58.53 235	87.13 109	90.49 137

CP-MVSNet		78.22 158	78.34 142	77.84 241	87.83 146	54.54 286	87.94 111	91.17 94	77.65 33	73.48 191	88.49 129	62.24 130	88.43 250	62.19 202	74.07 252	90.55 135

OMC-MVS		82.69 71	81.97 76	84.85 73	88.75 122	67.42 114	87.98 108	90.87 99	74.92 86	79.72 86	91.65 62	62.19 131	93.96 98	75.26 101	86.42 118	93.16 59

v11		77.45 185	76.06 188	81.59 179	84.22 202	64.52 169	87.11 142	89.02 159	72.76 129	68.76 248	81.90 263	62.09 132	91.71 189	71.98 134	66.73 290	88.56 210

testdata				79.97 206	90.90 66	64.21 181		84.71 226	59.27 275	85.40 22	92.91 44	62.02 133	89.08 240	68.95 154	91.37 61	86.63 253

MVSFormer		82.85 70	82.05 73	85.24 62	87.35 158	70.21 59	90.50 40	90.38 112	68.55 195	81.32 73	89.47 106	61.68 134	93.46 128	78.98 61	90.26 71	92.05 90

lupinMVS		81.39 91	80.27 96	84.76 75	87.35 158	70.21 59	85.55 190	86.41 210	62.85 247	81.32 73	88.61 125	61.68 134	92.24 173	78.41 67	90.26 71	91.83 94

cdsmvs_eth3d_5k		19.96 315	26.61 315	0.00 332	0.00 345	0.00 346	0.00 337	89.26 153	0.00 341	0.00 342	88.61 125	61.62 136	0.00 344	0.00 341	0.00 342	0.00 340

CDS-MVSNet		79.07 146	77.70 155	83.17 121	87.60 153	68.23 102	84.40 217	86.20 214	67.49 205	76.36 149	86.54 189	61.54 137	90.79 217	61.86 207	87.33 107	90.49 137
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
v10		79.74 132	78.67 130	82.97 136	84.06 217	64.95 159	87.88 114	90.62 105	73.11 121	75.11 179	86.56 188	61.46 138	94.05 96	73.68 111	75.55 238	89.90 166

v7		80.24 119	79.26 122	83.15 122	84.07 216	64.94 160	87.56 122	90.67 102	72.26 139	78.28 109	86.51 191	61.45 139	94.03 97	75.14 102	77.41 207	90.49 137

v1144		80.03 126	79.03 126	83.01 130	83.78 225	64.51 171	87.11 142	90.57 107	71.96 143	78.08 119	86.20 200	61.41 140	93.94 100	74.93 103	77.23 209	90.60 129

BH-w/o		78.21 159	77.33 161	80.84 192	88.81 118	65.13 156	84.87 203	87.85 194	69.75 172	74.52 186	84.74 228	61.34 141	93.11 145	58.24 238	85.84 125	84.27 276

Test_1112_low_res		76.40 200	75.44 193	79.27 217	89.28 103	58.09 244	81.69 243	87.07 204	59.53 273	72.48 202	86.67 180	61.30 142	89.33 235	60.81 217	80.15 186	90.41 141

Vis-MVSNet (Re-imp)		78.36 157	78.45 137	78.07 239	88.64 124	51.78 297	86.70 157	79.63 282	74.14 95	75.11 179	90.83 83	61.29 143	89.75 228	58.10 239	91.60 57	92.69 71

PEN-MVS		77.73 173	77.69 156	77.84 241	87.07 166	53.91 290	87.91 113	91.18 93	77.56 37	73.14 195	88.82 120	61.23 144	89.17 238	59.95 221	72.37 265	90.43 140

pm-mvs1		77.25 189	76.68 170	78.93 226	84.22 202	58.62 239	86.41 165	88.36 184	71.37 151	73.31 192	88.01 141	61.22 145	89.15 239	64.24 189	73.01 261	89.03 188

BH-untuned		79.47 138	78.60 132	82.05 162	89.19 107	65.91 137	86.07 174	88.52 182	72.18 140	75.42 168	87.69 149	61.15 146	93.54 125	60.38 218	86.83 112	86.70 251

v2v482		80.23 120	79.29 121	83.05 128	83.62 227	64.14 182	87.04 144	89.97 132	73.61 106	78.18 116	87.22 163	61.10 147	93.82 109	76.11 87	76.78 224	91.18 109

jason		81.39 91	80.29 95	84.70 76	86.63 172	69.90 66	85.95 176	86.77 206	63.24 241	81.07 79	89.47 106	61.08 148	92.15 174	78.33 68	90.07 75	92.05 90
jason: jason.
Vis-MVSNet		83.46 61	82.80 65	85.43 58	90.25 74	68.74 89	90.30 46	90.13 126	76.33 65	80.87 81	92.89 45	61.00 149	94.20 89	72.45 128	90.97 64	93.35 51
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
TAPA-MVS	73.13 9	79.15 144	77.94 149	82.79 148	89.59 90	62.99 209	88.16 106	91.51 84	65.77 220	77.14 137	91.09 75	60.91 150	93.21 137	50.26 275	87.05 110	92.17 87
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
PS-CasMVS		78.01 166	78.09 146	77.77 243	87.71 151	54.39 288	88.02 107	91.22 91	77.50 40	73.26 193	88.64 124	60.73 151	88.41 251	61.88 206	73.88 256	90.53 136

OPM-MVS		83.50 60	82.95 62	85.14 64	88.79 120	70.95 48	89.13 73	91.52 83	77.55 38	80.96 80	91.75 60	60.71 152	94.50 80	79.67 59	86.51 117	89.97 165

XVG-OURS-SEG-HR		80.81 100	79.76 103	83.96 101	85.60 182	68.78 86	83.54 230	90.50 109	70.66 161	76.71 141	91.66 61	60.69 153	91.26 203	76.94 82	81.58 169	91.83 94

v144192		79.47 138	78.37 141	82.78 149	83.35 232	63.96 188	86.96 146	90.36 115	69.99 168	77.50 127	85.67 209	60.66 154	93.77 114	74.27 107	76.58 225	90.62 127

V42		79.38 141	78.24 145	82.83 144	81.10 273	65.50 145	85.55 190	89.82 136	71.57 149	78.21 114	86.12 201	60.66 154	93.18 141	75.64 95	75.46 240	89.81 171

CPTT-MVS		83.73 56	83.33 57	84.92 72	93.28 32	70.86 52	92.09 20	90.38 112	68.75 192	79.57 87	92.83 47	60.60 156	93.04 150	80.92 50	91.56 59	90.86 117

DTE-MVSNet		76.99 191	76.80 168	77.54 247	86.24 175	53.06 294	87.52 124	90.66 104	77.08 46	72.50 201	88.67 123	60.48 157	89.52 232	57.33 246	70.74 276	90.05 157

HQP_MVS		83.64 58	83.14 58	85.14 64	90.08 78	68.71 91	91.25 28	92.44 44	79.12 23	78.92 94	91.00 80	60.42 158	95.38 50	78.71 63	86.32 119	91.33 105

plane_prior6					89.84 83	68.70 93						60.42 158

3Dnovator+	77.84 4	85.48 44	84.47 52	88.51 2	91.08 63	73.49 13	93.18 4	93.78 6	80.79 11	76.66 142	93.37 34	60.40 160	96.75 11	77.20 78	93.73 45	95.29 1

HQP2-MVS												60.17 161

HQP-MVS		82.61 73	82.02 74	84.37 83	89.33 97	66.98 122	89.17 68	92.19 54	76.41 59	77.23 134	90.23 91	60.17 161	95.11 58	77.47 75	85.99 123	91.03 111

VPNet		78.69 152	78.66 131	78.76 229	88.31 135	55.72 282	84.45 214	86.63 208	76.79 51	78.26 113	90.55 87	59.30 163	89.70 230	66.63 171	77.05 212	90.88 116

v1192		79.59 134	78.43 140	83.07 127	83.55 229	64.52 169	86.93 148	90.58 106	70.83 156	77.78 123	85.90 203	59.15 164	93.94 100	73.96 110	77.19 211	90.76 119

test222					91.50 59	68.26 101	84.16 221	83.20 244	54.63 300	79.74 85	91.63 64	58.97 165			91.42 60	86.77 249

CHOSEN 1792x2688		77.63 178	75.69 190	83.44 111	89.98 80	68.58 96	78.70 269	87.50 199	56.38 294	75.80 159	86.84 170	58.67 166	91.40 200	61.58 210	85.75 126	90.34 143

3Dnovator	76.31 5	83.38 64	82.31 70	86.59 43	87.94 143	72.94 22	90.64 37	92.14 56	77.21 42	75.47 165	92.83 47	58.56 167	94.72 75	73.24 119	92.71 51	92.13 88

v1921920		79.22 143	78.03 147	82.80 146	83.30 235	63.94 189	86.80 152	90.33 117	69.91 169	77.48 128	85.53 213	58.44 168	93.75 116	73.60 114	76.85 219	90.71 122

v748		77.97 167	76.65 171	81.92 166	82.29 257	63.28 200	87.53 123	90.35 116	73.50 111	70.76 220	85.55 212	58.28 169	92.81 158	68.81 156	72.76 264	89.67 174

114514_t		80.68 106	79.51 111	84.20 89	94.09 21	67.27 118	89.64 61	91.11 95	58.75 279	74.08 188	90.72 84	58.10 170	95.04 63	69.70 149	89.42 81	90.30 144

v7n		78.97 149	77.58 157	83.14 123	83.45 231	65.51 144	88.32 99	91.21 92	73.69 105	72.41 203	86.32 198	57.93 171	93.81 110	69.18 153	75.65 236	90.11 150

QAPM		80.88 95	79.50 112	85.03 67	88.01 142	68.97 83	91.59 24	92.00 62	66.63 212	75.15 178	92.16 52	57.70 172	95.45 45	63.52 191	88.76 87	90.66 126

HyFIR lowres test		77.53 179	75.40 195	83.94 102	89.59 90	66.62 126	80.36 253	88.64 180	56.29 295	76.45 145	85.17 219	57.64 173	93.28 135	61.34 213	83.10 154	91.91 92

CNLPA		78.08 163	76.79 169	81.97 164	90.40 72	71.07 46	87.59 118	84.55 228	66.03 219	72.38 204	89.64 102	57.56 174	86.04 268	59.61 224	83.35 150	88.79 196

sss		73.60 227	73.64 217	73.51 277	82.80 248	55.01 284	76.12 281	81.69 263	62.47 252	74.68 185	85.85 206	57.32 175	78.11 302	60.86 216	80.93 173	87.39 233

Effi-MVS+-dtu		80.03 126	78.57 135	84.42 82	85.13 189	68.74 89	88.77 82	88.10 188	74.99 84	74.97 182	83.49 240	57.27 176	93.36 133	73.53 115	80.88 174	91.18 109

mvs-test1		80.88 95	79.40 114	85.29 60	85.13 189	69.75 69	89.28 65	88.10 188	74.99 84	76.44 148	86.72 174	57.27 176	94.26 88	73.53 115	83.18 153	91.87 93

DI_MVS_plusplus_test		79.89 129	78.58 134	83.85 104	82.89 247	65.32 150	86.12 172	89.55 142	69.64 175	70.55 221	85.82 207	57.24 178	93.81 110	76.85 83	88.55 92	92.41 78

AdaColmap		80.58 110	79.42 113	84.06 93	93.09 38	68.91 84	89.36 64	88.97 166	69.27 180	75.70 164	89.69 100	57.20 179	95.77 37	63.06 195	88.41 96	87.50 232

test_normal		79.81 130	78.45 137	83.89 103	82.70 251	65.40 146	85.82 180	89.48 145	69.39 176	70.12 230	85.66 210	57.15 180	93.71 121	77.08 80	88.62 90	92.56 74

v1240		78.99 148	77.78 153	82.64 153	83.21 236	63.54 192	86.62 159	90.30 119	69.74 174	77.33 130	85.68 208	57.04 181	93.76 115	73.13 120	76.92 216	90.62 127

BH-RMVSNet		79.61 133	78.44 139	83.14 123	89.38 96	65.93 136	84.95 202	87.15 203	73.56 108	78.19 115	89.79 99	56.67 182	93.36 133	59.53 226	86.74 113	90.13 149

test_djsdf		80.30 118	79.32 117	83.27 118	83.98 219	65.37 149	90.50 40	90.38 112	68.55 195	76.19 152	88.70 121	56.44 183	93.46 128	78.98 61	80.14 187	90.97 114

pcd1.5k->3k		34.07 311	35.26 311	30.50 326	86.92 167	0.00 346	0.00 337	91.58 81	0.00 341	0.00 342	0.00 343	56.23 184	0.00 344	0.00 341	82.60 160	91.49 103

EPNet_dtu		75.46 215	74.86 204	77.23 251	82.57 254	54.60 285	86.89 149	83.09 245	71.64 145	66.25 273	85.86 205	55.99 185	88.04 255	54.92 256	86.55 116	89.05 186
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
V4		77.95 168	76.37 175	82.67 151	79.40 290	65.52 142	86.43 163	89.94 133	72.28 137	72.14 209	84.95 224	55.72 186	93.44 130	73.64 112	72.86 262	89.05 186

v52		77.94 170	76.37 175	82.67 151	79.39 291	65.52 142	86.43 163	89.94 133	72.28 137	72.15 208	84.94 225	55.70 187	93.44 130	73.64 112	72.84 263	89.06 182

CostFormer		75.24 218	73.90 216	79.27 217	82.65 253	58.27 243	80.80 248	82.73 249	61.57 258	75.33 174	83.13 242	55.52 188	91.07 213	64.98 185	78.34 201	88.45 213

tpmrst		72.39 240	72.13 227	73.18 279	80.54 278	49.91 308	79.91 258	79.08 285	63.11 242	71.69 213	79.95 278	55.32 189	82.77 285	65.66 180	73.89 255	86.87 246

1314		76.53 196	75.30 198	80.21 203	83.93 220	62.32 215	84.66 206	88.81 174	60.23 267	70.16 229	84.07 233	55.30 190	90.73 218	67.37 164	83.21 152	87.59 230

GBi-Net		78.40 155	77.40 159	81.40 183	87.60 153	63.01 206	88.39 96	89.28 150	71.63 146	75.34 171	87.28 159	54.80 191	91.11 207	62.72 196	79.57 189	90.09 152

test1		78.40 155	77.40 159	81.40 183	87.60 153	63.01 206	88.39 96	89.28 150	71.63 146	75.34 171	87.28 159	54.80 191	91.11 207	62.72 196	79.57 189	90.09 152

FMVSNet2		78.20 160	77.21 162	81.20 186	87.60 153	62.89 210	87.47 126	89.02 159	71.63 146	75.29 175	87.28 159	54.80 191	91.10 210	62.38 200	79.38 192	89.61 175

Fast-Effi-MVS+-dtu		78.02 165	76.49 172	82.62 154	83.16 240	66.96 124	86.94 147	87.45 201	72.45 134	71.49 216	84.17 231	54.79 194	91.58 198	67.61 161	80.31 184	89.30 179

MVSTER		79.01 147	77.88 150	82.38 157	83.07 241	64.80 163	84.08 224	88.95 168	69.01 189	78.69 96	87.17 166	54.70 195	92.43 165	74.69 104	80.57 180	89.89 167

OpenMVS	72.83 10	79.77 131	78.33 143	84.09 92	85.17 186	69.91 65	90.57 38	90.97 97	66.70 208	72.17 206	91.91 57	54.70 195	93.96 98	61.81 208	90.95 65	88.41 215

XVG-OURS		80.41 112	79.23 123	83.97 100	85.64 181	69.02 80	83.03 234	90.39 111	71.09 154	77.63 126	91.49 69	54.62 197	91.35 201	75.71 94	83.47 145	91.54 100

LPG-MVS_test		82.08 77	81.27 81	84.50 79	89.23 105	68.76 87	90.22 48	91.94 66	75.37 78	76.64 143	91.51 67	54.29 198	94.91 67	78.44 65	83.78 139	89.83 169

LGP-MVS_train				84.50 79	89.23 105	68.76 87		91.94 66	75.37 78	76.64 143	91.51 67	54.29 198	94.91 67	78.44 65	83.78 139	89.83 169

TR-MVS		77.44 186	76.18 181	81.20 186	88.24 136	63.24 201	84.61 209	86.40 211	67.55 204	77.81 122	86.48 192	54.10 200	93.15 142	57.75 242	82.72 159	87.20 239

FMVSNet3		77.88 171	76.85 167	80.97 191	86.84 169	62.36 213	86.52 162	88.77 175	71.13 152	75.34 171	86.66 181	54.07 201	91.10 210	62.72 196	79.57 189	89.45 177

DP-MVS		76.78 194	74.57 207	83.42 112	93.29 31	69.46 76	88.55 91	83.70 235	63.98 238	70.20 226	88.89 118	54.01 202	94.80 73	46.66 294	81.88 167	86.01 261

ACMP	74.13 6	81.51 90	80.57 90	84.36 84	89.42 94	68.69 94	89.97 51	91.50 86	74.46 90	75.04 181	90.41 88	53.82 203	94.54 77	77.56 74	82.91 155	89.86 168
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
PLC	70.83 11	78.05 164	76.37 175	83.08 126	91.88 57	67.80 109	88.19 104	89.46 146	64.33 234	69.87 236	88.38 132	53.66 204	93.58 123	58.86 231	82.73 158	87.86 224
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
CANet_DTU		80.61 107	79.87 100	82.83 144	85.60 182	63.17 205	87.36 128	88.65 179	76.37 63	75.88 157	88.44 131	53.51 205	93.07 147	73.30 118	89.74 78	92.25 83

ACMM	73.20 8	80.78 105	79.84 101	83.58 108	89.31 102	68.37 98	89.99 50	91.60 80	70.28 165	77.25 132	89.66 101	53.37 206	93.53 126	74.24 108	82.85 156	88.85 193
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
MVP-Stereo		76.12 207	74.46 211	81.13 189	85.37 185	69.79 67	84.42 216	87.95 192	65.03 227	67.46 261	85.33 217	53.28 207	91.73 185	58.01 240	83.27 151	81.85 296
Qingsong Yan: MVP-Stereo: A Parallel Multi-View Patchmatch Stereo Method with Dilation Matching for Photogrammetric Application.
anonymousdsp		78.60 153	77.15 163	82.98 132	80.51 279	67.08 120	87.24 135	89.53 143	65.66 222	75.16 177	87.19 165	52.52 208	92.25 172	77.17 79	79.34 193	89.61 175

CR-MVSNet		73.37 230	71.27 236	79.67 212	81.32 271	65.19 154	75.92 283	80.30 276	59.92 270	72.73 199	81.19 266	52.50 209	86.69 262	59.84 222	77.71 203	87.11 243

Patchmtry		70.74 250	69.16 249	75.49 263	80.72 275	54.07 289	74.94 292	80.30 276	58.34 280	70.01 231	81.19 266	52.50 209	86.54 264	53.37 263	71.09 274	85.87 263

pmmvs4		74.03 224	71.91 228	80.39 197	81.96 260	68.32 99	81.45 246	82.14 255	59.32 274	69.87 236	85.13 220	52.40 211	88.13 254	60.21 220	74.74 248	84.73 274

PatchFormer-LS_test		74.50 220	73.05 221	78.86 227	82.95 245	59.55 235	81.65 244	82.30 253	67.44 206	71.62 214	78.15 289	52.34 212	88.92 246	65.05 184	75.90 233	88.12 218

RPMNet		71.62 244	68.94 251	79.67 212	81.32 271	65.19 154	75.92 283	78.30 288	57.60 287	72.73 199	76.45 299	52.30 213	86.69 262	48.14 285	77.71 203	87.11 243

LFMVS		81.82 83	81.23 82	83.57 109	91.89 56	63.43 197	89.84 52	81.85 262	77.04 47	83.21 53	93.10 39	52.26 214	93.43 132	71.98 134	89.95 76	93.85 33

VDD-MVS		83.01 69	82.36 69	84.96 69	91.02 64	66.40 129	88.91 76	88.11 186	77.57 35	84.39 41	93.29 36	52.19 215	93.91 103	77.05 81	88.70 88	94.57 9

tfpn200view9		76.42 199	75.37 196	79.55 216	89.13 109	57.65 253	85.17 197	83.60 236	73.41 112	76.45 145	86.39 193	52.12 216	91.95 177	48.33 282	83.75 141	89.07 181

thres400		76.50 197	75.37 196	79.86 207	89.13 109	57.65 253	85.17 197	83.60 236	73.41 112	76.45 145	86.39 193	52.12 216	91.95 177	48.33 282	83.75 141	90.00 158

thres200		75.55 214	74.47 210	78.82 228	87.78 150	57.85 250	83.07 233	83.51 239	72.44 136	75.84 158	84.42 230	52.08 218	91.75 184	47.41 288	83.64 144	86.86 247

PMMVS		69.34 261	68.67 252	71.35 287	75.67 306	62.03 218	75.17 287	73.46 312	50.00 316	68.68 249	79.05 283	52.07 219	78.13 301	61.16 214	82.77 157	73.90 317

tpm cat1		70.57 252	68.31 255	77.35 249	82.41 256	57.95 248	78.08 274	80.22 278	52.04 311	68.54 253	77.66 294	52.00 220	87.84 257	51.77 267	72.07 269	86.25 255

Patchmatch-test1		73.49 228	71.85 230	78.41 236	84.05 218	62.17 217	79.96 257	79.29 284	66.30 215	72.38 204	79.58 282	51.95 221	85.08 275	55.46 254	77.67 205	87.99 220

thres600view7		76.50 197	75.44 193	79.68 211	89.40 95	57.16 258	85.53 192	83.23 243	73.79 104	76.26 150	87.09 167	51.89 222	91.89 179	48.05 286	83.72 143	90.00 158

view600		76.20 203	75.21 199	79.16 221	89.64 85	55.82 277	85.74 182	82.06 257	73.88 99	75.74 160	87.85 143	51.84 223	91.66 192	46.75 290	83.42 146	90.00 158

view800		76.20 203	75.21 199	79.16 221	89.64 85	55.82 277	85.74 182	82.06 257	73.88 99	75.74 160	87.85 143	51.84 223	91.66 192	46.75 290	83.42 146	90.00 158

conf0.05thres1000		76.20 203	75.21 199	79.16 221	89.64 85	55.82 277	85.74 182	82.06 257	73.88 99	75.74 160	87.85 143	51.84 223	91.66 192	46.75 290	83.42 146	90.00 158

tfpn		76.20 203	75.21 199	79.16 221	89.64 85	55.82 277	85.74 182	82.06 257	73.88 99	75.74 160	87.85 143	51.84 223	91.66 192	46.75 290	83.42 146	90.00 158

tpm2		73.26 233	71.46 233	78.63 230	83.34 233	56.71 265	80.65 251	80.40 275	56.63 293	73.55 190	82.02 256	51.80 227	91.24 204	56.35 251	78.42 200	87.95 221

LS3D		76.95 192	74.82 205	83.37 115	90.45 70	67.36 117	89.15 72	86.94 205	61.87 257	69.52 239	90.61 86	51.71 228	94.53 78	46.38 297	86.71 114	88.21 217

IterMVS		74.29 221	72.94 222	78.35 237	81.53 265	63.49 194	81.58 245	82.49 250	68.06 200	69.99 233	83.69 238	51.66 229	85.54 271	65.85 178	71.64 271	86.01 261
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
tpm		72.37 242	71.71 232	74.35 272	82.19 258	52.00 295	79.22 264	77.29 293	64.56 231	72.95 197	83.68 239	51.35 230	83.26 284	58.33 237	75.80 234	87.81 225

sam_mvs1												51.32 231

PatchmatchNet		73.12 235	71.33 235	78.49 235	83.18 238	60.85 225	79.63 259	78.57 286	64.13 235	71.73 212	79.81 281	51.20 232	85.97 269	57.40 245	76.36 229	88.66 199
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
tpmp4_e23		73.45 229	71.17 238	80.31 201	83.55 229	59.56 234	81.88 239	82.33 252	57.94 284	70.51 223	81.62 264	51.19 233	91.63 196	53.96 260	77.51 206	89.75 173

patchmatchnet-post											74.00 306	51.12 234	88.60 249

xiu_mvs_v1_base_debu		80.80 102	79.72 104	84.03 96	87.35 158	70.19 61	85.56 187	88.77 175	69.06 185	81.83 67	88.16 137	50.91 235	92.85 154	78.29 69	87.56 103	89.06 182

xiu_mvs_v1_base		80.80 102	79.72 104	84.03 96	87.35 158	70.19 61	85.56 187	88.77 175	69.06 185	81.83 67	88.16 137	50.91 235	92.85 154	78.29 69	87.56 103	89.06 182

xiu_mvs_v1_base_debi		80.80 102	79.72 104	84.03 96	87.35 158	70.19 61	85.56 187	88.77 175	69.06 185	81.83 67	88.16 137	50.91 235	92.85 154	78.29 69	87.56 103	89.06 182

Patchmatch-test		64.82 281	63.24 279	69.57 293	79.42 289	49.82 309	63.49 322	69.05 325	51.98 312	59.95 298	80.13 277	50.91 235	70.98 325	40.66 310	73.57 258	87.90 223

semantic-postprocess				80.11 204	82.69 252	64.85 162		83.47 240	69.16 183	70.49 224	84.15 232	50.83 239	88.15 253	69.23 152	72.14 268	87.34 235

Patchmatch-RL test		70.24 256	67.78 265	77.61 245	77.43 298	59.57 233	71.16 298	70.33 318	62.94 246	68.65 250	72.77 309	50.62 240	85.49 272	69.58 150	66.58 292	87.77 226

VDDNet		81.52 88	80.67 89	84.05 94	90.44 71	64.13 183	89.73 58	85.91 218	71.11 153	83.18 54	93.48 31	50.54 241	93.49 127	73.40 117	88.25 97	94.54 10

pmmvs6		74.69 219	73.39 218	78.61 231	81.38 268	57.48 256	86.64 158	87.95 192	64.99 228	70.18 227	86.61 185	50.43 242	89.52 232	62.12 204	70.18 278	88.83 194

test_post											5.46 338	50.36 243	84.24 278

sam_mvs												50.01 244

PatchT		68.46 266	67.85 262	70.29 291	80.70 276	43.93 318	72.47 296	74.88 303	60.15 268	70.55 221	76.57 298	49.94 245	81.59 288	50.58 271	74.83 247	85.34 266

tpmvs		71.09 248	69.29 248	76.49 255	82.04 259	56.04 275	78.92 267	81.37 267	64.05 236	67.18 265	78.28 288	49.74 246	89.77 227	49.67 278	72.37 265	83.67 281

CVMVSNet		72.99 237	72.58 224	74.25 273	84.28 199	50.85 304	86.41 165	83.45 241	44.56 320	73.23 194	87.54 154	49.38 247	85.70 270	65.90 177	78.44 199	86.19 256

MDTV_nov1_ep13_2view						37.79 328	75.16 288		55.10 298	66.53 270		49.34 248		53.98 259		87.94 222

UGNet		80.83 99	79.59 107	84.54 78	88.04 140	68.09 104	89.42 63	88.16 185	76.95 48	76.22 151	89.46 108	49.30 249	93.94 100	68.48 157	90.31 70	91.60 98
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
pmmvs5		71.55 245	70.20 245	75.61 261	77.83 296	56.39 270	81.74 242	80.89 268	57.76 285	67.46 261	84.49 229	49.26 250	85.32 274	57.08 248	75.29 243	85.11 270

LTVRE_ROB	69.57 13	76.25 202	74.54 209	81.41 182	88.60 125	64.38 180	79.24 263	89.12 157	70.76 159	69.79 238	87.86 142	49.09 251	93.20 139	56.21 252	80.16 185	86.65 252
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
FMVSNet1		77.44 186	76.12 182	81.40 183	86.81 170	63.01 206	88.39 96	89.28 150	70.49 163	74.39 187	87.28 159	49.06 252	91.11 207	60.91 215	78.52 197	90.09 152

MDTV_nov1_ep13			69.97 246		83.18 238	53.48 292	77.10 279	80.18 279	60.45 264	69.33 243	80.44 274	48.89 253	86.90 261	51.60 268	78.51 198

test_post1							78.90 268				5.43 339	48.81 254	85.44 273	59.25 228

test-LLR		72.94 238	72.43 225	74.48 270	81.35 269	58.04 245	78.38 270	77.46 291	66.66 209	69.95 234	79.00 285	48.06 255	79.24 296	66.13 173	84.83 129	86.15 257

test0.0.03 1		68.00 267	67.69 266	68.90 296	77.55 297	47.43 312	75.70 286	72.95 314	66.66 209	66.56 269	82.29 250	48.06 255	75.87 311	44.97 303	74.51 250	83.41 283

MS-PatchMatch		73.83 225	72.67 223	77.30 250	83.87 221	66.02 134	81.82 240	84.66 227	61.37 261	68.61 252	82.82 245	47.29 257	88.21 252	59.27 227	84.32 136	77.68 309

cascas		76.72 195	74.64 206	82.99 131	85.78 179	65.88 138	82.33 237	89.21 155	60.85 263	72.74 198	81.02 270	47.28 258	93.75 116	67.48 163	85.02 127	89.34 178

test20.03		67.45 269	66.95 268	68.94 295	75.48 309	44.84 316	77.50 276	77.67 290	66.66 209	63.01 289	83.80 235	47.02 259	78.40 300	42.53 307	68.86 284	83.58 282

test_0402		72.79 239	70.44 242	79.84 208	88.13 138	65.99 135	85.93 177	84.29 230	65.57 223	67.40 263	85.49 214	46.92 260	92.61 161	35.88 316	74.38 251	80.94 299

F-COLMAP		76.38 201	74.33 212	82.50 155	89.28 103	66.95 125	88.41 95	89.03 158	64.05 236	66.83 267	88.61 125	46.78 261	92.89 153	57.48 243	78.55 196	87.67 227

Anonymous20231206		68.60 263	67.80 264	71.02 289	80.23 282	50.75 305	78.30 273	80.47 273	56.79 292	66.11 274	82.63 247	46.35 262	78.95 298	43.62 305	75.70 235	83.36 284

CHOSEN 280x420		66.51 275	64.71 274	71.90 282	81.45 266	63.52 193	57.98 327	68.95 326	53.57 306	62.59 291	76.70 297	46.22 263	75.29 314	55.25 255	79.68 188	76.88 315

GA-MVS		76.87 193	75.17 203	81.97 164	82.75 249	62.58 212	81.44 247	86.35 213	72.16 142	74.74 184	82.89 243	46.20 264	92.02 176	68.85 155	81.09 172	91.30 107

MDA-MVSNet_test_wron		65.03 279	62.92 280	71.37 285	75.93 304	56.73 263	69.09 310	74.73 306	57.28 290	54.03 314	77.89 291	45.88 265	74.39 317	49.89 277	61.55 307	82.99 290

YYNet1		65.03 279	62.91 281	71.38 284	75.85 305	56.60 267	69.12 309	74.66 309	57.28 290	54.12 313	77.87 292	45.85 266	74.48 316	49.95 276	61.52 308	83.05 288

EPMVS		69.02 262	68.16 257	71.59 283	79.61 287	49.80 310	77.40 277	66.93 328	62.82 248	70.01 231	79.05 283	45.79 267	77.86 304	56.58 250	75.26 244	87.13 242

IB-MVS	68.01 15	75.85 211	73.36 219	83.31 116	84.76 193	66.03 133	83.38 231	85.06 224	70.21 167	69.40 240	81.05 269	45.76 268	94.66 76	65.10 183	75.49 239	89.25 180
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
jajsoiax		79.29 142	77.96 148	83.27 118	84.68 195	66.57 128	89.25 67	90.16 125	69.20 182	75.46 166	89.49 105	45.75 269	93.13 144	76.84 84	80.80 176	90.11 150

PatchMatch-RL		72.38 241	70.90 240	76.80 254	88.60 125	67.38 116	79.53 260	76.17 297	62.75 249	69.36 242	82.00 257	45.51 270	84.89 276	53.62 262	80.58 179	78.12 307

RPSCF		73.23 234	71.46 233	78.54 233	82.50 255	59.85 231	82.18 238	82.84 248	58.96 276	71.15 218	89.41 112	45.48 271	84.77 277	58.82 232	71.83 270	91.02 113

MSDG		73.36 232	70.99 239	80.49 196	84.51 197	65.80 139	80.71 250	86.13 216	65.70 221	65.46 276	83.74 237	44.60 272	90.91 215	51.13 270	76.89 217	84.74 273

PVSNet_0	57.27 20	61.67 287	59.27 288	68.85 297	79.61 287	57.44 257	68.01 313	73.44 313	55.93 296	58.54 301	70.41 314	44.58 273	77.55 305	47.01 289	35.91 327	71.55 319

DWT-MVSNet_test		73.70 226	71.86 229	79.21 219	82.91 246	58.94 237	82.34 236	82.17 254	65.21 224	71.05 219	78.31 287	44.21 274	90.17 225	63.29 194	77.28 208	88.53 212

mvs_tets		79.13 145	77.77 154	83.22 120	84.70 194	66.37 130	89.17 68	90.19 124	69.38 178	75.40 169	89.46 108	44.17 275	93.15 142	76.78 85	80.70 178	90.14 148

MDA-MVSNet-bldmvs		66.68 273	63.66 277	75.75 259	79.28 292	60.56 228	73.92 294	78.35 287	64.43 232	50.13 321	79.87 280	44.02 276	83.67 280	46.10 298	56.86 315	83.03 289

gg-mvs-nofinetune		69.95 258	67.96 260	75.94 258	83.07 241	54.51 287	77.23 278	70.29 319	63.11 242	70.32 225	62.33 321	43.62 277	88.69 248	53.88 261	87.76 100	84.62 275

GG-mvs-BLEND				75.38 264	81.59 264	55.80 281	79.32 262	69.63 321		67.19 264	73.67 308	43.24 278	88.90 247	50.41 272	84.50 133	81.45 298

CMPMVS	51.72 21	70.19 257	68.16 257	76.28 256	73.15 315	57.55 255	79.47 261	83.92 233	48.02 318	56.48 310	84.81 226	43.13 279	86.42 266	62.67 199	81.81 168	84.89 271
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
Test4		77.83 172	75.90 189	83.62 106	80.24 281	65.25 152	85.27 196	90.67 102	69.03 188	66.48 271	83.75 236	43.07 280	93.00 151	75.93 90	88.66 89	92.62 73

dp		66.80 272	65.43 272	70.90 290	79.74 286	48.82 311	75.12 290	74.77 305	59.61 272	64.08 285	77.23 295	42.89 281	80.72 291	48.86 280	66.58 292	83.16 286

PVSNet	64.34 18	72.08 243	70.87 241	75.69 260	86.21 176	56.44 269	74.37 293	80.73 271	62.06 256	70.17 228	82.23 251	42.86 282	83.31 283	54.77 257	84.45 135	87.32 236

pmmvs-eth3d		70.50 254	67.83 263	78.52 234	77.37 299	66.18 132	81.82 240	81.51 265	58.90 277	63.90 286	80.42 275	42.69 283	86.28 267	58.56 234	65.30 300	83.11 287

UnsupCasMVSNet_eth		67.33 270	65.99 271	71.37 285	73.48 312	51.47 300	75.16 288	85.19 223	65.20 225	60.78 294	80.93 273	42.35 284	77.20 306	57.12 247	53.69 320	85.44 265

ADS-MVSNet2		66.20 278	63.33 278	74.82 268	79.92 284	58.75 238	67.55 315	75.19 301	53.37 307	65.25 278	75.86 300	42.32 285	80.53 292	41.57 308	68.91 282	85.18 267

ADS-MVSNet		64.36 283	62.88 282	68.78 298	79.92 284	47.17 313	67.55 315	71.18 317	53.37 307	65.25 278	75.86 300	42.32 285	73.99 319	41.57 308	68.91 282	85.18 267

SixPastTwentyTwo		73.37 230	71.26 237	79.70 210	85.08 191	57.89 249	85.57 186	83.56 238	71.03 155	65.66 275	85.88 204	42.10 287	92.57 162	59.11 229	63.34 303	88.65 200

JIA-IIPM		66.32 277	62.82 283	76.82 253	77.09 302	61.72 221	65.34 319	75.38 299	58.04 283	64.51 282	62.32 322	42.05 288	86.51 265	51.45 269	69.22 281	82.21 294

ACMH	67.68 16	75.89 210	73.93 215	81.77 168	88.71 123	66.61 127	88.62 88	89.01 161	69.81 170	66.78 268	86.70 179	41.95 289	91.51 199	55.64 253	78.14 202	87.17 240
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
ACMH+	68.96 14	76.01 209	74.01 214	82.03 163	88.60 125	65.31 151	88.86 78	87.55 198	70.25 166	67.75 258	87.47 156	41.27 290	93.19 140	58.37 236	75.94 232	87.60 229

MIMVSNet		70.69 251	69.30 247	74.88 267	84.52 196	56.35 272	75.87 285	79.42 283	64.59 230	67.76 257	82.41 248	41.10 291	81.54 289	46.64 296	81.34 170	86.75 250

N_pmnet		52.79 300	53.26 298	51.40 319	78.99 294	7.68 343	69.52 305	3.89 344	51.63 314	57.01 308	74.98 303	40.83 292	65.96 332	37.78 314	64.67 301	80.56 302

testing_2		75.73 212	73.34 220	82.89 140	77.37 299	65.22 153	84.10 223	90.54 108	69.09 184	60.46 295	81.15 268	40.48 293	92.84 157	76.36 86	80.54 182	90.60 129

EU-MVSNet		68.53 265	67.61 267	71.31 288	78.51 295	47.01 314	84.47 211	84.27 231	42.27 321	66.44 272	84.79 227	40.44 294	83.76 279	58.76 233	68.54 286	83.17 285

DSMNet-mixed		57.77 294	56.90 294	60.38 310	67.70 325	35.61 329	69.18 307	53.97 335	32.30 331	57.49 306	79.88 279	40.39 295	68.57 329	38.78 312	72.37 265	76.97 312

OurMVSNet-221017-0		74.26 222	72.42 226	79.80 209	83.76 226	59.59 232	85.92 178	86.64 207	66.39 214	66.96 266	87.58 151	39.46 296	91.60 197	65.76 179	69.27 280	88.22 216

K. test v3		71.19 247	68.51 253	79.21 219	83.04 243	57.78 252	84.35 218	76.91 295	72.90 126	62.99 290	82.86 244	39.27 297	91.09 212	61.65 209	52.66 321	88.75 197

lessismore_v0				78.97 225	81.01 274	57.15 259		65.99 329		61.16 293	82.82 245	39.12 298	91.34 202	59.67 223	46.92 325	88.43 214

UnsupCasMVSNet_bld		63.70 285	61.53 287	70.21 292	73.69 311	51.39 301	72.82 295	81.89 261	55.63 297	57.81 304	71.80 311	38.67 299	78.61 299	49.26 279	52.21 322	80.63 300

new-patchmatchnet		61.73 286	61.73 286	61.70 309	72.74 316	24.50 339	69.16 308	78.03 289	61.40 259	56.72 309	75.53 302	38.42 300	76.48 309	45.95 299	57.67 313	84.13 279

MVS-HIRNet		59.14 290	57.67 292	63.57 306	81.65 263	43.50 319	71.73 297	65.06 331	39.59 325	51.43 319	57.73 325	38.34 301	82.58 286	39.53 311	73.95 254	64.62 324

COLMAP_ROB	66.92 17	73.01 236	70.41 243	80.81 193	87.13 165	65.63 141	88.30 100	84.19 232	62.96 245	63.80 287	87.69 149	38.04 302	92.56 163	46.66 294	74.91 246	84.24 277
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
TESTMET0.1,1		69.89 259	69.00 250	72.55 280	79.27 293	56.85 261	78.38 270	74.71 307	57.64 286	68.09 256	77.19 296	37.75 303	76.70 307	63.92 190	84.09 137	84.10 280

LP		61.36 288	57.78 291	72.09 281	75.54 308	58.53 240	67.16 317	75.22 300	51.90 313	54.13 312	69.97 315	37.73 304	80.45 293	32.74 320	55.63 317	77.29 311

OpenMVS_ROB	64.09 19	70.56 253	68.19 256	77.65 244	80.26 280	59.41 236	85.01 201	82.96 247	58.76 278	65.43 277	82.33 249	37.63 305	91.23 205	45.34 302	76.03 231	82.32 293

FMVSNet5		69.50 260	67.96 260	74.15 274	82.97 244	55.35 283	80.01 256	82.12 256	62.56 251	63.02 288	81.53 265	36.92 306	81.92 287	48.42 281	74.06 253	85.17 269

MIMVSNet1		68.58 264	66.78 269	73.98 275	80.07 283	51.82 296	80.77 249	84.37 229	64.40 233	59.75 299	82.16 252	36.47 307	83.63 281	42.73 306	70.33 277	86.48 254

ITE_SJBPF				78.22 238	81.77 262	60.57 227		83.30 242	69.25 181	67.54 260	87.20 164	36.33 308	87.28 260	54.34 258	74.62 249	86.80 248

test-mter		71.41 246	70.39 244	74.48 270	81.35 269	58.04 245	78.38 270	77.46 291	60.32 266	69.95 234	79.00 285	36.08 309	79.24 296	66.13 173	84.83 129	86.15 257

testgi		66.67 274	66.53 270	67.08 301	75.62 307	41.69 323	75.93 282	76.50 296	66.11 216	65.20 280	86.59 186	35.72 310	74.71 315	43.71 304	73.38 260	84.84 272

EG-PatchMatch MVS		74.04 223	71.82 231	80.71 195	84.92 192	67.42 114	85.86 179	88.08 190	66.04 218	64.22 284	83.85 234	35.10 311	92.56 163	57.44 244	80.83 175	82.16 295

XVG-ACMP-BASELINE		76.11 208	74.27 213	81.62 177	83.20 237	64.67 165	83.60 229	89.75 138	69.75 172	71.85 211	87.09 167	32.78 312	92.11 175	69.99 147	80.43 183	88.09 219

AllTest		70.96 249	68.09 259	79.58 214	85.15 187	63.62 190	84.58 210	79.83 280	62.31 253	60.32 296	86.73 172	32.02 313	88.96 244	50.28 273	71.57 272	86.15 257

TestCases				79.58 214	85.15 187	63.62 190		79.83 280	62.31 253	60.32 296	86.73 172	32.02 313	88.96 244	50.28 273	71.57 272	86.15 257

USDC		70.33 255	68.37 254	76.21 257	80.60 277	56.23 273	79.19 265	86.49 209	60.89 262	61.29 292	85.47 215	31.78 315	89.47 234	53.37 263	76.21 230	82.94 292

tmp_tt		18.61 316	21.40 317	10.23 329	4.82 342	10.11 342	34.70 334	30.74 342	1.48 338	23.91 333	26.07 336	28.42 316	13.41 341	27.12 329	15.35 337	7.17 335

testpf		56.51 296	57.58 293	53.30 316	71.99 318	41.19 324	46.89 332	69.32 324	58.06 282	52.87 318	69.45 317	27.99 317	72.73 321	59.59 225	62.07 305	45.98 329

Anonymous20231211		64.82 281	61.79 285	73.91 276	77.11 301	50.92 303	85.29 195	81.53 264	54.19 301	57.98 303	78.03 290	26.90 318	87.83 258	37.92 313	57.12 314	82.99 290

test2356		59.50 289	58.08 289	63.74 305	71.23 319	41.88 321	67.59 314	72.42 316	53.72 305	57.65 305	70.74 313	26.31 319	72.40 322	32.03 323	71.06 275	76.93 313

TDRefinement		67.49 268	64.34 275	76.92 252	73.47 313	61.07 222	84.86 204	82.98 246	59.77 271	58.30 302	85.13 220	26.06 320	87.89 256	47.92 287	60.59 311	81.81 297

test1235678		58.74 292	56.89 295	64.30 303	69.70 321	41.87 322	71.05 299	74.87 304	54.06 302	50.63 320	71.53 312	25.30 321	74.10 318	31.80 324	63.10 304	76.93 313

TinyColmap		67.30 271	64.81 273	74.76 269	81.92 261	56.68 266	80.29 254	81.49 266	60.33 265	56.27 311	83.22 241	24.77 322	87.66 259	45.52 300	69.47 279	79.95 303

LF4IMVS		64.02 284	62.19 284	69.50 294	70.90 320	53.29 293	76.13 280	77.18 294	52.65 310	58.59 300	80.98 271	23.55 323	76.52 308	53.06 265	66.66 291	78.68 306

1111		57.11 295	56.82 296	57.97 313	69.10 322	28.28 334	68.90 311	74.54 310	54.01 303	53.71 315	74.51 304	23.09 324	67.90 330	32.28 321	61.26 309	77.73 308

.test1245		45.55 305	50.02 302	32.14 325	69.10 322	28.28 334	68.90 311	74.54 310	54.01 303	53.71 315	74.51 304	23.09 324	67.90 330	32.28 321	0.02 339	0.25 338

new_pmnet		50.91 302	50.29 301	52.78 317	68.58 324	34.94 332	63.71 321	56.63 334	39.73 324	44.95 322	65.47 320	21.93 326	58.48 334	34.98 317	56.62 316	64.92 323

pmmvs3		57.79 293	54.26 297	68.37 299	64.02 327	56.72 264	75.12 290	65.17 330	40.20 323	52.93 317	69.86 316	20.36 327	75.48 313	45.45 301	55.25 319	72.90 318

PM-MVS		66.41 276	64.14 276	73.20 278	73.92 310	56.45 268	78.97 266	64.96 332	63.88 240	64.72 281	80.24 276	19.84 328	83.44 282	66.24 172	64.52 302	79.71 304

no-one		51.08 301	45.79 306	66.95 302	57.92 332	50.49 307	59.63 326	76.04 298	48.04 317	31.85 327	56.10 328	19.12 329	80.08 295	36.89 315	26.52 329	70.29 320

testus		59.00 291	57.91 290	62.25 308	72.25 317	39.09 326	69.74 303	75.02 302	53.04 309	57.21 307	73.72 307	18.76 330	70.33 326	32.86 319	68.57 285	77.35 310

ambc				75.24 265	73.16 314	50.51 306	63.05 323	87.47 200		64.28 283	77.81 293	17.80 331	89.73 229	57.88 241	60.64 310	85.49 264

ANet_high		50.57 303	46.10 305	63.99 304	48.67 337	39.13 325	70.99 301	80.85 269	61.39 260	31.18 329	57.70 326	17.02 332	73.65 320	31.22 325	15.89 336	79.18 305

FPMVS		53.68 299	51.64 299	59.81 311	65.08 326	51.03 302	69.48 306	69.58 322	41.46 322	40.67 324	72.32 310	16.46 333	70.00 327	24.24 331	65.42 299	58.40 326

testmv		53.85 298	51.03 300	62.31 307	61.46 329	38.88 327	70.95 302	74.69 308	51.11 315	41.26 323	66.85 318	14.28 334	72.13 323	29.19 326	49.51 324	75.93 316

test12356		49.28 304	48.51 304	51.59 318	62.06 328	19.11 340	60.40 324	72.45 315	47.60 319	40.64 325	65.68 319	13.84 335	68.72 328	27.29 328	46.67 326	66.94 322

EMVS		30.81 313	29.65 314	34.27 324	50.96 336	25.95 338	56.58 330	46.80 339	24.01 334	15.53 337	30.68 335	12.47 336	54.43 337	12.81 337	17.05 335	22.43 334

E-PMN		31.77 312	30.64 313	35.15 323	52.87 335	27.67 336	57.09 329	47.86 338	24.64 332	16.40 336	33.05 334	11.23 337	54.90 336	14.46 336	18.15 333	22.87 333

DeepMVS_CX				27.40 327	40.17 341	26.90 337		24.59 343	17.44 336	23.95 332	48.61 329	9.77 338	26.48 339	18.06 333	24.47 330	28.83 332

Gipuma		45.18 306	41.86 307	55.16 315	77.03 303	51.52 299	32.50 335	80.52 272	32.46 329	27.12 330	35.02 332	9.52 339	75.50 312	22.31 332	60.21 312	38.45 331
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
LCM-MVSNet		54.25 297	49.68 303	67.97 300	53.73 334	45.28 315	66.85 318	80.78 270	35.96 327	39.45 326	62.23 323	8.70 340	78.06 303	48.24 284	51.20 323	80.57 301

PMMVS2		40.82 308	38.86 309	46.69 321	53.84 333	16.45 341	48.61 331	49.92 337	37.49 326	31.67 328	60.97 324	8.14 341	56.42 335	28.42 327	30.72 328	67.19 321

PMVS	37.38 22	44.16 307	40.28 308	55.82 314	40.82 340	42.54 320	65.12 320	63.99 333	34.43 328	24.48 331	57.12 327	3.92 342	76.17 310	17.10 334	55.52 318	48.75 327
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
MVE	26.22 23	30.37 314	25.89 316	43.81 322	44.55 339	35.46 331	28.87 336	39.07 340	18.20 335	18.58 335	40.18 331	2.68 343	47.37 338	17.07 335	23.78 331	48.60 328
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
PNet_i23d		38.26 310	35.42 310	46.79 320	58.74 330	35.48 330	59.65 325	51.25 336	32.45 330	23.44 334	47.53 330	2.04 344	58.96 333	25.60 330	18.09 334	45.92 330

wuyk23d		16.82 317	15.94 318	19.46 328	58.74 330	31.45 333	39.22 333	3.74 345	6.84 337	6.04 339	2.70 340	1.27 345	24.29 340	10.54 338	14.40 338	2.63 336

wuykxyi23d		39.76 309	33.18 312	59.51 312	46.98 338	44.01 317	57.70 328	67.74 327	24.13 333	13.98 338	34.33 333	1.27 345	71.33 324	34.23 318	18.23 332	63.18 325

test123		6.12 319	8.11 320	0.14 330	0.06 344	0.09 344	71.05 299	0.03 347	0.04 340	0.25 341	1.30 342	0.05 347	0.03 343	0.21 340	0.01 341	0.29 337

testmvs		6.04 320	8.02 321	0.10 331	0.08 343	0.03 345	69.74 303	0.04 346	0.05 339	0.31 340	1.68 341	0.02 348	0.04 342	0.24 339	0.02 339	0.25 338

sosnet-low-res		0.00 322	0.00 323	0.00 332	0.00 345	0.00 346	0.00 337	0.00 348	0.00 341	0.00 342	0.00 343	0.00 349	0.00 344	0.00 341	0.00 342	0.00 340

sosnet		0.00 322	0.00 323	0.00 332	0.00 345	0.00 346	0.00 337	0.00 348	0.00 341	0.00 342	0.00 343	0.00 349	0.00 344	0.00 341	0.00 342	0.00 340

uncertanet		0.00 322	0.00 323	0.00 332	0.00 345	0.00 346	0.00 337	0.00 348	0.00 341	0.00 342	0.00 343	0.00 349	0.00 344	0.00 341	0.00 342	0.00 340

Regformer		0.00 322	0.00 323	0.00 332	0.00 345	0.00 346	0.00 337	0.00 348	0.00 341	0.00 342	0.00 343	0.00 349	0.00 344	0.00 341	0.00 342	0.00 340

ab-mvs-re		7.23 318	9.64 319	0.00 332	0.00 345	0.00 346	0.00 337	0.00 348	0.00 341	0.00 342	86.72 174	0.00 349	0.00 344	0.00 341	0.00 342	0.00 340

uanet		0.00 322	0.00 323	0.00 332	0.00 345	0.00 346	0.00 337	0.00 348	0.00 341	0.00 342	0.00 343	0.00 349	0.00 344	0.00 341	0.00 342	0.00 340

ESAPD								94.22 1

MTGPA								92.02 59

MTMP								32.83 341

gm-plane-assit					81.40 267	53.83 291			62.72 250		80.94 272		92.39 167	63.40 193

test9_res														84.90 17	95.70 12	92.87 68

agg_prior2														82.91 39	95.45 14	92.70 69

agg_prior					92.85 41	71.94 39		91.78 74		84.41 39			94.93 65

test_prior4						72.60 27	89.01 75

test_prior				86.33 46	92.61 46	69.59 71		92.97 30					95.48 43			93.91 30

旧先验2							86.56 161		58.10 281	87.04 13			88.98 242	74.07 109

新几何2							86.29 170

无先验							87.48 125	88.98 165	60.00 269				94.12 93	67.28 165		88.97 192

原ACMM2							86.86 150

testdata2													91.01 214	62.37 201

testdata1							84.14 222		75.71 71

plane_prior7					90.08 78	68.51 97

plane_prior5								92.44 44					95.38 50	78.71 63	86.32 119	91.33 105

plane_prior4											91.00 80

plane_prior3						68.60 95			78.44 30	78.92 94

plane_prior2							91.25 28		79.12 23

plane_prior1					89.90 82

plane_prior						68.71 91	90.38 44		77.62 34						86.16 121

n2								0.00 348

nn								0.00 348

door-mid								69.98 320

test11								92.23 51

door								69.44 323

HQP5-MVS						66.98 122

HQP-NCC					89.33 97		89.17 68		76.41 59	77.23 134

ACMP_Plane					89.33 97		89.17 68		76.41 59	77.23 134

BP-MVS														77.47 75

HQP4-MVS										77.24 133			95.11 58			91.03 111

HQP3-MVS								92.19 54							85.99 123

NP-MVS					89.62 89	68.32 99					90.24 90

ACMMP++_ref															81.95 166

ACMMP++															81.25 171

High-resolution multi-view benchmark