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.

EPP-MVSNet		97.75 54	98.71 51	96.63 77	95.68 114	99.56 48	97.51 101	93.10 89	99.22 43	94.99 51	97.18 85	97.30 70	98.65 66	98.83 45	98.93 29	99.84 5	99.92 1

LTVRE_ROB	93.20 16	92.84 165	94.92 147	90.43 193	92.83 152	98.63 123	97.08 119	87.87 161	97.91 136	68.42 220	93.54 144	79.46 212	96.62 124	97.55 128	97.40 115	99.74 49	99.92 1
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
canonicalmvs		97.31 69	97.81 82	96.72 73	96.20 96	99.45 60	98.21 77	91.60 102	99.22 43	95.39 42	98.48 51	90.95 123	99.16 43	97.66 122	99.05 23	99.76 44	99.90 3

v748		91.12 202	91.95 210	90.16 195	90.60 209	97.35 192	91.11 205	87.92 160	94.75 214	80.54 167	86.26 211	75.97 219	91.13 210	94.63 203	94.81 192	99.65 106	99.90 3

PVSNet_Blended_VisFu		97.41 64	98.49 56	96.15 86	97.49 64	99.76 1	96.02 137	93.75 81	99.26 38	93.38 85	93.73 143	99.35 49	96.47 129	98.96 35	98.46 54	99.77 42	99.90 3

CSCG		98.90 26	98.93 45	98.85 20	99.75 2	99.72 4	99.49 18	96.58 36	99.38 19	98.05 11	98.97 28	97.87 63	99.49 17	97.78 116	98.92 30	99.78 37	99.90 3

MVS_0304		98.14 46	99.03 40	97.10 54	98.05 58	99.63 29	99.27 31	94.33 58	99.63 6	93.06 89	97.32 79	99.05 53	98.09 85	98.82 46	98.87 34	99.81 26	99.89 7

PS-CasMVS		92.72 173	93.36 193	91.98 154	91.62 193	97.52 182	94.13 192	88.98 145	95.94 206	81.51 160	87.35 202	79.95 209	95.91 141	96.37 158	96.49 134	99.70 77	99.89 7

CP-MVSNet		93.25 156	94.00 170	92.38 143	91.65 191	97.56 179	94.38 188	89.20 143	96.05 203	83.16 149	89.51 175	81.97 197	96.16 136	96.43 156	96.56 132	99.71 69	99.89 7

WR-MVS_H		93.54 152	94.67 153	92.22 144	91.95 170	97.91 156	94.58 185	88.75 148	96.64 187	83.88 141	90.66 162	85.13 160	94.40 188	96.54 154	95.91 151	99.73 55	99.89 7

FC-MVSNet-train		97.04 74	97.91 81	96.03 91	96.00 104	98.41 138	96.53 130	93.42 84	99.04 67	93.02 90	98.03 65	94.32 100	97.47 102	97.93 109	97.77 98	99.75 45	99.88 11

IterMVS-LS		96.12 106	97.48 91	94.53 108	95.19 127	97.56 179	97.15 114	89.19 144	99.08 59	88.23 119	94.97 133	94.73 96	97.84 95	97.86 113	98.26 73	99.60 139	99.88 11
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
v7n		91.61 201	92.95 202	90.04 196	90.56 210	97.69 169	93.74 195	85.59 178	95.89 207	76.95 200	86.60 209	78.60 216	93.76 199	97.01 144	94.99 183	99.65 106	99.87 13

CHOSEN 1792x2688		96.41 96	96.99 105	95.74 97	98.01 59	99.72 4	97.70 97	90.78 118	99.13 56	90.03 113	87.35 202	95.36 88	98.33 77	98.59 67	98.91 32	99.59 145	99.87 13

CANet		98.46 38	99.16 29	97.64 43	98.48 51	99.64 25	99.35 28	94.71 51	99.53 12	95.17 46	97.63 76	99.59 46	98.38 75	98.88 43	98.99 26	99.74 49	99.86 15

diffmvs		97.50 63	98.63 52	96.18 84	95.88 108	99.26 85	98.19 79	91.08 113	99.36 23	94.32 69	98.24 60	96.83 74	98.22 81	98.45 73	98.42 56	99.42 175	99.86 15

HyFIR lowres test		95.99 108	96.56 112	95.32 102	97.99 60	99.65 20	96.54 128	88.86 146	98.44 112	89.77 116	84.14 214	97.05 72	99.03 57	98.55 69	98.19 76	99.73 55	99.86 15

v52		91.94 197	93.10 200	90.57 189	90.62 208	97.50 184	93.98 193	87.02 166	95.86 208	77.67 197	86.93 206	82.16 195	94.53 185	94.71 202	94.70 195	99.61 132	99.85 18

V4		91.92 198	93.10 200	90.55 190	90.64 207	97.51 183	93.93 194	87.02 166	95.81 210	77.61 198	86.93 206	82.19 194	94.50 186	94.72 201	94.68 196	99.62 129	99.85 18

tfpnnormal		93.85 149	94.12 165	93.54 127	93.22 151	98.24 146	95.45 147	91.96 97	94.61 215	83.91 140	90.74 160	81.75 200	97.04 110	97.49 130	96.16 144	99.68 88	99.84 20

Effi-MVS+		95.81 110	97.31 99	94.06 114	95.09 128	99.35 72	97.24 111	88.22 155	98.54 107	85.38 138	98.52 48	88.68 129	98.70 64	98.32 79	97.93 85	99.74 49	99.84 20

SD-MVS		99.25 8	99.50 6	98.96 16	98.79 47	99.55 50	99.33 29	98.29 9	99.75 1	97.96 13	99.15 19	99.95 14	99.61 5	99.17 25	99.06 22	99.81 26	99.84 20

EPNet		98.05 47	98.86 47	97.10 54	99.02 43	99.43 63	98.47 60	94.73 50	99.05 65	95.62 37	98.93 30	97.62 67	95.48 156	98.59 67	98.55 50	99.29 184	99.84 20
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
SteuartSystems-ACMMP		99.20 12	99.51 5	98.83 22	99.66 13	99.66 19	99.71 4	98.12 22	99.14 52	96.62 29	99.16 18	99.98 1	99.12 49	99.63 3	99.19 19	99.78 37	99.83 24
Skip Steuart: Steuart Systems R&D Blog.
TSAR-MVS + MP.		99.27 6	99.57 2	98.92 18	98.78 48	99.53 51	99.72 2	98.11 23	99.73 2	97.43 20	99.15 19	99.96 9	99.59 8	99.73 1	99.07 21	99.88 1	99.82 25

anonymousdsp		93.12 157	95.86 135	89.93 199	91.09 204	98.25 145	95.12 151	85.08 183	97.44 151	73.30 212	90.89 159	90.78 124	95.25 177	97.91 110	95.96 150	99.71 69	99.82 25

TSAR-MVS + ACMM		98.77 29	99.45 8	97.98 38	99.37 31	99.46 58	99.44 24	98.13 21	99.65 4	92.30 95	98.91 35	99.95 14	99.05 55	99.42 14	98.95 28	99.58 149	99.82 25

PEN-MVS		92.72 173	93.20 199	92.15 147	91.29 201	97.31 193	94.67 180	89.81 136	96.19 199	81.83 158	88.58 189	79.06 214	95.61 150	95.21 182	96.27 139	99.72 60	99.82 25

WR-MVS		93.43 155	94.48 158	92.21 145	91.52 196	97.69 169	94.66 181	89.98 133	96.86 178	83.43 146	90.12 164	85.03 161	93.94 196	96.02 170	95.82 152	99.71 69	99.82 25

HSP-MVS		99.31 3	99.43 13	99.17 2	99.68 10	99.75 2	99.72 2	98.31 5	99.45 16	98.16 9	99.28 12	99.98 1	99.30 30	99.34 19	98.41 58	99.81 26	99.81 30

DeepC-MVS	97.63 4	98.33 42	98.57 53	98.04 36	98.62 50	99.65 20	99.45 22	98.15 18	99.51 14	92.80 92	95.74 123	96.44 77	99.46 19	99.37 16	99.50 2	99.78 37	99.81 30
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
conf0.05thres1000		96.34 99	96.47 120	96.17 85	96.16 97	99.71 8	97.82 92	93.46 83	98.10 124	90.69 105	96.75 94	85.26 159	99.11 51	98.05 103	97.65 100	99.82 13	99.80 32

v8		92.87 164	93.87 178	91.72 164	92.05 166	97.50 184	94.79 169	88.20 156	96.85 179	80.11 175	90.01 167	82.86 184	95.48 156	95.15 193	94.90 189	99.66 100	99.80 32

v10		92.79 170	94.06 168	91.31 172	91.78 180	97.29 195	94.87 162	86.10 176	96.97 172	79.82 178	88.16 194	84.56 165	95.63 147	96.33 161	95.31 164	99.65 106	99.80 32

UniMVSNet_NR-MVSNet		94.59 134	95.47 142	93.55 126	91.85 174	97.89 157	95.03 153	92.00 95	97.33 159	86.12 131	93.19 149	87.29 133	96.60 125	96.12 167	96.70 127	99.72 60	99.80 32

DU-MVS		93.98 144	94.44 159	93.44 129	91.66 189	97.77 159	95.03 153	91.57 103	97.17 164	86.12 131	93.13 151	81.13 202	96.60 125	95.10 196	97.01 122	99.67 95	99.80 32

UGNet		97.66 57	99.07 35	96.01 92	97.19 73	99.65 20	97.09 118	93.39 85	99.35 25	94.40 66	98.79 40	99.59 46	94.24 191	98.04 105	98.29 72	99.73 55	99.80 32
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
IS_MVSNet		97.86 50	98.86 47	96.68 74	96.02 102	99.72 4	98.35 73	93.37 87	98.75 98	94.01 72	96.88 93	98.40 59	98.48 72	99.09 29	99.42 5	99.83 9	99.80 32

UniMVSNet (Re)		94.58 135	95.34 143	93.71 121	92.25 161	98.08 150	94.97 155	91.29 112	97.03 170	87.94 122	93.97 142	86.25 149	96.07 137	96.27 164	95.97 149	99.72 60	99.79 39

DELS-MVS		98.19 44	98.77 50	97.52 45	98.29 54	99.71 8	99.12 37	94.58 56	98.80 89	95.38 43	96.24 110	98.24 61	97.92 90	99.06 32	99.52 1	99.82 13	99.79 39
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
tfpn		96.22 103	95.62 140	96.93 71	96.29 90	99.72 4	98.34 74	93.94 75	97.96 132	93.94 74	96.45 106	79.09 213	99.22 32	98.28 81	98.06 80	99.83 9	99.78 41

v144192		92.38 185	93.55 190	91.00 180	91.44 197	97.47 187	94.27 189	87.41 164	96.52 193	78.03 194	87.50 201	82.65 186	95.32 172	95.82 173	95.15 169	99.55 157	99.78 41

V42		93.05 161	93.90 177	92.04 150	91.91 171	97.66 171	94.91 157	89.91 134	96.85 179	80.58 166	89.66 174	83.43 174	95.37 168	95.03 199	94.90 189	99.59 145	99.78 41

MVS_Test		97.30 70	98.54 54	95.87 93	95.74 112	99.28 83	98.19 79	91.40 107	99.18 49	91.59 101	98.17 61	96.18 80	98.63 68	98.61 63	98.55 50	99.66 100	99.78 41

TranMVSNet+NR-MVSNet		93.67 151	94.14 163	93.13 135	91.28 203	97.58 178	95.60 144	91.97 96	97.06 168	84.05 139	90.64 163	82.22 193	96.17 135	94.94 200	96.78 125	99.69 79	99.78 41

PVSNet_BlendedMVS		97.51 61	97.71 84	97.28 49	98.06 56	99.61 37	97.31 107	95.02 45	99.08 59	95.51 40	98.05 63	90.11 125	98.07 86	98.91 40	98.40 59	99.72 60	99.78 41

PVSNet_Blended		97.51 61	97.71 84	97.28 49	98.06 56	99.61 37	97.31 107	95.02 45	99.08 59	95.51 40	98.05 63	90.11 125	98.07 86	98.91 40	98.40 59	99.72 60	99.78 41

Fast-Effi-MVS+		95.38 119	96.52 115	94.05 115	94.15 138	99.14 95	97.24 111	86.79 169	98.53 108	87.62 125	94.51 138	87.06 134	98.76 62	98.60 66	98.04 82	99.72 60	99.77 48

v7		92.97 163	94.11 166	91.65 165	91.83 175	97.55 181	94.86 165	88.19 157	96.96 173	79.72 181	88.16 194	84.68 164	95.63 147	96.33 161	95.30 165	99.65 106	99.77 48

APDe-MVS		99.49 1	99.64 1	99.32 1	99.74 3	99.74 3	99.75 1	98.34 2	99.56 9	98.72 3	99.57 4	99.97 3	99.53 14	99.65 2	99.25 13	99.84 5	99.77 48

ACMMPR		99.30 5	99.54 3	99.03 12	99.66 13	99.64 25	99.68 5	98.25 12	99.56 9	97.12 25	99.19 16	99.95 14	99.72 1	99.43 13	99.25 13	99.72 60	99.77 48

view800		96.70 87	96.45 123	96.99 69	96.29 90	99.69 11	98.39 70	93.95 74	97.92 135	94.25 71	96.23 111	85.57 155	99.22 32	98.28 81	97.71 99	99.82 13	99.76 52

v1144		92.81 166	94.03 169	91.40 169	91.68 188	97.60 177	94.73 176	88.40 153	96.71 182	78.48 193	88.14 196	84.46 166	95.45 163	96.31 163	95.22 167	99.65 106	99.76 52

HFP-MVS		99.32 2	99.53 4	99.07 9	99.69 7	99.59 43	99.63 10	98.31 5	99.56 9	97.37 21	99.27 13	99.97 3	99.70 3	99.35 18	99.24 15	99.71 69	99.76 52

v1neww		93.06 159	93.94 173	92.03 151	91.99 168	97.70 165	94.79 169	90.14 129	96.93 175	80.13 173	89.97 169	83.01 179	95.48 156	95.16 189	95.01 178	99.63 123	99.76 52

v7new		93.06 159	93.94 173	92.03 151	91.99 168	97.70 165	94.79 169	90.14 129	96.93 175	80.13 173	89.97 169	83.01 179	95.48 156	95.16 189	95.01 178	99.63 123	99.76 52

v6		93.11 158	93.98 171	92.10 148	92.01 167	97.71 162	94.86 165	90.15 128	96.96 173	80.47 168	90.01 167	83.26 175	95.48 156	95.17 185	95.01 178	99.64 117	99.76 52

MSLP-MVS++		99.15 14	99.24 26	99.04 11	99.52 26	99.49 55	99.09 40	98.07 24	99.37 21	98.47 5	97.79 70	99.89 29	99.50 15	98.93 38	99.45 4	99.61 132	99.76 52

ACMMP		98.74 30	99.03 40	98.40 28	99.36 33	99.64 25	99.20 33	97.75 31	98.82 86	95.24 45	98.85 38	99.87 31	99.17 41	98.74 54	97.50 108	99.71 69	99.76 52
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
ACMH	95.42 14	95.27 123	95.96 132	94.45 110	96.83 78	98.78 111	94.72 177	91.67 101	98.95 71	86.82 130	96.42 107	83.67 171	97.00 111	97.48 131	96.68 128	99.69 79	99.76 52
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
X-MVS		98.93 25	99.37 18	98.42 27	99.67 11	99.62 33	99.60 12	98.15 18	99.08 59	93.81 79	98.46 53	99.95 14	99.59 8	99.49 10	99.21 18	99.68 88	99.75 61

NR-MVSNet		94.01 142	94.51 157	93.44 129	92.56 156	97.77 159	95.67 141	91.57 103	97.17 164	85.84 134	93.13 151	80.53 205	95.29 175	97.01 144	96.17 143	99.69 79	99.75 61

Vis-MVSNet (Re-imp)		97.40 65	98.89 46	95.66 99	95.99 105	99.62 33	97.82 92	93.22 88	98.82 86	91.40 102	96.94 91	98.56 57	95.70 145	99.14 27	99.41 6	99.79 34	99.75 61

ACMMP_Plus		99.05 21	99.45 8	98.58 26	99.73 4	99.60 41	99.64 8	98.28 10	99.23 42	94.57 57	99.35 11	99.97 3	99.55 12	99.63 3	98.66 43	99.70 77	99.74 64

v1192		92.43 182	93.61 184	91.05 177	91.53 195	97.43 188	94.61 183	87.99 159	96.60 188	76.72 201	87.11 204	82.74 185	95.85 142	96.35 160	95.30 165	99.60 139	99.74 64

PGM-MVS		98.86 27	99.35 22	98.29 30	99.77 1	99.63 29	99.67 6	95.63 39	98.66 101	95.27 44	99.11 22	99.82 36	99.67 4	99.33 20	99.19 19	99.73 55	99.74 64

CP-MVS		99.27 6	99.44 11	99.08 8	99.62 17	99.58 45	99.53 15	98.16 16	99.21 45	97.79 15	99.15 19	99.96 9	99.59 8	99.54 7	98.86 35	99.78 37	99.74 64

view600		96.70 87	96.44 125	97.01 65	96.28 93	99.67 13	98.42 62	93.99 67	97.87 138	94.34 68	95.99 115	85.94 151	99.20 35	98.26 83	97.64 101	99.82 13	99.73 68

MPTG		99.31 3	99.44 11	99.16 4	99.73 4	99.65 20	99.63 10	98.26 11	99.27 35	98.01 12	99.27 13	99.97 3	99.60 6	99.59 5	98.58 49	99.71 69	99.73 68

v1921920		92.36 187	93.57 187	90.94 182	91.39 199	97.39 190	94.70 178	87.63 163	96.60 188	76.63 202	86.98 205	82.89 183	95.75 143	96.26 165	95.14 170	99.55 157	99.73 68

DI_MVS_plusplus_trai		96.90 79	97.49 90	96.21 83	95.61 116	99.40 66	98.72 54	92.11 92	99.14 52	92.98 91	93.08 153	95.14 90	98.13 84	98.05 103	97.91 87	99.74 49	99.73 68

v1240		91.99 196	93.33 194	90.44 192	91.29 201	97.30 194	94.25 190	86.79 169	96.43 197	75.49 207	86.34 210	81.85 199	95.29 175	96.42 157	95.22 167	99.52 163	99.73 68

v1		92.81 166	93.57 187	91.94 155	91.79 179	97.70 165	94.80 168	90.32 123	96.52 193	79.75 179	88.47 190	82.46 190	95.32 172	95.14 195	94.96 185	99.63 123	99.73 68

thres600view7		96.69 89	96.43 127	97.00 67	96.28 93	99.67 13	98.41 63	93.99 67	97.85 141	94.29 70	95.96 116	85.91 152	99.19 37	98.26 83	97.63 102	99.82 13	99.73 68

MP-MVS		99.07 19	99.36 19	98.74 23	99.63 16	99.57 47	99.66 7	98.25 12	99.00 69	95.62 37	98.97 28	99.94 22	99.54 13	99.51 9	98.79 40	99.71 69	99.73 68

DTE-MVSNet		92.42 184	92.85 205	91.91 157	90.87 206	96.97 197	94.53 187	89.81 136	95.86 208	81.59 159	88.83 182	77.88 217	95.01 181	94.34 205	96.35 137	99.64 117	99.73 68

Baseline_NR-MVSNet		93.87 147	93.98 171	93.75 119	91.66 189	97.02 196	95.53 145	91.52 106	97.16 166	87.77 124	87.93 200	83.69 170	96.35 130	95.10 196	97.23 117	99.68 88	99.73 68

SixPastTwentyTwo		93.44 154	95.32 144	91.24 174	92.11 164	98.40 139	92.77 200	88.64 151	98.09 125	77.83 195	93.51 145	85.74 153	96.52 128	96.91 146	94.89 191	99.59 145	99.73 68

LGP-MVS_train		96.23 102	96.89 107	95.46 101	97.32 68	98.77 112	98.81 51	93.60 82	98.58 104	85.52 136	99.08 25	86.67 144	97.83 96	97.87 112	97.51 107	99.69 79	99.73 68

v1141		92.79 170	93.61 184	91.84 161	91.75 182	97.71 162	94.74 175	90.33 122	96.58 190	79.21 189	88.59 187	82.53 189	95.36 169	95.16 189	94.96 185	99.63 123	99.72 80

pm-mvs1		94.27 138	95.57 141	92.75 140	92.58 155	98.13 149	94.87 162	90.71 119	96.70 183	83.78 142	89.94 172	89.85 128	94.96 182	97.58 127	97.07 119	99.61 132	99.72 80

divwei89l23v2f112		92.80 168	93.60 186	91.86 160	91.75 182	97.71 162	94.75 174	90.32 123	96.54 192	79.35 186	88.59 187	82.55 188	95.35 170	95.15 193	94.96 185	99.63 123	99.72 80

IterMVS		94.81 129	97.71 84	91.42 168	94.83 134	97.63 173	97.38 104	85.08 183	98.93 76	75.67 205	94.02 140	97.64 65	96.66 123	98.45 73	97.60 104	98.90 193	99.72 80
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
MCST-MVS		99.11 16	99.27 25	98.93 17	99.67 11	99.33 76	99.51 17	98.31 5	99.28 33	96.57 31	99.10 23	99.90 27	99.71 2	99.19 24	98.35 65	99.82 13	99.71 84

ACMP	96.25 10	96.62 94	96.72 109	96.50 81	96.96 77	98.75 115	97.80 94	94.30 59	98.85 81	93.12 88	98.78 41	86.61 145	97.23 107	97.73 120	96.61 130	99.62 129	99.71 84
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
tfpn111		96.96 78	96.91 106	97.03 59	96.31 83	99.67 13	98.41 63	93.99 67	97.35 153	94.50 60	98.65 45	86.93 137	99.14 44	98.26 83	97.80 93	99.82 13	99.70 86

tfpn1000		97.60 59	98.21 68	96.89 72	96.32 82	99.60 41	97.99 89	93.85 77	99.21 45	95.03 49	98.49 50	93.69 109	98.31 78	98.50 72	98.31 71	99.86 2	99.70 86

conf200view11		96.75 82	96.51 116	97.03 59	96.31 83	99.67 13	98.41 63	93.99 67	97.35 153	94.50 60	95.90 118	86.93 137	99.14 44	98.26 83	97.80 93	99.82 13	99.70 86

tfpn200view9		96.75 82	96.51 116	97.03 59	96.31 83	99.67 13	98.41 63	93.99 67	97.35 153	94.52 58	95.90 118	86.93 137	99.14 44	98.26 83	97.80 93	99.82 13	99.70 86

v11		92.43 182	93.77 180	90.85 186	91.72 186	95.58 212	94.87 162	84.07 198	96.98 171	79.28 187	88.03 197	84.22 168	95.53 154	96.55 153	95.36 162	99.65 106	99.70 86

thres400		96.71 86	96.45 123	97.02 63	96.28 93	99.63 29	98.41 63	94.00 66	97.82 143	94.42 65	95.74 123	86.26 148	99.18 39	98.20 90	97.79 97	99.81 26	99.70 86

conf0.01		96.35 98	95.71 137	97.10 54	96.30 89	99.65 20	98.41 63	94.10 63	97.35 153	94.82 54	95.44 131	81.88 198	99.14 44	98.16 92	97.80 93	99.82 13	99.69 92

conf0.002		96.31 100	95.63 139	97.11 53	96.29 90	99.64 25	98.41 63	94.11 62	97.35 153	94.86 52	95.49 130	81.06 203	99.14 44	98.14 93	98.02 83	99.82 13	99.69 92

v148		92.36 187	92.88 203	91.75 162	91.63 192	97.66 171	92.64 201	90.55 121	96.09 201	83.34 147	88.19 193	80.00 208	92.74 204	93.98 206	94.58 197	99.58 149	99.69 92

v2v482		92.77 172	93.52 191	91.90 158	91.59 194	97.63 173	94.57 186	90.31 125	96.80 181	79.22 188	88.74 184	81.55 201	96.04 139	95.26 178	94.97 184	99.66 100	99.69 92

CPTT-MVS		99.14 15	99.20 28	99.06 10	99.58 20	99.53 51	99.45 22	97.80 30	99.19 48	98.32 8	98.58 47	99.95 14	99.60 6	99.28 22	98.20 75	99.64 117	99.69 92

FMVSNet1		95.77 111	96.41 128	95.03 103	93.42 150	97.86 158	97.11 117	89.89 135	98.53 108	92.00 97	89.17 177	93.23 113	98.15 83	98.07 99	98.34 67	99.61 132	99.69 92

Vis-MVSNet		96.16 105	98.22 67	93.75 119	95.33 125	99.70 10	97.27 109	90.85 115	98.30 116	85.51 137	95.72 125	96.45 75	93.69 200	98.70 56	99.00 25	99.84 5	99.69 92
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
v17		92.55 178	93.65 183	91.27 173	92.11 164	95.63 206	94.89 158	85.15 181	97.12 167	80.39 172	90.02 166	83.02 178	95.45 163	95.17 185	94.92 188	99.66 100	99.68 99

v16		92.66 176	93.80 179	91.32 171	92.13 162	95.62 207	94.89 158	85.12 182	97.20 162	80.66 164	89.96 171	83.93 169	95.49 155	95.17 185	95.04 173	99.63 123	99.68 99

v13		92.16 194	93.28 198	90.85 186	91.75 182	95.58 212	94.65 182	84.23 196	96.49 196	79.51 185	88.40 192	82.58 187	95.31 174	95.21 182	95.03 175	99.66 100	99.68 99

v12		92.18 193	93.29 197	90.88 184	91.70 187	95.59 210	94.61 183	84.36 193	96.65 186	79.59 183	88.85 181	82.03 196	95.35 170	95.22 179	95.04 173	99.65 106	99.68 99

MVSTER		97.16 72	97.71 84	96.52 79	95.97 106	98.48 131	98.63 56	92.10 93	98.68 100	95.96 36	99.23 15	91.79 120	96.87 116	98.76 51	97.37 116	99.57 153	99.68 99

ESAPD		99.23 10	99.41 15	99.01 14	99.70 6	99.69 11	99.40 26	98.31 5	98.94 74	97.70 17	99.40 9	99.97 3	99.17 41	99.54 7	98.67 42	99.78 37	99.67 104

v18		92.63 177	93.67 182	91.43 167	92.13 162	95.65 205	95.09 152	85.44 180	97.06 168	80.78 163	90.06 165	83.06 177	95.47 161	95.16 189	95.01 178	99.64 117	99.67 104

v15		92.27 190	93.33 194	91.04 178	91.83 175	95.60 208	94.79 169	84.88 186	96.66 185	79.66 182	88.72 185	82.45 191	95.40 166	95.19 184	95.00 182	99.65 106	99.67 104

V14		92.31 189	93.41 192	91.03 179	91.80 178	95.59 210	94.79 169	84.70 187	96.58 190	79.83 177	88.79 183	82.98 181	95.41 165	95.22 179	95.02 177	99.65 106	99.67 104

V9		92.24 191	93.32 196	90.98 181	91.76 181	95.58 212	94.83 167	84.50 191	96.68 184	79.73 180	88.66 186	82.39 192	95.39 167	95.22 179	95.03 175	99.65 106	99.67 104

GBi-Net		96.98 76	98.00 78	95.78 94	93.81 143	97.98 151	98.09 82	91.32 108	98.80 89	93.92 75	97.21 82	95.94 84	97.89 91	98.07 99	98.34 67	99.68 88	99.67 104

test1		96.98 76	98.00 78	95.78 94	93.81 143	97.98 151	98.09 82	91.32 108	98.80 89	93.92 75	97.21 82	95.94 84	97.89 91	98.07 99	98.34 67	99.68 88	99.67 104

FMVSNet2		96.64 92	97.50 89	95.63 100	93.81 143	97.98 151	98.09 82	90.87 114	98.99 70	93.48 83	93.17 150	95.25 89	97.89 91	98.63 62	98.80 39	99.68 88	99.67 104

3Dnovator+	96.92 7	98.71 32	99.05 36	98.32 29	99.53 24	99.34 74	99.06 42	94.61 53	99.65 4	97.49 19	96.75 94	99.86 32	99.44 21	98.78 49	99.30 11	99.81 26	99.67 104

HPM-MVS++		99.10 17	99.30 23	98.86 19	99.69 7	99.48 56	99.59 13	98.34 2	99.26 38	96.55 32	99.10 23	99.96 9	99.36 25	99.25 23	98.37 63	99.64 117	99.66 113

thres200		96.76 81	96.53 114	97.03 59	96.31 83	99.67 13	98.37 71	93.99 67	97.68 148	94.49 62	95.83 122	86.77 142	99.18 39	98.26 83	97.82 92	99.82 13	99.66 113

3Dnovator	96.92 7	98.67 33	99.05 36	98.23 33	99.57 21	99.45 60	99.11 38	94.66 52	99.69 3	96.80 28	96.55 104	99.61 45	99.40 23	98.87 44	99.49 3	99.85 4	99.66 113

TSAR-MVS + GP.		98.66 35	99.36 19	97.85 40	97.16 74	99.46 58	99.03 44	94.59 55	99.09 57	97.19 24	99.73 3	99.95 14	99.39 24	98.95 36	98.69 41	99.75 45	99.65 116

FMVSNet3		97.02 75	98.12 73	95.73 98	93.59 149	97.98 151	98.34 74	91.32 108	98.80 89	93.92 75	97.21 82	95.94 84	97.63 98	98.61 63	98.62 45	99.61 132	99.65 116

CDS-MVSNet		96.59 95	98.02 77	94.92 105	94.45 136	98.96 102	97.46 103	91.75 99	97.86 140	90.07 112	96.02 114	97.25 71	96.21 132	98.04 105	98.38 61	99.60 139	99.65 116
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
ACMH+	95.51 13	95.40 118	96.00 130	94.70 107	96.33 81	98.79 109	96.79 123	91.32 108	98.77 95	87.18 127	95.60 128	85.46 156	96.97 112	97.15 140	96.59 131	99.59 145	99.65 116

QAPM		98.62 36	99.04 39	98.13 34	99.57 21	99.48 56	99.17 35	94.78 49	99.57 8	96.16 34	96.73 96	99.80 37	99.33 27	98.79 48	99.29 12	99.75 45	99.64 120

DeepC-MVS_fast	98.34 1	99.17 13	99.45 8	98.85 20	99.55 23	99.37 69	99.64 8	98.05 25	99.53 12	96.58 30	98.93 30	99.92 24	99.49 17	99.46 11	99.32 10	99.80 33	99.64 120
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
thres100view900		96.72 85	96.47 120	97.00 67	96.31 83	99.52 54	98.28 76	94.01 65	97.35 153	94.52 58	95.90 118	86.93 137	99.09 54	98.07 99	97.87 89	99.81 26	99.63 122

Effi-MVS+-dtu		95.74 112	98.04 75	93.06 136	93.92 139	99.16 94	97.90 90	88.16 158	99.07 64	82.02 157	98.02 66	94.32 100	96.74 120	98.53 70	97.56 105	99.61 132	99.62 123

HQP-MVS		96.37 97	96.58 111	96.13 88	97.31 70	98.44 135	98.45 61	95.22 43	98.86 79	88.58 118	98.33 57	87.00 136	97.67 97	97.23 137	96.56 132	99.56 156	99.62 123

tfpn_n400		97.32 66	98.38 59	96.09 89	96.07 99	99.30 80	98.00 87	93.84 78	99.35 25	90.50 108	98.93 30	94.24 102	98.30 79	98.65 58	98.60 47	99.83 9	99.60 125

tfpnconf		97.32 66	98.38 59	96.09 89	96.07 99	99.30 80	98.00 87	93.84 78	99.35 25	90.50 108	98.93 30	94.24 102	98.30 79	98.65 58	98.60 47	99.83 9	99.60 125

tfpnview11		97.32 66	98.33 62	96.14 87	96.07 99	99.31 79	98.08 85	93.96 73	99.25 40	90.50 108	98.93 30	94.24 102	98.38 75	98.61 63	98.36 64	99.84 5	99.59 127

IB-MVS	93.96 15	95.02 126	96.44 125	93.36 132	97.05 76	99.28 83	90.43 210	93.39 85	98.02 127	96.02 35	94.92 134	92.07 119	83.52 222	95.38 176	95.82 152	99.72 60	99.59 127
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
train_agg		98.73 31	99.11 31	98.28 31	99.36 33	99.35 72	99.48 20	97.96 27	98.83 84	93.86 78	98.70 44	99.86 32	99.44 21	99.08 31	98.38 61	99.61 132	99.58 129

CDPH-MVS		98.41 39	99.10 32	97.61 44	99.32 37	99.36 70	99.49 18	96.15 38	98.82 86	91.82 98	98.41 54	99.66 44	99.10 52	98.93 38	98.97 27	99.75 45	99.58 129

APD-MVS		99.25 8	99.38 17	99.09 7	99.69 7	99.58 45	99.56 14	98.32 4	98.85 81	97.87 14	98.91 35	99.92 24	99.30 30	99.45 12	99.38 8	99.79 34	99.58 129
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
MVS_111021_LR		98.67 33	99.41 15	97.81 41	99.37 31	99.53 51	98.51 59	95.52 41	99.27 35	94.85 53	99.56 5	99.69 43	99.04 56	99.36 17	98.88 33	99.60 139	99.58 129

PHI-MVS		99.08 18	99.43 13	98.67 24	99.15 40	99.59 43	99.11 38	97.35 33	99.14 52	97.30 22	99.44 8	99.96 9	99.32 28	98.89 42	99.39 7	99.79 34	99.58 129

MVS_111021_HR		98.59 37	99.36 19	97.68 42	99.42 29	99.61 37	98.14 81	94.81 48	99.31 30	95.00 50	99.51 6	99.79 39	99.00 59	98.94 37	98.83 37	99.69 79	99.57 134

tfpn_ndepth		97.71 55	98.30 63	97.02 63	96.31 83	99.56 48	98.05 86	93.94 75	98.95 71	95.59 39	98.40 55	94.79 95	98.39 74	98.40 77	98.42 56	99.86 2	99.56 135

DeepPCF-MVS	97.74 3	98.34 41	99.46 7	97.04 58	98.82 46	99.33 76	96.28 134	97.47 32	99.58 7	94.70 56	98.99 27	99.85 35	97.24 106	99.55 6	99.34 9	97.73 210	99.56 135

CLD-MVS		96.74 84	96.51 116	97.01 65	96.71 79	98.62 124	98.73 53	94.38 57	98.94 74	94.46 63	97.33 78	87.03 135	98.07 86	97.20 139	96.87 124	99.72 60	99.54 137
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
CANet_DTU		96.64 92	99.08 33	93.81 118	97.10 75	99.42 64	98.85 49	90.01 132	99.31 30	79.98 176	99.78 2	99.10 52	97.42 103	98.35 78	98.05 81	99.47 168	99.53 138

pmmvs6		91.90 199	92.53 209	91.17 175	91.81 177	97.63 173	93.23 196	88.37 154	93.43 220	80.61 165	77.32 224	87.47 132	94.12 192	96.58 151	95.72 155	98.88 194	99.53 138

FC-MVSNet-test		96.07 107	97.94 80	93.89 116	93.60 148	98.67 121	96.62 127	90.30 127	98.76 96	88.62 117	95.57 129	97.63 66	94.48 187	97.97 107	97.48 111	99.71 69	99.52 140

CNVR-MVS		99.23 10	99.28 24	99.17 2	99.65 15	99.34 74	99.46 21	98.21 14	99.28 33	98.47 5	98.89 37	99.94 22	99.50 15	99.42 14	98.61 46	99.73 55	99.52 140

PLC	97.93 2	99.02 24	98.94 44	99.11 6	99.46 28	99.24 88	99.06 42	97.96 27	99.31 30	99.16 1	97.90 68	99.79 39	99.36 25	98.71 55	98.12 78	99.65 106	99.52 140
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
MAR-MVS		97.71 55	98.04 75	97.32 47	99.35 35	98.91 104	97.65 98	91.68 100	98.00 128	97.01 26	97.72 74	94.83 93	98.85 61	98.44 75	98.86 35	99.41 176	99.52 140
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
NCCC		99.05 21	99.08 33	99.02 13	99.62 17	99.38 67	99.43 25	98.21 14	99.36 23	97.66 18	97.79 70	99.90 27	99.45 20	99.17 25	98.43 55	99.77 42	99.51 144

OpenMVS	96.23 11	97.95 49	98.45 57	97.35 46	99.52 26	99.42 64	98.91 48	94.61 53	98.87 78	92.24 96	94.61 137	99.05 53	99.10 52	98.64 61	99.05 23	99.74 49	99.51 144

thresconf0.02		97.18 71	97.81 82	96.45 82	96.11 98	99.20 93	98.21 77	94.26 60	99.14 52	91.72 100	98.65 45	91.51 122	98.57 69	98.22 89	98.47 53	99.82 13	99.50 146

Fast-Effi-MVS+-dtu		95.38 119	98.20 69	92.09 149	93.91 140	98.87 106	97.35 106	85.01 185	99.08 59	81.09 161	98.10 62	96.36 78	95.62 149	98.43 76	97.03 120	99.55 157	99.50 146

ACMM	96.26 9	96.67 91	96.69 110	96.66 75	97.29 71	98.46 132	96.48 131	95.09 44	99.21 45	93.19 87	98.78 41	86.73 143	98.17 82	97.84 114	96.32 138	99.74 49	99.49 148
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
CHOSEN 280x420		97.99 48	99.24 26	96.53 78	98.34 53	99.61 37	98.36 72	89.80 138	99.27 35	95.08 48	99.81 1	98.58 56	98.64 67	99.02 33	98.92 30	98.93 192	99.48 149

TAPA-MVS	97.53 5	98.41 39	98.84 49	97.91 39	99.08 42	99.33 76	99.15 36	97.13 34	99.34 28	93.20 86	97.75 72	99.19 51	99.20 35	98.66 57	98.13 77	99.66 100	99.48 149
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
GA-MVS		93.93 146	96.31 129	91.16 176	93.61 147	98.79 109	95.39 149	90.69 120	98.25 118	73.28 213	96.15 112	88.42 130	94.39 189	97.76 118	95.35 163	99.58 149	99.45 151

CVMVSNet		95.33 122	97.09 102	93.27 134	95.23 126	98.39 140	95.49 146	92.58 91	97.71 147	83.00 151	94.44 139	93.28 112	93.92 197	97.79 115	98.54 52	99.41 176	99.45 151

LS3D		97.79 51	98.25 65	97.26 51	98.40 52	99.63 29	99.53 15	98.63 1	99.25 40	88.13 120	96.93 92	94.14 105	99.19 37	99.14 27	99.23 16	99.69 79	99.42 153

abl_6				98.09 35	99.33 36	99.22 90	98.79 52	94.96 47	98.52 110	97.00 27	97.30 80	99.86 32	98.76 62			99.69 79	99.41 154

test0.0.03 1		96.69 89	98.12 73	95.01 104	95.49 119	98.99 99	95.86 139	90.82 116	98.38 114	92.54 94	96.66 98	97.33 68	95.75 143	97.75 119	98.34 67	99.60 139	99.40 155

testgi		95.67 113	97.48 91	93.56 125	95.07 129	99.00 97	95.33 150	88.47 152	98.80 89	86.90 129	97.30 80	92.33 117	95.97 140	97.66 122	97.91 87	99.60 139	99.38 156

TAMVS		95.53 115	96.50 119	94.39 111	93.86 142	99.03 96	96.67 125	89.55 141	97.33 159	90.64 106	93.02 154	91.58 121	96.21 132	97.72 121	97.43 114	99.43 173	99.36 157

AdaColmap		99.06 20	98.98 43	99.15 5	99.60 19	99.30 80	99.38 27	98.16 16	99.02 68	98.55 4	98.71 43	99.57 48	99.58 11	99.09 29	97.84 91	99.64 117	99.36 157

PM-MVS		89.55 209	90.30 214	88.67 205	87.06 217	95.60 208	90.88 208	84.51 190	96.14 200	75.75 204	86.89 208	63.47 229	94.64 183	96.85 147	93.89 202	99.17 189	99.29 159

pmmvs4		95.09 124	95.90 133	94.14 113	92.29 159	97.70 165	95.45 147	90.31 125	98.60 102	90.70 104	93.25 148	89.90 127	96.67 122	97.13 141	95.42 160	99.44 172	99.28 160

EG-PatchMatch MVS		92.45 180	93.92 176	90.72 188	92.56 156	98.43 137	94.88 161	84.54 189	97.18 163	79.55 184	86.12 212	83.23 176	93.15 203	97.22 138	96.00 146	99.67 95	99.27 161

UA-Net		97.13 73	99.14 30	94.78 106	97.21 72	99.38 67	97.56 99	92.04 94	98.48 111	88.03 121	98.39 56	99.91 26	94.03 194	99.33 20	99.23 16	99.81 26	99.25 162

pmmvs-eth3d		89.81 208	89.65 215	90.00 197	86.94 218	95.38 215	91.08 206	86.39 174	94.57 216	82.27 156	83.03 218	64.94 226	93.96 195	96.57 152	93.82 203	99.35 180	99.24 163

gg-mvs-nofinetune		90.85 203	94.14 163	87.02 209	94.89 133	99.25 86	98.64 55	76.29 226	88.24 227	57.50 231	79.93 222	95.45 87	95.18 178	98.77 50	98.07 79	99.62 129	99.24 163

PMMVS		97.52 60	98.39 58	96.51 80	95.82 111	98.73 118	97.80 94	93.05 90	98.76 96	94.39 67	99.07 26	97.03 73	98.55 70	98.31 80	97.61 103	99.43 173	99.21 165

CNLPA		99.03 23	99.05 36	99.01 14	99.27 38	99.22 90	99.03 44	97.98 26	99.34 28	99.00 2	98.25 59	99.71 42	99.31 29	98.80 47	98.82 38	99.48 166	99.17 166

CR-MVSNet		94.57 136	97.34 95	91.33 170	94.90 132	98.59 126	97.15 114	79.14 215	97.98 129	80.42 169	96.59 103	93.50 111	96.85 117	98.10 94	97.49 109	99.50 165	99.15 167

PatchT		93.96 145	97.36 94	90.00 197	94.76 135	98.65 122	90.11 213	78.57 220	97.96 132	80.42 169	96.07 113	94.10 106	96.85 117	98.10 94	97.49 109	99.26 185	99.15 167

COLMAP_ROB	96.15 12	97.78 52	98.17 70	97.32 47	98.84 45	99.45 60	99.28 30	95.43 42	99.48 15	91.80 99	94.83 135	98.36 60	98.90 60	98.09 96	97.85 90	99.68 88	99.15 167
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
MSDG		98.27 43	98.29 64	98.24 32	99.20 39	99.22 90	99.20 33	97.82 29	99.37 21	94.43 64	95.90 118	97.31 69	99.12 49	98.76 51	98.35 65	99.67 95	99.14 170

test-mter		94.86 128	97.32 96	92.00 153	92.41 158	98.82 108	96.18 136	86.35 175	98.05 126	82.28 155	96.48 105	94.39 99	95.46 162	98.17 91	96.20 142	99.32 182	99.13 171

RPMNet		94.66 131	97.16 101	91.75 162	94.98 130	98.59 126	97.00 121	78.37 221	97.98 129	83.78 142	96.27 109	94.09 107	96.91 114	97.36 133	96.73 126	99.48 166	99.09 172

OMC-MVS		98.84 28	99.01 42	98.65 25	99.39 30	99.23 89	99.22 32	96.70 35	99.40 18	97.77 16	97.89 69	99.80 37	99.21 34	99.02 33	98.65 44	99.57 153	99.07 173

TSAR-MVS + COLMAP		96.79 80	96.55 113	97.06 57	97.70 63	98.46 132	99.07 41	96.23 37	99.38 19	91.32 103	98.80 39	85.61 154	98.69 65	97.64 125	96.92 123	99.37 179	99.06 174

tpm		92.38 185	94.79 151	89.56 200	94.30 137	97.50 184	94.24 191	78.97 218	97.72 146	74.93 209	97.97 67	82.91 182	96.60 125	93.65 209	94.81 192	98.33 201	98.98 175

PatchMatch-RL		97.77 53	98.25 65	97.21 52	99.11 41	99.25 86	97.06 120	94.09 64	98.72 99	95.14 47	98.47 52	96.29 79	98.43 73	98.65 58	97.44 113	99.45 170	98.94 176

pmmvs5		92.71 175	94.27 162	90.90 183	91.42 198	97.74 161	93.23 196	86.66 172	95.99 205	78.96 192	91.45 156	83.44 173	95.55 151	97.30 135	95.05 172	99.58 149	98.93 177

test-LLR		95.50 116	97.32 96	93.37 131	95.49 119	98.74 116	96.44 132	90.82 116	98.18 120	82.75 152	96.60 101	94.67 97	95.54 152	98.09 96	96.00 146	99.20 187	98.93 177

TESTMET0.1,1		94.95 127	97.32 96	92.20 146	92.62 154	98.74 116	96.44 132	86.67 171	98.18 120	82.75 152	96.60 101	94.67 97	95.54 152	98.09 96	96.00 146	99.20 187	98.93 177

EU-MVSNet		92.80 168	94.76 152	90.51 191	91.88 172	96.74 201	92.48 202	88.69 149	96.21 198	79.00 191	91.51 155	87.82 131	91.83 208	95.87 172	96.27 139	99.21 186	98.92 180

Anonymous20231211		83.86 217	83.39 223	84.40 216	85.29 221	93.44 223	86.29 221	84.24 194	85.55 230	68.63 219	61.25 230	59.57 232	84.33 221	92.50 215	92.52 213	97.65 212	98.89 181

PCF-MVS	97.50 6	98.18 45	98.35 61	97.99 37	98.65 49	99.36 70	98.94 47	98.14 20	98.59 103	93.62 82	96.61 100	99.76 41	99.03 57	97.77 117	97.45 112	99.57 153	98.89 181
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
EPNet_dtu		96.30 101	98.53 55	93.70 122	98.97 44	98.24 146	97.36 105	94.23 61	98.85 81	79.18 190	99.19 16	98.47 58	94.09 193	97.89 111	98.21 74	98.39 200	98.85 183
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
OPM-MVS		96.22 103	95.85 136	96.65 76	97.75 61	98.54 129	99.00 46	95.53 40	96.88 177	89.88 114	95.95 117	86.46 147	98.07 86	97.65 124	96.63 129	99.67 95	98.83 184

GG-mvs-BLEND		69.11 227	98.13 72	35.26 233	3.49 239	98.20 148	94.89 158	2.38 237	98.42 113	5.82 242	96.37 108	98.60 55	5.97 237	98.75 53	97.98 84	99.01 191	98.61 185

ambc			80.99 225		80.04 231	90.84 224	90.91 207		96.09 201	74.18 210	62.81 229	30.59 240	82.44 223	96.25 166	91.77 220	95.91 228	98.56 186

MDTV_nov1_ep13_2view		92.44 181	95.66 138	88.68 204	91.05 205	97.92 155	92.17 203	79.64 210	98.83 84	76.20 203	91.45 156	93.51 110	95.04 180	95.68 174	93.70 204	97.96 206	98.53 187

USDC		94.26 139	94.83 150	93.59 124	96.02 102	98.44 135	97.84 91	88.65 150	98.86 79	82.73 154	94.02 140	80.56 204	96.76 119	97.28 136	96.15 145	99.55 157	98.50 188

MDA-MVSNet-bldmvs		87.84 215	89.22 216	86.23 211	81.74 228	96.77 200	83.74 223	89.57 140	94.50 217	72.83 215	96.64 99	64.47 228	92.71 205	81.43 229	92.28 217	96.81 224	98.47 189

gm-plane-assit		89.44 210	92.82 207	85.49 213	91.37 200	95.34 216	79.55 227	82.12 201	91.68 223	64.79 226	87.98 198	80.26 207	95.66 146	98.51 71	97.56 105	99.45 170	98.41 190

MS-PatchMatch		95.99 108	97.26 100	94.51 109	97.46 65	98.76 114	97.27 109	86.97 168	99.09 57	89.83 115	93.51 145	97.78 64	96.18 134	97.53 129	95.71 156	99.35 180	98.41 190

TransMVSNet (Re)		93.45 153	94.08 167	92.72 141	92.83 152	97.62 176	94.94 156	91.54 105	95.65 211	83.06 150	88.93 180	83.53 172	94.25 190	97.41 132	97.03 120	99.67 95	98.40 192

TinyColmap		94.00 143	94.35 161	93.60 123	95.89 107	98.26 144	97.49 102	88.82 147	98.56 106	83.21 148	91.28 158	80.48 206	96.68 121	97.34 134	96.26 141	99.53 162	98.24 193

TDRefinement		93.04 162	93.57 187	92.41 142	96.58 80	98.77 112	97.78 96	91.96 97	98.12 123	80.84 162	89.13 179	79.87 210	87.78 213	96.44 155	94.50 198	99.54 161	98.15 194

MDTV_nov1_ep13		95.57 114	97.48 91	93.35 133	95.43 121	98.97 101	97.19 113	83.72 199	98.92 77	87.91 123	97.75 72	96.12 82	97.88 94	96.84 148	95.64 157	97.96 206	98.10 195

tpmp4_e23		93.84 150	94.58 156	92.98 138	95.41 124	98.29 143	96.81 122	80.57 205	98.15 122	90.53 107	97.00 88	84.39 167	96.91 114	93.69 207	92.45 214	97.67 211	98.06 196

MIMVSNet		94.49 137	97.59 88	90.87 185	91.74 185	98.70 120	94.68 179	78.73 219	97.98 129	83.71 145	97.71 75	94.81 94	96.96 113	97.97 107	97.92 86	99.40 178	98.04 197

CostFormer		94.25 140	94.88 149	93.51 128	95.43 121	98.34 142	96.21 135	80.64 204	97.94 134	94.01 72	98.30 58	86.20 150	97.52 99	92.71 211	92.69 211	97.23 222	98.02 198

RPSCF		97.61 58	98.16 71	96.96 70	98.10 55	99.00 97	98.84 50	93.76 80	99.45 16	94.78 55	99.39 10	99.31 50	98.53 71	96.61 149	95.43 159	97.74 208	97.93 199

testus		88.77 212	92.77 208	84.10 217	88.24 215	93.95 220	87.16 219	84.24 194	97.37 152	61.54 230	95.70 126	73.10 222	84.90 219	95.56 175	95.82 152	98.51 196	97.88 200

Anonymous20231206		90.70 205	93.93 175	86.92 210	90.21 213	96.79 199	90.30 212	86.61 173	96.05 203	69.25 218	88.46 191	84.86 163	85.86 217	97.11 142	96.47 135	99.30 183	97.80 201

DWT-MVSNet_training		95.38 119	95.05 146	95.78 94	95.86 109	98.88 105	97.55 100	90.09 131	98.23 119	96.49 33	97.62 77	86.92 141	97.16 108	92.03 219	94.12 200	97.52 214	97.50 202

pmmvs3		88.19 214	91.27 211	84.60 215	85.60 220	93.66 221	85.68 222	81.13 202	92.36 222	63.66 228	89.51 175	77.10 218	93.22 202	96.37 158	92.40 215	98.30 202	97.46 203

test2356		88.81 211	92.86 204	84.09 218	87.85 216	93.46 222	87.07 220	83.60 200	96.50 195	62.08 229	97.06 87	75.04 220	85.17 218	95.08 198	95.42 160	98.75 195	97.46 203

N_pmnet		92.21 192	94.60 154	89.42 201	91.88 172	97.38 191	89.15 215	89.74 139	97.89 137	73.75 211	87.94 199	92.23 118	93.85 198	96.10 168	93.20 207	98.15 204	97.43 205

PatchmatchNet		94.70 130	97.08 103	91.92 156	95.53 117	98.85 107	95.77 140	79.54 212	98.95 71	85.98 133	98.52 48	96.45 75	97.39 104	95.32 177	94.09 201	97.32 218	97.38 206
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
ADS-MVSNet		94.65 132	97.04 104	91.88 159	95.68 114	98.99 99	95.89 138	79.03 217	99.15 50	85.81 135	96.96 90	98.21 62	97.10 109	94.48 204	94.24 199	97.74 208	97.21 207

MVS-HIRNet		92.51 179	95.97 131	88.48 206	93.73 146	98.37 141	90.33 211	75.36 229	98.32 115	77.78 196	89.15 178	94.87 92	95.14 179	97.62 126	96.39 136	98.51 196	97.11 208

dps		94.63 133	95.31 145	93.84 117	95.53 117	98.71 119	96.54 128	80.12 207	97.81 145	97.21 23	96.98 89	92.37 116	96.34 131	92.46 216	91.77 220	97.26 220	97.08 209

test20.03		90.65 206	93.71 181	87.09 208	90.44 211	96.24 202	89.74 214	85.46 179	95.59 212	72.99 214	90.68 161	85.33 157	84.41 220	95.94 171	95.10 171	99.52 163	97.06 210

EPMVS		95.05 125	96.86 108	92.94 139	95.84 110	98.96 102	96.68 124	79.87 208	99.05 65	90.15 111	97.12 86	95.99 83	97.49 101	95.17 185	94.75 194	97.59 213	96.96 211

testpf		91.80 200	94.43 160	88.74 203	93.89 141	95.30 217	92.05 204	71.77 230	97.52 150	87.24 126	94.77 136	92.68 115	91.48 209	91.75 222	92.11 219	96.02 227	96.89 212

LP		92.12 195	94.60 154	89.22 202	94.96 131	98.45 134	93.01 198	77.58 222	97.85 141	77.26 199	89.80 173	93.00 114	94.54 184	93.69 207	92.58 212	98.00 205	96.83 213

tpmrst		93.86 148	95.88 134	91.50 166	95.69 113	98.62 124	95.64 143	79.41 213	98.80 89	83.76 144	95.63 127	96.13 81	97.25 105	92.92 210	92.31 216	97.27 219	96.74 214

new-patchmatchnet		86.12 216	87.30 217	84.74 214	86.92 219	95.19 219	83.57 224	84.42 192	92.67 221	65.66 223	80.32 221	64.72 227	89.41 212	92.33 218	89.21 224	98.43 199	96.69 215

tpm cat1		94.06 141	94.90 148	93.06 136	95.42 123	98.52 130	96.64 126	80.67 203	97.82 143	92.63 93	93.39 147	95.00 91	96.06 138	91.36 223	91.58 222	96.98 223	96.66 216

FMVSNet5		95.42 117	96.47 120	94.20 112	92.26 160	95.99 204	95.66 142	87.15 165	97.87 138	93.46 84	96.68 97	93.79 108	97.52 99	97.10 143	97.21 118	99.11 190	96.62 217

DeepMVS_CX						96.85 198	87.43 218	89.27 142	98.30 116	75.55 206	95.05 132	79.47 211	92.62 206	89.48 225		95.18 229	95.96 218

MIMVSNet1		88.61 213	90.68 213	86.19 212	81.56 229	95.30 217	87.78 217	85.98 177	94.19 218	72.30 216	78.84 223	78.90 215	90.06 211	96.59 150	95.47 158	99.46 169	95.49 219

CMPMVS	70.31 18	90.74 204	91.06 212	90.36 194	97.32 68	97.43 188	92.97 199	87.82 162	93.50 219	75.34 208	83.27 217	84.90 162	92.19 207	92.64 214	91.21 223	96.50 225	94.46 220
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
new_pmnet		90.45 207	92.84 206	87.66 207	88.96 214	96.16 203	88.71 216	84.66 188	97.56 149	71.91 217	85.60 213	86.58 146	93.28 201	96.07 169	93.54 205	98.46 198	94.39 221

Gipuma		81.40 222	81.78 224	80.96 221	83.21 224	85.61 233	79.73 226	76.25 227	97.33 159	64.21 227	55.32 231	55.55 234	86.04 216	92.43 217	92.20 218	96.32 226	93.99 222
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
testmv		81.83 220	86.26 218	76.66 223	84.10 222	89.42 228	74.29 231	79.65 209	90.61 224	51.85 235	82.11 219	63.06 231	72.61 227	91.94 220	92.75 209	97.49 215	93.94 223

test1235678		81.83 220	86.26 218	76.66 223	84.10 222	89.41 229	74.29 231	79.64 210	90.60 225	51.84 236	82.11 219	63.07 230	72.61 227	91.94 220	92.75 209	97.49 215	93.94 223

PMMVS2		77.26 224	79.47 226	74.70 227	76.00 232	88.37 231	74.22 233	76.34 225	78.31 232	54.13 232	69.96 228	52.50 235	70.14 230	84.83 227	88.71 225	97.35 217	93.58 225

test12356		80.53 223	84.80 221	75.54 225	82.31 225	88.05 232	75.99 228	79.31 214	88.53 226	53.24 234	83.30 216	56.38 233	65.16 233	90.87 224	93.10 208	97.25 221	93.34 226

1111		82.87 219	85.67 220	79.62 222	81.86 226	89.62 226	74.44 229	68.81 232	87.44 228	66.59 221	76.83 225	70.33 224	87.71 214	92.65 212	93.37 206	98.28 203	89.42 227

MVE	67.97 19	65.53 231	67.43 232	63.31 232	59.33 236	74.20 235	53.09 238	70.43 231	66.27 235	43.13 237	45.98 236	30.62 239	70.65 229	79.34 231	86.30 226	83.25 236	89.33 228
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
FPMVS		83.82 218	84.61 222	82.90 219	90.39 212	90.71 225	90.85 209	84.10 197	95.47 213	65.15 224	83.44 215	74.46 221	75.48 224	81.63 228	79.42 230	91.42 231	87.14 229

no-one		66.79 230	67.62 231	65.81 231	73.06 235	81.79 234	51.90 239	76.20 228	61.07 236	54.05 233	51.62 235	41.72 237	49.18 234	67.26 233	82.83 228	90.47 232	87.07 230

EMVS		68.12 229	68.11 230	68.14 229	75.51 233	71.76 236	55.38 237	77.20 224	77.78 233	37.79 239	53.59 232	43.61 236	74.72 225	67.05 234	76.70 232	88.27 235	86.24 231

E-PMN		68.30 228	68.43 229	68.15 228	74.70 234	71.56 237	55.64 236	77.24 223	77.48 234	39.46 238	51.95 234	41.68 238	73.28 226	70.65 232	79.51 229	88.61 234	86.20 232

PMVS	72.60 17	76.39 225	77.66 227	74.92 226	81.04 230	69.37 238	68.47 234	80.54 206	85.39 231	65.07 225	73.52 227	72.91 223	65.67 232	80.35 230	76.81 231	88.71 233	85.25 233
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
.test1245		69.67 226	72.22 228	66.70 230	81.86 226	89.62 226	74.44 229	68.81 232	87.44 228	66.59 221	76.83 225	70.33 224	87.71 214	92.65 212	37.65 233	20.79 237	51.04 234

testmvs		31.24 232	40.15 233	20.86 234	12.61 237	17.99 239	25.16 240	13.30 235	48.42 237	24.82 240	53.07 233	30.13 241	28.47 235	42.73 235	37.65 233	20.79 237	51.04 234

test123		26.75 233	34.25 234	18.01 235	7.93 238	17.18 240	24.85 241	12.36 236	44.83 238	16.52 241	41.80 237	18.10 242	28.29 236	33.08 236	34.79 235	18.10 239	49.95 236

sosnet-low-res		0.00 234	0.00 235	0.00 236	0.00 240	0.00 241	0.00 242	0.00 238	0.00 239	0.00 243	0.00 238	0.00 243	0.00 238	0.00 237	0.00 236	0.00 240	0.00 237

sosnet		0.00 234	0.00 235	0.00 236	0.00 240	0.00 241	0.00 242	0.00 238	0.00 239	0.00 243	0.00 238	0.00 243	0.00 238	0.00 237	0.00 236	0.00 240	0.00 237

MTAPA										98.09 10		99.97 3

MTMP										98.46 7		99.96 9

Patchmatch-RL test							66.86 235

tmp_tt				82.25 220	97.73 62	88.71 230	80.18 225	68.65 234	99.15 50	86.98 128	99.47 7	85.31 158	68.35 231	87.51 226	83.81 227	91.64 230

XVS					97.42 66	99.62 33	98.59 57			93.81 79		99.95 14				99.69 79

X-MVStestdata					97.42 66	99.62 33	98.59 57			93.81 79		99.95 14				99.69 79

mPP-MVS					99.53 24							99.89 29

NP-MVS									98.57 105

Patchmtry						98.59 126	97.15 114	79.14 215		80.42 169

High-resolution multi-view benchmark