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	courty.	delive.	electro	facade	kicker	meadow	office	pipes	playgr.	relief	relief.	terrace	terrai.

conf0.01		96.20 64	97.19 90	95.05 59	93.11 83	99.83 52	97.67 63	89.06 67	98.62 117	91.38 48	99.19 41	89.09 121	99.28 41	97.48 94	96.10 113	99.76 73	100.00 1

conf0.002		96.51 57	97.75 77	95.07 57	93.11 83	99.83 52	97.67 63	89.10 66	98.62 117	91.47 46	99.39 33	91.68 110	99.28 41	97.49 93	97.24 86	99.76 73	100.00 1

thres100view900		95.86 70	96.62 102	94.97 60	93.10 85	99.83 52	97.76 62	89.15 65	98.62 117	90.69 57	99.00 47	84.86 132	99.30 39	97.57 90	96.48 103	99.81 47	100.00 1

ESAPD		99.25 6	99.69 18	98.74 8	99.62 6	99.94 17	99.79 2	96.87 9	99.93 24	96.33 14	99.59 23	100.00 1	99.84 8	99.88 8	98.50 53	100.00 1	100.00 1

Fast-Effi-MVS+-dtu		92.73 125	97.62 79	87.02 144	88.91 145	98.83 118	95.79 102	73.98 188	99.89 31	68.62 162	97.73 79	93.30 106	95.21 125	97.67 87	95.96 121	99.59 127	100.00 1

Effi-MVS+-dtu		93.13 117	97.13 91	88.47 134	88.86 147	99.19 100	96.79 89	79.08 157	99.64 66	70.01 157	97.51 80	89.38 117	96.53 113	97.60 88	96.55 102	99.57 130	100.00 1

MVS_0304		97.04 49	98.72 51	95.08 56	96.32 60	99.90 43	99.15 36	89.61 60	99.89 31	87.22 93	95.47 99	98.22 74	98.22 78	98.63 43	98.90 36	99.93 16	100.00 1

TSAR-MVS + MP.		98.99 20	99.61 26	98.27 19	97.88 49	99.92 35	99.71 19	96.80 14	99.96 16	95.58 20	98.71 61	100.00 1	99.68 19	99.91 5	98.78 44	99.99 6	100.00 1

Effi-MVS+		93.06 120	95.94 123	89.70 118	90.82 129	99.45 91	95.71 104	78.94 159	98.72 113	74.71 139	97.92 76	80.73 149	98.35 75	97.72 86	97.05 98	99.70 110	100.00 1

GG-mvs-BLEND		69.85 226	99.39 36	35.39 236	3.67 240	99.94 17	99.10 38	1.69 238	99.85 41	3.19 243	98.13 74	99.46 53	4.92 238	99.23 28	99.14 29	99.80 50	100.00 1

gm-plane-assit		84.93 204	91.61 155	77.14 214	84.14 201	91.29 218	66.18 229	69.70 203	85.22 225	47.95 227	78.58 200	89.24 119	94.90 128	98.82 35	98.12 61	99.99 6	100.00 1

train_agg		98.62 26	99.76 8	97.28 29	99.03 37	99.93 29	99.65 23	96.37 29	99.98 3	89.24 78	99.53 27	99.83 39	99.59 26	99.85 10	99.19 28	99.80 50	100.00 1

DI_MVS_plusplus_trai		95.29 88	97.02 94	93.28 88	91.76 121	99.52 85	97.84 60	85.67 94	99.08 104	87.29 91	87.76 149	97.46 80	97.31 98	97.83 85	97.48 82	99.83 34	100.00 1

SD-MVS		99.16 12	99.73 14	98.49 15	97.93 48	99.95 12	99.74 13	96.94 4	99.96 16	96.60 11	99.47 32	100.00 1	99.88 6	99.15 29	99.59 12	99.84 29	100.00 1

MSLP-MVS++		99.39 2	99.76 8	98.95 2	99.60 12	99.99 1	99.83 1	96.82 13	99.92 29	97.58 6	99.58 25	100.00 1	99.93 1	98.98 31	99.86 7	99.96 11	100.00 1

APDe-MVS		99.40 1	99.81 2	98.92 3	99.62 6	99.96 7	99.76 5	96.87 9	99.95 20	97.66 4	99.57 26	100.00 1	99.63 24	99.88 8	99.28 25	100.00 1	100.00 1

thres200		95.77 77	96.55 104	94.86 64	93.09 89	99.82 56	97.63 69	88.85 70	98.49 126	90.66 61	98.99 50	84.86 132	99.20 47	97.41 98	96.28 109	99.76 73	100.00 1

test0.0.03 1		95.15 92	97.87 76	91.99 102	91.69 123	98.82 119	93.04 139	83.60 116	99.65 64	88.80 81	94.15 116	97.67 78	94.97 126	96.62 114	98.16 59	99.83 34	100.00 1

MCST-MVS		99.08 17	99.72 16	98.33 18	99.59 14	99.97 3	99.78 3	96.96 2	99.95 20	93.72 31	99.67 11	100.00 1	99.90 4	99.91 5	98.55 51	100.00 1	100.00 1

MVS_Test		95.74 80	98.18 67	92.90 95	92.16 113	99.49 88	97.36 77	84.30 109	99.79 50	84.94 102	96.65 88	93.63 104	98.85 63	98.61 45	99.10 31	99.81 47	100.00 1

CDPH-MVS		97.88 40	99.59 27	95.89 46	98.90 38	99.95 12	99.40 30	92.86 39	99.86 40	85.33 99	98.62 63	99.45 55	99.06 57	99.29 26	99.94 3	99.81 47	100.00 1

UniMVSNet (Re)		90.41 144	91.96 152	88.59 133	85.71 162	96.73 154	90.82 156	84.11 114	95.23 167	78.54 126	88.91 144	76.41 160	92.84 150	93.40 176	93.05 164	99.55 134	100.00 1

TSAR-MVS + GP.		98.06 36	99.55 32	96.32 43	94.72 74	99.92 35	99.22 35	89.98 54	99.97 6	94.77 25	99.94 9	100.00 1	99.43 31	98.52 55	98.53 52	99.79 54	100.00 1

DWT-MVSNet_training		96.26 63	98.44 59	93.72 82	92.58 101	99.34 96	96.15 98	83.00 123	99.76 55	93.63 32	97.89 77	99.46 53	97.23 100	94.43 154	98.19 58	99.70 110	100.00 1

APD-MVS		99.33 5	99.85 1	98.73 9	99.61 10	99.92 35	99.77 4	96.91 6	99.93 24	96.31 15	99.59 23	99.95 33	99.84 8	99.73 16	99.84 8	99.95 13	100.00 1
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
CP-MVS		99.14 13	99.67 20	98.53 13	99.45 20	99.94 17	99.63 25	96.62 24	99.82 46	95.92 17	99.65 16	100.00 1	99.71 15	99.76 14	98.56 50	99.83 34	100.00 1

tpm cat1		93.29 116	96.53 108	89.50 121	91.84 118	99.18 101	94.70 123	67.70 205	98.38 131	86.67 94	89.16 141	99.38 58	96.66 111	94.33 155	95.30 133	99.43 165	100.00 1

SteuartSystems-ACMMP		98.95 22	99.80 3	97.95 24	99.43 23	99.96 7	99.76 5	96.45 28	99.82 46	93.63 32	99.64 19	100.00 1	98.56 72	99.90 7	99.31 23	99.84 29	100.00 1
Skip Steuart: Steuart Systems R&D Blog.
tpmp4_e23		92.95 121	96.28 116	89.06 125	91.80 119	98.81 120	94.95 119	67.56 208	99.21 96	82.97 112	96.54 89	88.52 124	97.47 95	94.47 153	96.42 106	99.61 125	100.00 1

CostFormer		93.50 111	96.50 109	90.00 116	91.69 123	98.65 124	93.88 133	67.64 206	98.97 106	89.16 79	97.79 78	88.92 122	97.97 85	95.14 145	96.06 118	99.63 122	100.00 1

CR-MVSNet		92.32 133	97.97 73	85.74 162	90.63 133	98.95 113	95.46 112	65.50 217	99.09 102	67.51 168	94.20 114	98.18 75	95.59 122	98.16 69	97.20 91	99.74 90	100.00 1

PatchT		91.06 141	97.66 78	83.36 198	90.32 135	98.96 112	82.30 214	64.72 222	98.45 129	67.51 168	93.28 125	97.60 79	95.59 122	98.16 69	97.20 91	99.70 110	100.00 1

tpmrst		92.52 130	97.45 83	86.77 148	92.15 114	99.36 95	92.53 144	65.95 215	99.53 73	72.50 142	92.22 131	99.83 39	97.81 90	95.18 144	96.05 119	99.69 116	100.00 1

DELS-MVS		97.05 47	98.05 70	95.88 48	97.09 54	99.99 1	98.82 45	90.30 51	98.44 130	91.40 47	92.91 126	96.57 84	97.68 94	98.56 47	99.88 5	100.00 1	100.00 1
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
RPMNet		92.64 128	97.88 75	86.53 150	90.79 130	98.95 113	95.13 117	64.44 223	99.09 102	72.36 143	93.58 122	99.01 61	96.74 110	98.05 75	96.45 105	99.71 108	100.00 1

PVSNet_Blended_VisFu		95.37 87	97.44 84	92.95 94	95.20 67	99.80 68	92.68 141	88.41 84	99.12 100	87.64 87	88.31 146	99.10 60	94.07 139	98.27 65	97.51 81	99.73 102	100.00 1

PVSNet_BlendedMVS		96.01 67	96.48 110	95.46 50	96.47 58	99.89 47	95.64 106	91.23 45	99.75 57	91.59 43	96.80 84	82.44 145	98.05 81	98.53 52	97.92 68	99.80 50	100.00 1

PVSNet_Blended		96.01 67	96.48 110	95.46 50	96.47 58	99.89 47	95.64 106	91.23 45	99.75 57	91.59 43	96.80 84	82.44 145	98.05 81	98.53 52	97.92 68	99.80 50	100.00 1

MDTV_nov1_ep13		94.32 101	98.77 49	89.14 124	91.70 122	99.52 85	95.21 116	72.09 201	99.80 49	78.91 124	96.32 91	99.62 48	97.71 93	98.39 59	97.71 77	99.22 203	100.00 1

IterMVS		91.65 137	96.62 102	85.85 159	90.27 136	95.80 178	95.32 115	74.15 184	98.91 111	60.95 206	88.79 145	97.76 77	94.69 132	98.04 77	97.07 96	99.73 102	100.00 1
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
HyFIR lowres test		93.13 117	94.48 137	91.56 104	96.12 63	99.68 76	93.52 136	79.98 148	97.24 149	81.73 118	72.66 215	95.74 93	98.29 77	98.27 65	97.79 71	99.70 110	100.00 1

IB-MVS	90.59 15	92.70 126	95.70 125	89.21 123	94.62 76	99.45 91	83.77 209	88.92 69	99.53 73	92.82 36	98.86 55	86.08 129	75.24 221	92.81 185	93.17 161	99.89 22	100.00 1
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
IS_MVSNet		96.66 55	98.62 53	94.38 75	92.41 109	99.70 75	97.19 80	87.67 89	99.05 105	91.27 51	95.09 104	98.46 69	97.95 86	98.64 41	99.37 18	99.79 54	100.00 1

AdaColmap		99.21 9	99.45 35	98.92 3	99.67 4	99.95 12	99.65 23	96.77 17	99.97 6	97.67 3	100.00 1	99.69 46	99.93 1	99.26 27	97.25 85	99.85 27	100.00 1

CLD-MVS		94.53 97	94.45 138	94.61 72	93.85 80	98.36 127	98.12 58	89.68 58	99.35 92	89.62 77	95.19 102	77.08 154	96.66 111	95.51 137	95.67 127	99.74 90	100.00 1
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
ACMH	92.34 14	91.59 138	93.02 147	89.92 117	93.97 79	97.98 137	90.10 179	84.70 103	98.46 128	76.80 131	93.38 124	71.94 179	94.39 134	95.34 141	94.04 150	99.54 135	100.00 1
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
tfpn111		95.79 73	96.55 104	94.89 61	93.10 85	99.82 56	97.67 63	88.85 70	98.62 117	90.69 57	99.07 44	84.86 132	99.28 41	97.41 98	96.10 113	99.76 73	99.99 47

tfpn_ndepth		96.84 53	98.58 54	94.81 67	93.18 82	99.62 81	96.83 88	88.75 76	99.73 59	92.38 39	98.45 65	96.34 88	97.90 87	98.34 63	97.59 79	99.84 29	99.99 47

conf200view11		95.78 74	96.54 106	94.89 61	93.10 85	99.82 56	97.67 63	88.85 70	98.62 117	90.69 57	99.00 47	84.86 132	99.28 41	97.41 98	96.10 113	99.76 73	99.99 47

tfpn200view9		95.78 74	96.54 106	94.89 61	93.10 85	99.82 56	97.67 63	88.85 70	98.62 117	90.69 57	99.00 47	84.86 132	99.28 41	97.41 98	96.10 113	99.76 73	99.99 47

view600		95.64 82	96.38 113	94.79 68	92.96 93	99.82 56	97.48 74	88.85 70	98.38 131	90.52 64	98.84 57	84.61 138	99.15 51	97.41 98	95.60 130	99.76 73	99.99 47

CANet		97.62 41	98.94 42	96.08 44	97.19 53	99.93 29	99.29 33	90.38 50	99.87 37	91.00 54	95.79 97	99.51 52	98.72 70	98.53 52	99.00 34	99.90 21	99.99 47

test-LLR		93.71 107	97.23 88	89.60 119	91.69 123	99.10 103	94.68 125	83.60 116	99.36 90	71.94 147	93.82 119	96.51 85	95.96 117	97.42 96	94.37 145	99.74 90	99.99 47

TESTMET0.1,1		92.87 123	97.23 88	87.79 140	86.96 153	99.10 103	94.68 125	77.46 168	99.36 90	71.94 147	93.82 119	96.51 85	95.96 117	97.42 96	94.37 145	99.74 90	99.99 47

test-mter		92.67 127	97.13 91	87.47 142	86.72 155	99.07 105	94.28 129	76.90 172	99.21 96	71.53 151	93.63 121	96.32 89	95.67 119	97.32 105	94.36 147	99.74 90	99.99 47

thres600view7		95.64 82	96.38 113	94.79 68	92.96 93	99.82 56	97.48 74	88.85 70	98.38 131	90.52 64	98.84 57	84.61 138	99.15 51	97.41 98	95.60 130	99.76 73	99.99 47

thres400		95.72 81	96.48 110	94.84 65	93.00 92	99.83 52	97.55 70	88.93 68	98.49 126	90.61 62	98.86 55	84.63 137	99.20 47	97.45 95	96.10 113	99.77 66	99.99 47

diffmvs		96.35 61	98.76 50	93.54 85	92.41 109	99.55 84	97.22 79	83.75 115	99.57 70	89.64 76	96.86 83	98.33 70	98.37 74	98.42 57	98.61 48	99.88 25	99.99 47

abl_6				97.06 32	99.17 35	99.82 56	98.68 49	90.86 47	100.00 1	94.53 27	97.40 82	100.00 1	99.17 50			99.93 16	99.99 47

testpf		91.26 140	97.28 87	84.23 185	89.52 139	97.45 145	88.08 198	56.08 232	99.76 55	78.71 125	95.06 108	98.26 71	93.44 145	94.72 150	95.69 126	99.57 130	99.99 47

IterMVS-LS		93.50 111	96.22 118	90.33 114	90.93 128	95.50 188	94.83 121	80.54 144	98.92 110	79.11 121	90.64 137	93.70 103	96.79 107	96.93 109	97.85 70	99.78 61	99.99 47
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
HQP-MVS		94.48 99	95.39 129	93.42 86	95.10 68	98.35 128	98.19 57	91.41 43	99.77 53	79.79 119	99.30 38	77.08 154	96.25 114	96.93 109	96.28 109	99.76 73	99.99 47

EPP-MVSNet		96.29 62	98.34 62	93.90 80	91.77 120	99.38 94	95.45 114	87.25 92	99.38 89	91.36 49	94.86 111	98.49 67	97.83 89	98.01 80	98.23 57	99.75 86	99.99 47

PHI-MVS		98.85 23	99.67 20	97.89 25	98.63 44	99.93 29	98.95 42	95.20 34	99.84 44	94.94 23	99.74 10	100.00 1	99.69 17	98.40 58	99.75 10	99.93 16	99.99 47

PLC	98.06 1	99.17 11	99.38 37	98.92 3	99.47 18	99.90 43	99.48 28	96.47 27	99.96 16	98.73 1	99.52 29	100.00 1	99.55 29	98.54 51	97.73 76	99.84 29	99.99 47
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
ACMP	94.49 9	94.19 105	94.74 134	93.56 84	94.25 78	98.32 130	96.02 101	89.35 63	98.90 112	87.28 92	99.14 43	76.41 160	94.94 127	96.07 131	94.35 148	99.49 146	99.99 47
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
view800		95.62 84	96.38 113	94.73 70	92.96 93	99.81 64	97.38 76	88.75 76	98.35 136	90.43 67	98.81 59	84.54 140	99.13 54	97.35 104	95.82 125	99.76 73	99.98 67

tfpn		95.93 69	97.06 93	94.62 71	92.94 97	99.81 64	97.25 78	88.71 79	98.32 137	89.98 70	98.79 60	88.55 123	99.11 55	97.26 108	96.71 101	99.75 86	99.98 67

FC-MVSNet-train		94.61 96	96.27 117	92.68 99	92.35 112	97.14 149	93.45 138	87.73 88	98.93 109	87.31 90	96.42 90	89.35 118	95.67 119	96.06 132	96.01 120	99.56 132	99.98 67

TSAR-MVS + ACMM		98.30 30	99.64 22	96.74 36	99.08 36	99.94 17	99.67 22	96.73 19	99.97 6	86.30 96	98.30 67	99.99 26	98.78 66	99.73 16	99.57 14	99.88 25	99.98 67

CHOSEN 1792x2688		93.69 108	94.89 133	92.28 100	96.17 61	99.84 51	95.69 105	83.17 121	98.54 124	82.04 115	77.58 204	91.15 112	96.90 104	98.36 61	98.82 41	99.73 102	99.98 67

tpm		89.60 149	94.93 132	83.39 196	89.94 137	97.11 150	90.09 180	65.28 220	98.67 115	60.03 210	96.79 86	84.38 142	95.66 121	91.90 190	95.65 129	99.32 193	99.98 67

MVSTER		97.00 51	98.85 46	94.83 66	92.71 99	97.43 146	99.03 40	85.52 95	99.82 46	92.74 37	99.15 42	99.94 34	99.19 49	98.66 39	96.99 99	99.79 54	99.98 67

CPTT-MVS		99.08 17	99.53 33	98.57 12	99.44 22	99.93 29	99.60 26	95.92 32	99.77 53	97.01 9	99.67 11	100.00 1	99.72 14	99.56 21	97.76 73	99.70 110	99.98 67

dps		94.29 103	97.33 85	90.75 108	92.02 116	99.21 99	94.31 128	66.97 209	99.50 75	95.61 19	96.22 93	98.64 63	96.08 115	93.71 169	94.03 151	99.52 139	99.98 67

QAPM		97.90 39	98.89 43	96.74 36	99.35 27	99.80 68	98.84 44	90.20 53	99.94 22	92.85 35	94.17 115	99.78 42	99.42 32	98.71 37	99.87 6	99.79 54	99.98 67

DeepC-MVS_fast	98.03 2	99.05 19	99.78 6	98.21 21	99.47 18	99.97 3	99.75 11	96.80 14	99.97 6	93.58 34	98.68 62	99.94 34	99.69 17	99.93 4	99.95 2	99.96 11	99.98 67
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
3Dnovator	95.01 8	97.98 37	98.89 43	96.92 35	99.36 26	99.76 72	98.72 48	89.98 54	99.98 3	93.99 29	94.60 112	99.43 56	99.50 30	98.55 48	99.91 4	99.99 6	99.98 67

OpenMVS	94.03 11	96.87 52	98.10 69	95.44 52	99.29 30	99.78 70	98.46 56	89.92 57	99.47 81	85.78 97	91.05 136	98.50 65	99.30 39	98.49 56	99.41 17	99.89 22	99.98 67

CHOSEN 280x420		97.16 46	99.58 29	94.35 78	96.95 56	99.97 3	97.19 80	81.55 140	99.92 29	91.75 42	100.00 1	100.00 1	98.84 64	98.55 48	98.65 47	99.79 54	99.97 80

gg-mvs-nofinetune		86.69 192	91.30 158	81.30 206	90.42 134	99.64 79	98.50 54	61.68 228	79.23 229	40.35 232	66.58 223	97.14 81	96.92 103	98.64 41	97.94 64	99.91 20	99.97 80

HFP-MVS		99.19 10	99.77 7	98.51 14	99.55 16	99.94 17	99.76 5	96.84 12	99.88 34	95.27 22	99.67 11	100.00 1	99.85 7	99.56 21	99.36 20	99.79 54	99.97 80

X-MVS		98.62 26	99.75 11	97.29 28	99.50 17	99.94 17	99.71 19	96.55 25	99.85 41	88.58 83	99.65 16	99.98 28	99.67 20	99.60 20	99.26 26	99.77 66	99.97 80

MP-MVS		98.82 24	99.63 23	97.88 26	99.41 24	99.91 42	99.74 13	96.76 18	99.88 34	91.89 41	99.50 30	99.94 34	99.65 22	99.71 19	98.49 54	99.82 39	99.97 80

UniMVSNet_NR-MVSNet		90.50 143	92.31 150	88.38 135	85.04 176	96.34 162	90.94 153	85.32 96	95.87 160	75.69 132	87.68 150	78.49 150	93.78 142	93.21 178	94.60 138	99.53 138	99.97 80

DU-MVS		89.49 151	90.60 162	88.19 138	84.71 190	96.20 163	90.94 153	84.58 104	95.54 162	75.69 132	87.52 152	68.74 209	93.78 142	91.10 208	95.13 135	99.47 153	99.97 80

NR-MVSNet		89.52 150	90.71 161	88.14 139	86.19 160	96.20 163	92.07 146	84.58 104	95.54 162	75.27 136	87.52 152	67.96 210	91.24 183	94.33 155	93.45 158	99.49 146	99.97 80

LGP-MVS_train		93.60 109	95.05 130	91.90 103	94.90 71	98.29 131	97.93 59	88.06 85	99.14 99	74.83 137	99.26 39	76.50 158	96.07 116	96.31 124	95.90 124	99.59 127	99.97 80

EPNet		98.11 35	99.63 23	96.34 42	98.44 46	99.88 49	98.55 51	90.25 52	99.93 24	92.60 38	100.00 1	99.73 43	98.41 73	98.87 33	99.02 33	99.82 39	99.97 80
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
CNVR-MVS		99.39 2	99.75 11	98.98 1	99.69 1	99.95 12	99.76 5	96.91 6	99.98 3	97.59 5	99.64 19	100.00 1	99.93 1	99.94 2	98.75 46	99.97 10	99.97 80

NCCC		99.24 7	99.75 11	98.65 10	99.63 5	99.96 7	99.76 5	96.91 6	99.97 6	95.86 18	99.67 11	100.00 1	99.75 13	99.85 10	98.80 42	99.98 9	99.97 80

MVS_111021_HR		97.94 38	99.59 27	96.02 45	99.27 32	99.97 3	97.03 84	90.44 49	99.89 31	90.75 56	100.00 1	99.73 43	98.68 71	98.67 38	98.89 37	99.95 13	99.97 80

ACMMP		98.16 33	99.01 40	97.18 30	98.86 39	99.92 35	98.77 47	95.73 33	99.31 94	91.15 53	100.00 1	99.81 41	98.82 65	98.11 72	95.91 122	99.77 66	99.97 80
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
CNLPA		99.24 7	99.58 29	98.85 6	99.34 28	99.95 12	99.32 31	96.65 22	99.96 16	98.44 2	98.97 51	100.00 1	99.57 27	98.66 39	99.56 15	99.76 73	99.97 80

DeepPCF-MVS	97.16 4	97.58 43	99.72 16	95.07 57	98.45 45	99.96 7	93.83 134	95.93 31	100.00 1	90.79 55	98.38 66	99.85 38	95.28 124	99.94 2	99.97 1	96.15 220	99.97 80

DeepC-MVS	96.33 6	97.05 47	97.59 80	96.42 40	97.37 52	99.92 35	99.10 38	96.54 26	99.34 93	86.64 95	91.93 133	93.15 107	99.11 55	99.11 30	99.68 11	99.73 102	99.97 80
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
PCF-MVS	97.20 3	97.49 44	98.20 66	96.66 38	97.62 51	99.92 35	98.93 43	96.64 23	98.53 125	88.31 86	94.04 117	99.58 50	98.94 58	97.53 91	97.79 71	99.54 135	99.97 80
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
3Dnovator+	95.21 7	98.17 32	99.08 38	97.12 31	99.28 31	99.78 70	98.61 50	89.93 56	99.93 24	95.36 21	95.50 98	100.00 1	99.56 28	98.58 46	99.80 9	99.95 13	99.97 80

tfpn1000		96.58 56	98.37 60	94.50 74	93.04 90	99.59 82	96.53 93	88.54 80	99.73 59	91.59 43	98.28 69	95.76 92	97.46 96	98.19 68	97.10 94	99.82 39	99.96 99

HSP-MVS		99.36 4	99.79 4	98.85 6	99.61 10	99.96 7	99.71 19	96.94 4	99.97 6	97.11 8	99.60 22	100.00 1	99.70 16	99.96 1	99.12 30	100.00 1	99.96 99

Fast-Effi-MVS+		92.11 135	94.33 139	89.52 120	89.06 144	99.00 110	95.13 117	76.72 174	98.59 123	78.21 128	89.99 139	77.35 153	98.34 76	97.97 81	97.44 83	99.67 118	99.96 99

HPM-MVS++		98.98 21	99.62 25	98.22 20	99.62 6	99.94 17	99.74 13	96.95 3	99.87 37	93.76 30	99.49 31	100.00 1	99.39 35	99.73 16	98.35 55	99.89 22	99.96 99

ACMMPR		99.12 15	99.76 8	98.36 17	99.45 20	99.94 17	99.75 11	96.70 21	99.93 24	94.65 26	99.65 16	99.96 31	99.84 8	99.51 23	99.35 21	99.79 54	99.96 99

PS-CasMVS		87.24 180	88.52 197	85.73 163	84.58 194	95.35 192	89.03 193	80.17 146	93.11 207	68.86 161	77.71 203	66.89 213	92.30 154	93.13 181	93.50 157	99.46 156	99.96 99

CP-MVSNet		88.09 161	89.57 173	86.36 151	84.63 193	95.46 190	89.48 190	80.53 145	93.42 201	71.26 154	81.25 176	69.90 202	92.78 151	93.30 177	93.69 155	99.47 153	99.96 99

WR-MVS_H		88.47 155	90.55 163	86.04 153	85.13 173	96.07 171	89.86 188	79.80 150	94.37 183	72.32 145	83.12 162	74.44 168	89.60 193	93.52 173	92.40 171	99.51 143	99.96 99

PMMVS		96.45 59	98.24 63	94.36 77	92.58 101	99.01 109	97.08 82	87.42 90	99.88 34	90.06 69	99.39 33	94.63 96	99.33 38	97.85 84	96.99 99	99.70 110	99.96 99

SMA-MVS		99.14 13	99.79 4	98.39 16	99.68 2	99.94 17	99.74 13	96.86 11	99.97 6	94.36 28	99.22 40	100.00 1	99.89 5	99.84 12	99.58 13	99.83 34	99.95 108

thresconf0.02		96.46 58	98.87 45	93.64 83	92.77 98	99.11 102	97.05 83	89.36 61	99.64 66	85.14 100	99.07 44	96.84 82	97.72 91	98.72 36	98.76 45	99.78 61	99.95 108

PGM-MVS		98.47 29	99.73 14	97.00 33	99.68 2	99.94 17	99.76 5	91.74 41	99.84 44	91.17 52	100.00 1	99.69 46	99.81 11	99.38 25	99.30 24	99.82 39	99.95 108

MVS_111021_LR		98.15 34	99.69 18	96.36 41	99.23 33	99.93 29	97.79 61	91.84 40	99.87 37	90.53 63	100.00 1	99.57 51	98.93 59	99.44 24	99.08 32	99.85 27	99.95 108

LTVRE_ROB	88.65 16	87.87 166	91.11 159	84.10 188	86.64 157	97.47 144	94.40 127	78.41 163	96.13 158	52.02 221	87.95 147	65.92 217	93.59 144	95.29 142	95.09 136	99.52 139	99.95 108
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
zzz-MVS		99.12 15	99.52 34	98.65 10	99.58 15	99.93 29	99.74 13	96.72 20	99.44 83	96.47 12	99.62 21	100.00 1	99.63 24	99.74 15	97.97 63	99.77 66	99.94 113

v148		86.63 193	87.79 205	85.28 170	84.65 192	95.97 175	86.46 203	82.84 124	92.91 208	71.52 152	78.99 194	66.74 215	86.83 208	89.28 216	90.69 205	99.41 183	99.94 113

CSCG		98.22 31	98.37 60	98.04 22	99.60 12	99.82 56	99.45 29	93.59 37	99.16 98	96.46 13	98.22 73	95.86 91	99.41 34	96.33 123	99.22 27	99.75 86	99.94 113

tfpnnormal		89.09 153	89.71 169	88.38 135	87.37 152	96.78 153	91.46 152	85.20 99	90.33 216	72.35 144	83.45 160	69.30 207	94.45 133	95.29 142	92.86 166	99.44 164	99.93 116

v1144		87.49 172	89.64 170	84.97 172	84.73 189	95.84 177	90.17 178	79.30 154	93.96 187	64.65 194	78.83 197	73.38 174	91.51 171	93.77 167	91.77 180	99.45 159	99.93 116

v2v482		87.46 174	88.90 190	85.78 161	84.58 194	95.95 176	89.90 187	82.43 129	94.19 185	65.65 189	79.80 185	69.12 208	92.67 152	91.88 191	91.46 186	99.45 159	99.93 116

v1		87.48 173	88.91 189	85.81 160	84.93 180	96.07 171	90.33 170	82.45 128	93.65 196	66.39 180	79.38 192	70.40 198	91.33 178	91.58 194	91.38 194	99.42 171	99.93 116

conf0.05thres1000		94.50 98	95.70 125	93.11 90	92.68 100	99.67 77	96.04 99	87.81 86	97.52 148	83.71 105	96.20 94	84.52 141	98.73 69	96.39 121	95.66 128	99.71 108	99.92 120

canonicalmvs		95.80 72	97.02 94	94.37 76	92.96 93	99.47 89	97.49 71	84.58 104	99.44 83	92.05 40	98.54 64	86.65 127	99.37 36	96.18 126	98.93 35	99.77 66	99.92 120

v1141		87.45 176	88.98 185	85.67 165	84.86 184	96.08 169	90.23 176	82.46 126	93.75 191	65.64 191	79.57 189	70.52 196	91.41 176	91.63 192	91.39 192	99.42 171	99.92 120

v1neww		87.88 164	89.51 178	85.97 157	85.32 169	96.12 166	90.33 170	82.17 133	94.51 176	66.96 174	81.84 171	71.21 187	91.64 165	91.52 197	91.43 188	99.42 171	99.92 120

v7new		87.88 164	89.51 178	85.97 157	85.32 169	96.12 166	90.33 170	82.17 133	94.51 176	66.96 174	81.84 171	71.21 187	91.64 165	91.52 197	91.43 188	99.42 171	99.92 120

divwei89l23v2f112		87.46 174	88.97 186	85.70 164	84.85 185	96.08 169	90.23 176	82.46 126	93.69 195	65.83 187	79.57 189	70.54 195	91.39 177	91.60 193	91.39 192	99.43 165	99.92 120

v7		87.72 170	89.75 168	85.35 169	85.01 177	95.79 179	90.43 165	78.98 158	94.50 179	66.39 180	78.87 195	73.65 172	91.85 161	93.69 170	91.86 179	99.45 159	99.92 120

v6		87.96 163	89.58 172	86.08 152	85.34 168	96.14 165	90.44 163	82.19 132	94.56 175	67.43 172	81.90 169	71.57 184	91.62 168	91.54 195	91.43 188	99.43 165	99.92 120

V42		87.84 167	89.42 180	85.99 156	85.16 172	96.01 174	90.52 159	81.78 138	94.43 182	67.59 166	81.32 175	71.87 180	91.48 172	91.25 207	91.16 202	99.43 165	99.92 120

CDS-MVSNet		94.32 101	97.00 96	91.19 106	89.82 138	98.71 121	95.51 111	85.14 101	96.85 152	82.33 114	92.48 130	96.40 87	94.71 130	96.86 111	97.76 73	99.63 122	99.92 120
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
MAR-MVS		97.03 50	98.00 72	95.89 46	99.32 29	99.74 73	96.76 90	84.89 102	99.97 6	94.86 24	98.29 68	90.58 114	99.67 20	98.02 79	99.50 16	99.82 39	99.92 120
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
v1240		86.24 199	88.56 196	83.54 193	84.05 203	95.21 195	89.27 192	76.76 173	93.42 201	60.68 209	75.99 211	69.80 204	91.21 184	93.83 166	91.76 181	99.29 197	99.91 131

v144192		86.80 189	88.90 190	84.35 180	84.33 198	95.56 185	89.34 191	77.74 166	93.60 197	64.03 195	77.82 202	70.76 194	91.28 180	92.91 184	91.74 182	99.37 189	99.90 132

v1192		86.93 187	89.01 183	84.50 178	84.46 196	95.51 187	89.93 186	78.65 161	93.75 191	62.29 202	77.19 205	70.88 193	92.28 155	93.84 164	91.96 176	99.38 187	99.90 132

v748		84.47 206	86.06 211	82.62 202	82.85 211	95.02 197	83.73 210	78.48 162	90.20 218	67.45 171	75.86 213	61.27 223	83.84 212	89.87 214	90.28 208	99.34 190	99.90 132

test123		48.14 234	58.11 231	36.51 235	8.71 239	56.81 240	59.55 232	24.08 237	77.50 232	14.41 242	49.20 231	11.94 243	80.98 215	41.62 237	69.81 233	31.32 240	99.90 132

ACMMP_Plus		98.68 25	99.58 29	97.62 27	99.62 6	99.92 35	99.72 18	96.78 16	99.71 61	90.13 68	99.66 15	99.99 26	99.64 23	99.78 13	98.14 60	99.82 39	99.89 136

TranMVSNet+NR-MVSNet		88.88 154	89.90 167	87.69 141	84.06 202	95.68 180	91.88 147	85.23 98	95.16 168	72.54 141	83.06 163	70.14 201	92.93 149	90.81 211	94.53 140	99.48 150	99.89 136

FMVSNet3		95.59 86	97.51 82	93.34 87	89.48 140	96.57 157	97.67 63	84.17 110	99.48 78	89.76 71	95.09 104	94.35 97	99.14 53	98.37 60	98.86 38	99.82 39	99.89 136

CANet_DTU		94.90 95	98.98 41	90.13 115	94.74 73	99.81 64	98.53 53	82.23 131	99.97 6	66.76 177	100.00 1	98.50 65	98.74 68	97.52 92	97.19 93	99.76 73	99.88 139

v1921920		86.81 188	88.93 188	84.33 183	84.23 200	95.41 191	90.09 180	78.10 164	93.74 193	62.17 203	76.98 207	71.14 190	92.05 157	93.69 170	91.69 183	99.32 193	99.88 139

EPMVS		94.08 106	98.54 56	88.87 126	92.51 107	99.47 89	94.18 130	66.53 210	99.68 63	82.40 113	95.24 101	99.40 57	97.86 88	98.12 71	97.99 62	99.75 86	99.88 139

TAPA-MVS	96.62 5	97.60 42	98.46 58	96.60 39	98.73 42	99.90 43	99.30 32	94.96 35	99.46 82	87.57 88	96.05 96	98.53 64	99.26 46	98.04 77	97.33 84	99.77 66	99.88 139
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
v7n		85.39 203	87.70 208	82.70 200	82.77 212	95.64 181	88.27 197	74.83 180	92.30 211	62.58 201	76.37 210	64.80 220	88.38 203	94.29 157	90.61 206	99.34 190	99.87 143

v8		87.54 171	89.33 181	85.45 168	85.41 166	95.50 188	90.32 173	78.94 159	94.35 184	66.93 176	81.90 169	70.99 192	91.62 168	91.49 200	91.22 199	99.48 150	99.87 143

ACMH+	92.61 13	91.80 136	93.03 146	90.37 112	93.03 91	98.17 133	94.00 132	84.13 113	98.12 140	77.39 129	91.95 132	74.62 165	94.36 136	94.62 152	93.82 153	99.32 193	99.87 143

v10		87.40 177	89.62 171	84.80 175	84.93 180	95.07 196	90.44 163	75.63 178	94.51 176	66.52 178	78.87 195	73.47 173	91.86 160	93.69 170	91.87 178	99.45 159	99.86 146

Baseline_NR-MVSNet		89.13 152	89.53 176	88.66 132	84.71 190	94.43 199	91.79 149	84.49 107	95.54 162	78.28 127	78.52 201	72.46 178	93.29 147	91.10 208	94.82 137	99.42 171	99.86 146

ADS-MVSNet		92.91 122	97.97 73	87.01 145	92.07 115	99.27 98	92.70 140	65.39 219	99.85 41	75.40 135	94.93 109	98.26 71	96.86 105	96.09 129	97.52 80	99.65 120	99.84 148

PEN-MVS		87.20 181	88.22 201	86.01 154	84.01 204	94.93 198	90.00 182	81.52 143	93.46 200	69.29 159	79.69 187	65.51 218	91.72 162	91.01 210	93.12 162	99.49 146	99.84 148

WR-MVS		88.23 160	90.15 165	86.00 155	84.39 197	95.64 181	89.96 184	81.80 136	94.46 181	71.60 150	82.10 165	74.36 169	88.76 201	92.48 187	92.20 173	99.46 156	99.83 150

v52		85.80 200	87.74 206	83.53 194	82.87 210	95.31 194	88.71 194	77.04 171	92.23 212	63.53 198	76.91 208	69.80 204	89.78 190	90.05 213	90.07 209	99.26 201	99.82 151

V4		85.78 201	87.74 206	83.50 195	82.90 209	95.33 193	88.62 195	77.05 170	92.14 214	63.45 199	76.91 208	69.85 203	89.72 191	90.07 212	90.05 210	99.27 200	99.81 152

GBi-Net		95.19 90	96.99 97	93.09 91	89.11 141	96.47 159	96.90 85	84.17 110	99.48 78	89.76 71	95.09 104	94.35 97	98.87 60	96.50 115	97.21 87	99.74 90	99.81 152

test1		95.19 90	96.99 97	93.09 91	89.11 141	96.47 159	96.90 85	84.17 110	99.48 78	89.76 71	95.09 104	94.35 97	98.87 60	96.50 115	97.21 87	99.74 90	99.81 152

FMVSNet2		94.48 99	95.95 122	92.77 98	89.11 141	96.47 159	96.90 85	83.38 118	99.11 101	88.64 82	87.50 154	92.26 109	98.87 60	97.91 82	98.60 49	99.74 90	99.81 152

tfpnview11		95.78 74	98.17 68	93.01 93	92.58 101	99.04 108	96.64 91	88.72 78	99.63 68	83.08 109	98.90 52	94.24 100	97.25 99	98.35 62	97.21 87	99.77 66	99.80 156

LS3D		96.44 60	97.31 86	95.41 53	97.06 55	99.87 50	99.51 27	97.48 1	99.57 70	79.00 122	95.39 100	89.19 120	99.81 11	98.55 48	98.84 39	99.62 124	99.78 157

Vis-MVSNet (Re-imp)		95.60 85	98.52 57	92.19 101	92.37 111	99.56 83	96.37 96	87.41 91	98.95 108	84.77 104	94.88 110	98.48 68	92.44 153	98.63 43	99.37 18	99.76 73	99.77 158

tfpn_n400		95.76 78	98.21 64	92.90 95	92.57 105	99.05 106	96.42 94	88.50 81	99.49 76	83.08 109	98.90 52	94.24 100	97.07 101	98.10 73	97.93 66	99.74 90	99.76 159

tfpnconf		95.76 78	98.21 64	92.90 95	92.57 105	99.05 106	96.42 94	88.50 81	99.49 76	83.08 109	98.90 52	94.24 100	97.07 101	98.10 73	97.93 66	99.74 90	99.76 159

anonymousdsp		87.98 162	92.38 149	82.85 199	83.68 206	96.79 152	90.78 157	74.06 187	95.29 166	57.91 214	83.33 161	83.12 144	91.15 185	95.96 133	92.37 172	99.52 139	99.76 159

MSDG		97.29 45	97.55 81	97.00 33	98.66 43	99.71 74	99.03 40	96.15 30	99.59 69	89.67 75	92.77 129	94.86 95	98.75 67	98.22 67	97.94 64	99.72 106	99.76 159

UGNet		96.05 65	98.55 55	93.13 89	94.64 75	99.65 78	94.70 123	87.78 87	99.40 88	89.69 74	98.25 70	99.25 59	92.12 156	96.50 115	97.08 95	99.84 29	99.72 163
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
FMVSNet1		92.55 129	93.66 142	91.26 105	87.91 151	96.12 166	94.75 122	81.69 139	97.67 145	85.63 98	80.56 179	87.88 126	98.15 80	96.50 115	97.21 87	99.41 183	99.71 164

DTE-MVSNet		86.70 191	87.66 209	85.58 166	83.30 207	94.29 200	89.74 189	81.53 141	92.77 209	68.93 160	80.13 181	64.00 221	90.62 188	89.45 215	93.34 160	99.32 193	99.67 165

pm-mvs1		89.68 148	92.00 151	86.96 146	86.23 159	96.62 156	90.36 168	83.05 122	93.97 186	72.15 146	81.77 174	82.10 147	90.69 187	95.38 140	94.50 141	99.29 197	99.65 166

.test1245		70.78 224	79.90 219	60.13 229	69.34 227	81.48 228	59.42 233	52.46 234	78.55 230	50.42 222	62.43 227	71.67 182	80.43 216	86.79 218	78.71 227	48.74 238	99.65 166

testmvs		61.76 229	72.90 227	48.76 233	21.21 238	68.61 236	66.11 230	37.38 236	94.83 172	33.06 238	64.31 225	29.72 237	86.08 210	74.44 228	78.71 227	48.74 238	99.65 166

TSAR-MVS + COLMAP		95.20 89	95.03 131	95.41 53	96.17 61	98.69 122	99.11 37	93.40 38	99.97 6	84.89 103	98.23 72	75.01 164	99.34 37	97.27 107	96.37 108	99.58 129	99.64 169

v11		86.74 190	89.01 183	84.09 190	84.79 187	91.79 215	90.39 167	72.53 200	94.47 180	65.75 188	78.64 198	72.96 175	91.66 163	93.92 162	91.69 183	99.42 171	99.61 170

v18		87.14 184	88.96 187	85.01 171	85.57 163	92.03 208	90.89 155	74.62 182	94.80 173	67.90 163	82.02 167	71.28 186	91.63 167	91.53 196	91.44 187	99.47 153	99.60 171

v17		86.99 185	88.90 190	84.76 176	85.52 165	91.96 209	90.50 160	74.17 183	94.88 171	67.33 173	81.94 168	71.21 187	91.57 170	91.49 200	91.20 200	99.48 150	99.60 171

v16		87.15 183	89.13 182	84.83 174	85.55 164	91.94 210	90.50 160	74.13 186	95.06 169	67.72 165	81.84 171	72.55 176	91.65 164	91.50 199	91.42 191	99.42 171	99.60 171

v15		86.50 195	88.32 199	84.37 179	85.00 178	91.86 211	90.30 174	73.76 189	93.90 189	66.28 183	79.78 186	70.37 199	91.45 174	91.48 202	91.27 196	99.43 165	99.58 174

V14		86.54 194	88.41 198	84.35 180	84.94 179	91.83 212	90.28 175	73.48 191	93.73 194	66.50 179	79.89 184	71.12 191	91.46 173	91.48 202	91.25 197	99.42 171	99.58 174

SixPastTwentyTwo		88.35 156	91.51 156	84.66 177	85.39 167	96.96 151	86.57 202	79.62 152	96.57 155	63.73 197	87.86 148	75.18 163	93.43 146	94.03 159	90.37 207	99.24 202	99.58 174

OMC-MVS		98.59 28	99.07 39	98.03 23	99.41 24	99.90 43	99.26 34	94.33 36	99.94 22	96.03 16	96.68 87	99.72 45	99.42 32	98.86 34	98.84 39	99.72 106	99.58 174

v12		86.32 197	88.22 201	84.10 188	84.76 188	91.80 213	89.94 185	72.97 197	93.85 190	66.18 184	79.98 183	69.72 206	91.33 178	91.40 205	91.20 200	99.42 171	99.56 178

v13		86.27 198	88.16 203	84.06 191	84.85 185	91.77 216	90.00 182	72.77 199	93.56 198	66.06 186	79.25 193	70.50 197	91.25 182	91.35 206	91.15 203	99.42 171	99.55 179

V9		86.42 196	88.26 200	84.27 184	84.88 182	91.80 213	90.34 169	73.18 195	93.92 188	66.37 182	79.68 188	70.25 200	91.42 175	91.43 204	91.23 198	99.42 171	99.55 179

PatchMatch-RL		96.84 53	98.03 71	95.47 49	98.84 40	99.81 64	95.61 109	89.20 64	99.65 64	91.28 50	99.39 33	93.46 105	98.18 79	98.05 75	96.28 109	99.69 116	99.55 179

ACMM	94.44 10	94.26 104	94.62 135	93.84 81	94.86 72	97.73 141	93.48 137	90.76 48	99.27 95	87.46 89	99.04 46	76.60 157	96.76 109	96.37 122	93.76 154	99.74 90	99.55 179
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
Vis-MVSNet		93.08 119	96.76 101	88.78 130	91.14 127	99.63 80	94.85 120	83.34 119	97.19 150	74.78 138	91.92 134	93.15 107	88.81 200	97.59 89	98.35 55	99.78 61	99.49 183
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
GA-MVS		90.38 145	94.59 136	85.46 167	88.30 149	98.44 126	92.18 145	83.30 120	97.89 144	58.05 213	92.86 127	84.25 143	91.27 181	96.65 113	92.61 170	99.66 119	99.43 184

PatchmatchNet		93.48 114	98.84 47	87.22 143	91.93 117	99.39 93	92.55 143	66.06 214	99.71 61	75.61 134	98.24 71	99.59 49	97.35 97	97.87 83	97.64 78	99.83 34	99.43 184
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
TAMVS		92.43 132	94.21 140	90.35 113	88.68 148	98.85 117	94.15 131	81.53 141	95.58 161	83.61 107	87.05 155	86.45 128	94.71 130	96.27 125	95.91 122	99.42 171	99.38 186

FC-MVSNet-test		92.78 124	96.19 120	88.80 129	88.00 150	97.54 143	93.60 135	82.36 130	98.16 138	79.71 120	91.55 135	95.41 94	89.65 192	96.09 129	95.23 134	99.49 146	99.31 187

pmmvs4		91.41 139	93.05 145	89.49 122	85.85 161	96.52 158	91.70 150	82.49 125	98.14 139	83.17 108	87.57 151	81.76 148	94.39 134	95.47 138	92.62 169	99.33 192	99.29 188

MDTV_nov1_ep13_2view		87.75 169	93.32 143	81.26 207	83.74 205	96.64 155	85.66 205	66.20 213	98.36 135	61.61 204	84.34 158	87.95 125	91.12 186	94.01 160	92.66 168	99.22 203	99.27 189

pmmvs6		85.75 202	86.97 210	84.34 182	84.88 182	95.59 183	87.41 200	79.19 156	87.81 222	67.56 167	63.05 226	77.76 152	89.15 196	93.45 175	91.90 177	97.83 217	99.21 190

testgi		92.47 131	95.68 127	88.73 131	90.68 131	98.35 128	91.67 151	79.50 153	98.96 107	77.12 130	95.17 103	85.84 130	93.95 140	95.75 135	96.47 104	99.45 159	99.21 190

CVMVSNet		92.13 134	95.40 128	88.32 137	91.29 126	97.29 148	91.85 148	86.42 93	96.71 154	71.84 149	89.56 140	91.18 111	88.98 199	96.17 127	97.76 73	99.51 143	99.14 192

EG-PatchMatch MVS		86.96 186	89.56 174	83.93 192	86.29 158	97.61 142	90.75 158	73.31 193	95.43 165	66.08 185	75.88 212	71.31 185	87.55 206	94.79 149	92.74 167	99.61 125	99.13 193

PM-MVS		82.79 212	84.51 214	80.77 208	77.22 218	92.13 207	83.61 211	73.31 193	93.50 199	61.06 205	77.15 206	46.52 230	90.55 189	94.14 158	89.05 215	98.85 211	99.12 194

pmmvs-eth3d		82.92 211	83.31 216	82.47 203	76.97 219	91.76 217	83.79 208	76.10 177	90.33 216	69.95 158	71.04 219	48.09 227	89.02 198	93.85 163	89.14 213	99.02 209	98.96 195

COLMAP_ROB	93.56 12	96.03 66	96.83 100	95.11 55	97.87 50	99.52 85	98.81 46	91.40 44	99.42 85	84.97 101	90.46 138	96.82 83	98.05 81	96.46 119	96.19 112	99.54 135	98.92 196
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
USDC		90.36 146	91.68 154	88.82 128	92.58 101	98.02 135	96.27 97	79.83 149	98.37 134	70.61 156	89.05 142	67.50 211	94.17 137	95.77 134	94.43 143	99.46 156	98.62 197

UA-Net		94.95 94	98.66 52	90.63 109	94.60 77	98.94 115	96.03 100	85.28 97	98.01 143	78.92 123	97.42 81	99.96 31	89.09 197	98.95 32	98.80 42	99.82 39	98.57 198

OPM-MVS		93.50 111	93.00 148	94.07 79	95.82 64	98.26 132	98.49 55	91.62 42	94.69 174	81.93 116	92.82 128	76.18 162	96.82 106	96.12 128	94.57 139	99.74 90	98.39 199

LP		88.31 158	93.18 144	82.63 201	90.66 132	97.98 137	87.32 201	63.49 226	97.17 151	63.02 200	82.08 166	90.47 115	91.92 158	92.75 186	93.42 159	99.38 187	98.37 200

EPNet_dtu		95.10 93	98.81 48	90.78 107	98.38 47	98.47 125	96.54 92	89.36 61	99.78 52	65.65 189	99.31 37	98.24 73	94.79 129	98.28 64	99.35 21	99.93 16	98.27 201
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
TinyColmap		89.94 147	90.88 160	88.84 127	92.43 108	97.91 139	95.59 110	80.10 147	98.12 140	71.33 153	84.56 156	67.46 212	94.15 138	95.57 136	94.27 149	99.43 165	98.26 202

ambc			74.33 225		66.84 232	84.26 227	84.17 206		93.39 203	58.99 211	45.93 232	18.06 242	70.61 226	93.94 161	86.62 222	92.61 230	98.13 203

pmmvs5		87.33 178	90.01 166	84.20 186	84.31 199	96.04 173	87.63 199	76.59 175	93.17 206	65.35 193	84.30 159	71.68 181	91.91 159	95.41 139	91.37 195	99.39 185	98.13 203

MS-PatchMatch		93.46 115	95.91 124	90.61 110	95.48 65	99.31 97	95.62 108	77.23 169	99.42 85	81.88 117	88.92 143	96.06 90	93.80 141	96.45 120	93.11 163	99.65 120	98.10 205

EU-MVSNet		87.20 181	90.47 164	83.38 197	85.11 175	93.85 204	86.10 204	79.76 151	93.30 205	65.39 192	84.41 157	78.43 151	85.04 211	92.20 189	93.03 165	98.86 210	98.05 206

Anonymous20231211		74.10 218	74.22 226	73.97 220	74.36 222	87.76 224	75.92 222	72.78 198	74.83 234	52.25 220	44.18 233	42.42 233	73.07 225	86.16 221	86.24 223	95.44 225	97.94 207

MDA-MVSNet-bldmvs		80.30 216	82.83 217	77.34 213	69.16 229	94.29 200	72.16 224	81.97 135	90.14 219	57.32 216	94.01 118	47.97 228	86.81 209	68.74 231	86.82 221	96.63 219	97.86 208

new-patchmatchnet		78.17 217	80.82 218	75.07 217	76.93 220	91.20 219	71.90 225	73.32 192	86.59 224	48.91 224	67.11 222	47.85 229	81.19 214	88.18 217	87.02 220	98.19 215	97.79 209

N_pmnet		87.31 179	91.51 156	82.41 204	85.13 173	95.57 184	80.59 216	81.79 137	96.20 157	58.52 212	78.62 199	85.66 131	89.36 195	94.64 151	92.14 174	99.08 208	97.72 210

TransMVSNet (Re)		88.33 157	89.55 175	86.91 147	86.65 156	95.56 185	90.48 162	84.44 108	92.02 215	71.07 155	80.13 181	72.48 177	89.41 194	95.05 147	94.44 142	99.39 185	97.14 211

MIMVSNet		91.01 142	96.22 118	84.93 173	85.24 171	98.09 134	90.40 166	64.96 221	97.55 147	72.65 140	96.23 92	90.81 113	96.79 107	96.69 112	97.06 97	99.52 139	97.09 212

TDRefinement		87.79 168	88.76 194	86.66 149	93.54 81	98.02 135	95.76 103	85.18 100	96.57 155	67.90 163	80.51 180	66.51 216	78.37 218	93.20 179	89.73 211	99.22 203	96.75 213

MVS-HIRNet		88.27 159	94.05 141	81.51 205	88.90 146	98.93 116	83.38 212	60.52 231	98.06 142	63.78 196	80.67 178	90.36 116	92.94 148	97.29 106	96.41 107	99.56 132	96.66 214

testus		82.22 213	88.82 193	74.52 219	79.14 216	89.37 221	78.38 219	72.99 196	97.57 146	44.54 231	93.44 123	58.13 225	74.20 223	92.96 183	93.67 156	97.89 216	96.58 215

RPSCF		95.86 70	96.94 99	94.61 72	96.52 57	98.67 123	98.54 52	88.43 83	99.56 72	90.51 66	99.39 33	98.70 62	97.72 91	93.77 167	92.00 175	95.93 221	96.50 216

Anonymous20231206		84.28 207	89.53 176	78.17 211	82.31 214	94.16 202	82.57 213	76.51 176	93.38 204	52.98 219	79.47 191	73.74 171	75.45 220	95.07 146	94.41 144	99.18 206	96.46 217

test2356		83.84 210	91.77 153	74.59 218	78.71 217	89.10 222	78.24 221	72.07 202	96.78 153	45.18 230	96.19 95	76.77 156	74.87 222	93.17 180	94.01 152	98.44 214	96.38 218

pmmvs3		80.91 214	85.62 212	75.42 216	75.01 221	89.09 223	75.31 223	68.70 204	86.99 223	46.74 229	81.18 177	62.91 222	87.95 204	93.84 164	89.06 214	98.80 212	96.23 219

test20.03		83.86 209	88.73 195	78.16 212	82.60 213	93.00 205	81.61 215	74.68 181	92.36 210	57.50 215	83.01 164	74.48 167	73.30 224	92.40 188	91.14 204	99.29 197	94.75 220

FMVSNet5		93.53 110	96.09 121	90.56 111	86.74 154	92.84 206	92.64 142	77.50 167	99.41 87	88.97 80	98.02 75	97.81 76	98.00 84	94.85 148	95.43 132	99.50 145	94.25 221

DeepMVS_CX						97.31 147	79.48 218	89.65 59	98.66 116	60.89 207	94.40 113	66.89 213	87.65 205	81.69 226		92.76 229	94.24 222

MIMVSNet1		80.64 215	83.97 215	76.76 215	68.91 230	91.15 220	78.32 220	75.47 179	89.58 221	56.64 217	65.10 224	65.17 219	82.14 213	93.51 174	91.64 185	99.10 207	91.66 223

new_pmnet		84.12 208	87.89 204	79.72 209	80.43 215	94.14 203	80.26 217	74.14 185	96.01 159	56.30 218	74.94 214	76.45 159	88.59 202	93.11 182	89.31 212	98.59 213	91.27 224

CMPMVS	65.66 17	84.62 205	85.02 213	84.15 187	95.40 66	97.79 140	88.35 196	79.22 155	89.66 220	60.71 208	72.20 216	73.94 170	87.32 207	86.73 220	84.55 225	93.90 227	90.31 225
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
PMMVS2		65.18 227	68.25 229	61.59 227	61.37 233	79.72 232	59.18 237	61.80 227	64.72 235	37.33 233	53.82 230	35.59 236	54.46 234	73.94 229	80.52 226	95.40 226	89.43 226

testmv		71.50 221	77.62 221	64.36 224	72.64 223	81.28 230	59.32 235	66.24 211	83.91 226	35.02 236	69.74 220	46.18 231	57.12 230	85.60 223	87.48 218	95.84 222	89.16 227

test1235678		71.50 221	77.61 222	64.36 224	72.64 223	81.26 231	59.31 236	66.22 212	83.90 227	35.02 236	69.74 220	46.18 231	57.12 230	85.60 223	87.47 219	95.84 222	89.15 228

Gipuma		71.02 223	72.60 228	69.19 222	71.31 225	75.11 234	66.36 228	61.65 229	94.93 170	47.29 228	38.74 234	38.52 235	75.52 219	86.09 222	85.92 224	93.01 228	88.87 229
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
test12356		69.94 225	75.85 223	63.04 226	70.04 226	79.32 233	61.62 231	65.84 216	80.56 228	36.30 235	71.45 218	39.38 234	48.79 236	83.64 225	88.02 217	95.64 224	88.56 230

MVE	58.81 19	52.07 233	55.15 234	48.48 234	42.45 237	62.35 237	36.41 241	54.70 233	49.88 238	27.65 239	29.98 238	18.08 241	54.87 233	65.93 233	77.26 229	74.79 237	82.59 231
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
EMVS		55.14 231	55.29 233	54.97 230	60.87 234	57.52 238	38.58 240	63.57 225	64.54 236	23.36 241	36.96 235	27.99 238	60.69 228	51.17 235	66.61 234	82.73 235	82.25 232

E-PMN		55.33 230	55.79 232	54.81 231	59.81 235	57.23 239	38.83 239	63.59 224	64.06 237	24.66 240	35.33 236	26.40 239	58.69 229	55.41 234	70.54 232	83.26 233	81.56 233

1111		73.79 220	78.62 220	68.16 223	69.34 227	81.48 228	59.42 233	52.46 234	78.55 230	50.42 222	62.43 227	71.67 182	80.43 216	86.79 218	88.22 216	96.87 218	81.17 234

FPMVS		73.80 219	74.62 224	72.84 221	83.09 208	84.44 226	83.89 207	73.64 190	92.20 213	48.50 225	72.19 217	59.51 224	63.16 227	69.13 230	66.26 235	84.74 232	78.59 235

no-one		52.34 232	53.36 235	51.14 232	57.63 236	69.39 235	35.07 242	61.58 230	44.14 239	37.06 234	34.80 237	26.36 240	32.65 237	50.68 236	70.83 231	82.88 234	77.30 236

PMVS	60.14 18	62.67 228	64.05 230	61.06 228	68.32 231	53.27 241	52.23 238	67.63 207	75.07 233	48.30 226	58.27 229	57.43 226	49.99 235	67.20 232	62.42 236	79.87 236	74.68 237
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
sosnet-low-res		0.00 235	0.00 236	0.00 237	0.00 241	0.00 242	0.00 243	0.00 239	0.00 240	0.00 244	0.00 239	0.00 244	0.00 239	0.00 238	0.00 237	0.00 241	0.00 238

sosnet		0.00 235	0.00 236	0.00 237	0.00 241	0.00 242	0.00 243	0.00 239	0.00 240	0.00 244	0.00 239	0.00 244	0.00 239	0.00 238	0.00 237	0.00 241	0.00 238

MTAPA										96.61 10		100.00 1

MTMP										97.42 7		100.00 1

Patchmatch-RL test							68.01 227

tmp_tt				78.81 210	98.80 41	85.73 225	70.08 226	77.87 165	98.68 114	83.71 105	99.53 27	74.55 166	54.97 232	78.28 227	72.43 230	87.45 231

XVS					95.09 69	99.94 17	97.49 71			88.58 83		99.98 28				99.78 61

X-MVStestdata					95.09 69	99.94 17	97.49 71			88.58 83		99.98 28				99.78 61

mPP-MVS					99.23 33							99.87 37

NP-MVS									99.79 50

Patchmtry						99.00 110	95.46 112	65.50 217		67.51 168

High-resolution multi-view benchmark