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 55	98.71 52	96.63 78	95.68 115	99.56 49	97.51 102	93.10 90	99.22 44	94.99 52	97.18 86	97.30 71	98.65 67	98.83 46	98.93 30	99.84 5	99.92 1

LTVRE_ROB	93.20 16	92.84 166	94.92 148	90.43 194	92.83 153	98.63 124	97.08 120	87.87 162	97.91 137	68.42 221	93.54 145	79.46 213	96.62 125	97.55 129	97.40 116	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 70	97.81 83	96.72 74	96.20 97	99.45 61	98.21 78	91.60 103	99.22 44	95.39 43	98.48 52	90.95 124	99.16 44	97.66 123	99.05 24	99.76 44	99.90 3

v748		91.12 203	91.95 211	90.16 196	90.60 210	97.35 193	91.11 206	87.92 161	94.75 215	80.54 168	86.26 212	75.97 220	91.13 211	94.63 204	94.81 193	99.65 107	99.90 3

PVSNet_Blended_VisFu		97.41 65	98.49 57	96.15 87	97.49 65	99.76 1	96.02 138	93.75 82	99.26 39	93.38 86	93.73 144	99.35 50	96.47 130	98.96 36	98.46 55	99.77 42	99.90 3

CSCG		98.90 27	98.93 46	98.85 21	99.75 3	99.72 4	99.49 18	96.58 37	99.38 20	98.05 11	98.97 29	97.87 64	99.49 18	97.78 117	98.92 31	99.78 37	99.90 3

MVS_0304		98.14 47	99.03 41	97.10 55	98.05 59	99.63 30	99.27 32	94.33 59	99.63 6	93.06 90	97.32 80	99.05 54	98.09 86	98.82 47	98.87 35	99.81 26	99.89 7

PS-CasMVS		92.72 174	93.36 194	91.98 155	91.62 194	97.52 183	94.13 193	88.98 146	95.94 207	81.51 161	87.35 203	79.95 210	95.91 142	96.37 159	96.49 135	99.70 78	99.89 7

CP-MVSNet		93.25 157	94.00 171	92.38 144	91.65 192	97.56 180	94.38 189	89.20 144	96.05 204	83.16 150	89.51 176	81.97 198	96.16 137	96.43 157	96.56 133	99.71 69	99.89 7

WR-MVS_H		93.54 153	94.67 154	92.22 145	91.95 171	97.91 157	94.58 186	88.75 149	96.64 188	83.88 142	90.66 163	85.13 161	94.40 189	96.54 155	95.91 152	99.73 55	99.89 7

FC-MVSNet-train		97.04 75	97.91 82	96.03 92	96.00 105	98.41 139	96.53 131	93.42 85	99.04 68	93.02 91	98.03 66	94.32 101	97.47 103	97.93 110	97.77 99	99.75 45	99.88 11

IterMVS-LS		96.12 107	97.48 92	94.53 109	95.19 128	97.56 180	97.15 115	89.19 145	99.08 60	88.23 120	94.97 134	94.73 97	97.84 96	97.86 114	98.26 74	99.60 140	99.88 11
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
v7n		91.61 202	92.95 203	90.04 197	90.56 211	97.69 170	93.74 196	85.59 179	95.89 208	76.95 201	86.60 210	78.60 217	93.76 200	97.01 145	94.99 184	99.65 107	99.87 13

CHOSEN 1792x2688		96.41 97	96.99 106	95.74 98	98.01 60	99.72 4	97.70 98	90.78 119	99.13 57	90.03 114	87.35 203	95.36 89	98.33 78	98.59 68	98.91 33	99.59 146	99.87 13

CANet		98.46 39	99.16 30	97.64 44	98.48 52	99.64 25	99.35 29	94.71 52	99.53 12	95.17 47	97.63 77	99.59 47	98.38 76	98.88 44	98.99 27	99.74 49	99.86 15

diffmvs		97.50 64	98.63 53	96.18 85	95.88 109	99.26 86	98.19 80	91.08 114	99.36 24	94.32 70	98.24 61	96.83 75	98.22 82	98.45 74	98.42 57	99.42 176	99.86 15

HyFIR lowres test		95.99 109	96.56 113	95.32 103	97.99 61	99.65 20	96.54 129	88.86 147	98.44 113	89.77 117	84.14 215	97.05 73	99.03 58	98.55 70	98.19 77	99.73 55	99.86 15

v52		91.94 198	93.10 201	90.57 190	90.62 209	97.50 185	93.98 194	87.02 167	95.86 209	77.67 198	86.93 207	82.16 196	94.53 186	94.71 203	94.70 196	99.61 133	99.85 18

V4		91.92 199	93.10 201	90.55 191	90.64 208	97.51 184	93.93 195	87.02 167	95.81 211	77.61 199	86.93 207	82.19 195	94.50 187	94.72 202	94.68 197	99.62 130	99.85 18

tfpnnormal		93.85 150	94.12 166	93.54 128	93.22 152	98.24 147	95.45 148	91.96 98	94.61 216	83.91 141	90.74 161	81.75 201	97.04 111	97.49 131	96.16 145	99.68 89	99.84 20

Effi-MVS+		95.81 111	97.31 100	94.06 115	95.09 129	99.35 73	97.24 112	88.22 156	98.54 108	85.38 139	98.52 49	88.68 130	98.70 65	98.32 80	97.93 86	99.74 49	99.84 20

SD-MVS		99.25 9	99.50 7	98.96 16	98.79 48	99.55 51	99.33 30	98.29 9	99.75 1	97.96 13	99.15 20	99.95 15	99.61 6	99.17 26	99.06 23	99.81 26	99.84 20

EPNet		98.05 48	98.86 48	97.10 55	99.02 44	99.43 64	98.47 61	94.73 51	99.05 66	95.62 38	98.93 31	97.62 68	95.48 157	98.59 68	98.55 51	99.29 185	99.84 20
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
SteuartSystems-ACMMP		99.20 13	99.51 6	98.83 23	99.66 14	99.66 19	99.71 4	98.12 23	99.14 53	96.62 30	99.16 19	99.98 1	99.12 50	99.63 3	99.19 20	99.78 37	99.83 24
Skip Steuart: Steuart Systems R&D Blog.
TSAR-MVS + MP.		99.27 7	99.57 3	98.92 19	98.78 49	99.53 52	99.72 2	98.11 24	99.73 2	97.43 20	99.15 20	99.96 10	99.59 9	99.73 1	99.07 22	99.88 1	99.82 25

anonymousdsp		93.12 158	95.86 136	89.93 200	91.09 205	98.25 146	95.12 152	85.08 184	97.44 152	73.30 213	90.89 160	90.78 125	95.25 178	97.91 111	95.96 151	99.71 69	99.82 25

TSAR-MVS + ACMM		98.77 30	99.45 9	97.98 39	99.37 32	99.46 59	99.44 24	98.13 22	99.65 4	92.30 96	98.91 36	99.95 15	99.05 56	99.42 15	98.95 29	99.58 150	99.82 25

PEN-MVS		92.72 174	93.20 200	92.15 148	91.29 202	97.31 194	94.67 181	89.81 137	96.19 200	81.83 159	88.58 190	79.06 215	95.61 151	95.21 183	96.27 140	99.72 60	99.82 25

WR-MVS		93.43 156	94.48 159	92.21 146	91.52 197	97.69 170	94.66 182	89.98 134	96.86 179	83.43 147	90.12 165	85.03 162	93.94 197	96.02 171	95.82 153	99.71 69	99.82 25

HSP-MVS		99.31 3	99.43 14	99.17 2	99.68 11	99.75 2	99.72 2	98.31 5	99.45 17	98.16 9	99.28 13	99.98 1	99.30 31	99.34 20	98.41 59	99.81 26	99.81 30

DeepC-MVS	97.63 4	98.33 43	98.57 54	98.04 37	98.62 51	99.65 20	99.45 22	98.15 19	99.51 15	92.80 93	95.74 124	96.44 78	99.46 20	99.37 17	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		96.34 100	96.47 121	96.17 86	96.16 98	99.71 8	97.82 93	93.46 84	98.10 125	90.69 106	96.75 95	85.26 160	99.11 52	98.05 104	97.65 101	99.82 13	99.80 32

v8		92.87 165	93.87 179	91.72 165	92.05 167	97.50 185	94.79 170	88.20 157	96.85 180	80.11 176	90.01 168	82.86 185	95.48 157	95.15 194	94.90 190	99.66 101	99.80 32

v10		92.79 171	94.06 169	91.31 173	91.78 181	97.29 196	94.87 163	86.10 177	96.97 173	79.82 179	88.16 195	84.56 166	95.63 148	96.33 162	95.31 165	99.65 107	99.80 32

UniMVSNet_NR-MVSNet		94.59 135	95.47 143	93.55 127	91.85 175	97.89 158	95.03 154	92.00 96	97.33 160	86.12 132	93.19 150	87.29 134	96.60 126	96.12 168	96.70 128	99.72 60	99.80 32

DU-MVS		93.98 145	94.44 160	93.44 130	91.66 190	97.77 160	95.03 154	91.57 104	97.17 165	86.12 132	93.13 152	81.13 203	96.60 126	95.10 197	97.01 123	99.67 96	99.80 32

UGNet		97.66 58	99.07 36	96.01 93	97.19 74	99.65 20	97.09 119	93.39 86	99.35 26	94.40 67	98.79 41	99.59 47	94.24 192	98.04 106	98.29 73	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 51	98.86 48	96.68 75	96.02 103	99.72 4	98.35 74	93.37 88	98.75 99	94.01 73	96.88 94	98.40 60	98.48 73	99.09 30	99.42 5	99.83 9	99.80 32

UniMVSNet (Re)		94.58 136	95.34 144	93.71 122	92.25 162	98.08 151	94.97 156	91.29 113	97.03 171	87.94 123	93.97 143	86.25 150	96.07 138	96.27 165	95.97 150	99.72 60	99.79 39

DELS-MVS		98.19 45	98.77 51	97.52 46	98.29 55	99.71 8	99.12 38	94.58 57	98.80 90	95.38 44	96.24 111	98.24 62	97.92 91	99.06 33	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
SMA-MVS		99.30 5	99.62 2	98.93 17	99.76 2	99.64 25	99.44 24	98.21 14	99.53 12	96.79 29	99.41 9	99.98 1	99.67 4	99.63 3	99.37 9	99.71 69	99.78 41

tfpn		96.22 104	95.62 141	96.93 72	96.29 91	99.72 4	98.34 75	93.94 76	97.96 133	93.94 75	96.45 107	79.09 214	99.22 33	98.28 82	98.06 81	99.83 9	99.78 41

v144192		92.38 186	93.55 191	91.00 181	91.44 198	97.47 188	94.27 190	87.41 165	96.52 194	78.03 195	87.50 202	82.65 187	95.32 173	95.82 174	95.15 170	99.55 158	99.78 41

V42		93.05 162	93.90 178	92.04 151	91.91 172	97.66 172	94.91 158	89.91 135	96.85 180	80.58 167	89.66 175	83.43 175	95.37 169	95.03 200	94.90 190	99.59 146	99.78 41

MVS_Test		97.30 71	98.54 55	95.87 94	95.74 113	99.28 84	98.19 80	91.40 108	99.18 50	91.59 102	98.17 62	96.18 81	98.63 69	98.61 64	98.55 51	99.66 101	99.78 41

TranMVSNet+NR-MVSNet		93.67 152	94.14 164	93.13 136	91.28 204	97.58 179	95.60 145	91.97 97	97.06 169	84.05 140	90.64 164	82.22 194	96.17 136	94.94 201	96.78 126	99.69 80	99.78 41

PVSNet_BlendedMVS		97.51 62	97.71 85	97.28 50	98.06 57	99.61 38	97.31 108	95.02 46	99.08 60	95.51 41	98.05 64	90.11 126	98.07 87	98.91 41	98.40 60	99.72 60	99.78 41

PVSNet_Blended		97.51 62	97.71 85	97.28 50	98.06 57	99.61 38	97.31 108	95.02 46	99.08 60	95.51 41	98.05 64	90.11 126	98.07 87	98.91 41	98.40 60	99.72 60	99.78 41

Fast-Effi-MVS+		95.38 120	96.52 116	94.05 116	94.15 139	99.14 96	97.24 112	86.79 170	98.53 109	87.62 126	94.51 139	87.06 135	98.76 63	98.60 67	98.04 83	99.72 60	99.77 49

v7		92.97 164	94.11 167	91.65 166	91.83 176	97.55 182	94.86 166	88.19 158	96.96 174	79.72 182	88.16 195	84.68 165	95.63 148	96.33 162	95.30 166	99.65 107	99.77 49

APDe-MVS		99.49 1	99.64 1	99.32 1	99.74 4	99.74 3	99.75 1	98.34 2	99.56 9	98.72 3	99.57 4	99.97 4	99.53 15	99.65 2	99.25 14	99.84 5	99.77 49

ACMMPR		99.30 5	99.54 4	99.03 12	99.66 14	99.64 25	99.68 5	98.25 12	99.56 9	97.12 25	99.19 17	99.95 15	99.72 1	99.43 14	99.25 14	99.72 60	99.77 49

view800		96.70 88	96.45 124	96.99 70	96.29 91	99.69 11	98.39 71	93.95 75	97.92 136	94.25 72	96.23 112	85.57 156	99.22 33	98.28 82	97.71 100	99.82 13	99.76 53

v1144		92.81 167	94.03 170	91.40 170	91.68 189	97.60 178	94.73 177	88.40 154	96.71 183	78.48 194	88.14 197	84.46 167	95.45 164	96.31 164	95.22 168	99.65 107	99.76 53

HFP-MVS		99.32 2	99.53 5	99.07 9	99.69 8	99.59 44	99.63 10	98.31 5	99.56 9	97.37 21	99.27 14	99.97 4	99.70 3	99.35 19	99.24 16	99.71 69	99.76 53

v1neww		93.06 160	93.94 174	92.03 152	91.99 169	97.70 166	94.79 170	90.14 130	96.93 176	80.13 174	89.97 170	83.01 180	95.48 157	95.16 190	95.01 179	99.63 124	99.76 53

v7new		93.06 160	93.94 174	92.03 152	91.99 169	97.70 166	94.79 170	90.14 130	96.93 176	80.13 174	89.97 170	83.01 180	95.48 157	95.16 190	95.01 179	99.63 124	99.76 53

v6		93.11 159	93.98 172	92.10 149	92.01 168	97.71 163	94.86 166	90.15 129	96.96 174	80.47 169	90.01 168	83.26 176	95.48 157	95.17 186	95.01 179	99.64 118	99.76 53

MSLP-MVS++		99.15 15	99.24 27	99.04 11	99.52 27	99.49 56	99.09 41	98.07 25	99.37 22	98.47 5	97.79 71	99.89 30	99.50 16	98.93 39	99.45 4	99.61 133	99.76 53

ACMMP		98.74 31	99.03 41	98.40 29	99.36 34	99.64 25	99.20 34	97.75 32	98.82 87	95.24 46	98.85 39	99.87 32	99.17 42	98.74 55	97.50 109	99.71 69	99.76 53
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	95.27 124	95.96 133	94.45 111	96.83 79	98.78 112	94.72 178	91.67 102	98.95 72	86.82 131	96.42 108	83.67 172	97.00 112	97.48 132	96.68 129	99.69 80	99.76 53
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
X-MVS		98.93 26	99.37 19	98.42 28	99.67 12	99.62 34	99.60 12	98.15 19	99.08 60	93.81 80	98.46 54	99.95 15	99.59 9	99.49 11	99.21 19	99.68 89	99.75 62

NR-MVSNet		94.01 143	94.51 158	93.44 130	92.56 157	97.77 160	95.67 142	91.57 104	97.17 165	85.84 135	93.13 152	80.53 206	95.29 176	97.01 145	96.17 144	99.69 80	99.75 62

Vis-MVSNet (Re-imp)		97.40 66	98.89 47	95.66 100	95.99 106	99.62 34	97.82 93	93.22 89	98.82 87	91.40 103	96.94 92	98.56 58	95.70 146	99.14 28	99.41 6	99.79 34	99.75 62

ACMMP_Plus		99.05 22	99.45 9	98.58 27	99.73 5	99.60 42	99.64 8	98.28 10	99.23 43	94.57 58	99.35 12	99.97 4	99.55 13	99.63 3	98.66 44	99.70 78	99.74 65

v1192		92.43 183	93.61 185	91.05 178	91.53 196	97.43 189	94.61 184	87.99 160	96.60 189	76.72 202	87.11 205	82.74 186	95.85 143	96.35 161	95.30 166	99.60 140	99.74 65

PGM-MVS		98.86 28	99.35 23	98.29 31	99.77 1	99.63 30	99.67 6	95.63 40	98.66 102	95.27 45	99.11 23	99.82 37	99.67 4	99.33 21	99.19 20	99.73 55	99.74 65

CP-MVS		99.27 7	99.44 12	99.08 8	99.62 18	99.58 46	99.53 15	98.16 17	99.21 46	97.79 15	99.15 20	99.96 10	99.59 9	99.54 8	98.86 36	99.78 37	99.74 65

view600		96.70 88	96.44 126	97.01 66	96.28 94	99.67 13	98.42 63	93.99 68	97.87 139	94.34 69	95.99 116	85.94 152	99.20 36	98.26 84	97.64 102	99.82 13	99.73 69

zzz-MVS		99.31 3	99.44 12	99.16 4	99.73 5	99.65 20	99.63 10	98.26 11	99.27 36	98.01 12	99.27 14	99.97 4	99.60 7	99.59 6	98.58 50	99.71 69	99.73 69

v1921920		92.36 188	93.57 188	90.94 183	91.39 200	97.39 191	94.70 179	87.63 164	96.60 189	76.63 203	86.98 206	82.89 184	95.75 144	96.26 166	95.14 171	99.55 158	99.73 69

DI_MVS_plusplus_trai		96.90 80	97.49 91	96.21 84	95.61 117	99.40 67	98.72 55	92.11 93	99.14 53	92.98 92	93.08 154	95.14 91	98.13 85	98.05 104	97.91 88	99.74 49	99.73 69

v1240		91.99 197	93.33 195	90.44 193	91.29 202	97.30 195	94.25 191	86.79 170	96.43 198	75.49 208	86.34 211	81.85 200	95.29 176	96.42 158	95.22 168	99.52 164	99.73 69

v1		92.81 167	93.57 188	91.94 156	91.79 180	97.70 166	94.80 169	90.32 124	96.52 194	79.75 180	88.47 191	82.46 191	95.32 173	95.14 196	94.96 186	99.63 124	99.73 69

thres600view7		96.69 90	96.43 128	97.00 68	96.28 94	99.67 13	98.41 64	93.99 68	97.85 142	94.29 71	95.96 117	85.91 153	99.19 38	98.26 84	97.63 103	99.82 13	99.73 69

MP-MVS		99.07 20	99.36 20	98.74 24	99.63 17	99.57 48	99.66 7	98.25 12	99.00 70	95.62 38	98.97 29	99.94 23	99.54 14	99.51 10	98.79 41	99.71 69	99.73 69

DTE-MVSNet		92.42 185	92.85 206	91.91 158	90.87 207	96.97 198	94.53 188	89.81 137	95.86 209	81.59 160	88.83 183	77.88 218	95.01 182	94.34 206	96.35 138	99.64 118	99.73 69

Baseline_NR-MVSNet		93.87 148	93.98 172	93.75 120	91.66 190	97.02 197	95.53 146	91.52 107	97.16 167	87.77 125	87.93 201	83.69 171	96.35 131	95.10 197	97.23 118	99.68 89	99.73 69

SixPastTwentyTwo		93.44 155	95.32 145	91.24 175	92.11 165	98.40 140	92.77 201	88.64 152	98.09 126	77.83 196	93.51 146	85.74 154	96.52 129	96.91 147	94.89 192	99.59 146	99.73 69

LGP-MVS_train		96.23 103	96.89 108	95.46 102	97.32 69	98.77 113	98.81 52	93.60 83	98.58 105	85.52 137	99.08 26	86.67 145	97.83 97	97.87 113	97.51 108	99.69 80	99.73 69

v1141		92.79 171	93.61 185	91.84 162	91.75 183	97.71 163	94.74 176	90.33 123	96.58 191	79.21 190	88.59 188	82.53 190	95.36 170	95.16 190	94.96 186	99.63 124	99.72 81

pm-mvs1		94.27 139	95.57 142	92.75 141	92.58 156	98.13 150	94.87 163	90.71 120	96.70 184	83.78 143	89.94 173	89.85 129	94.96 183	97.58 128	97.07 120	99.61 133	99.72 81

divwei89l23v2f112		92.80 169	93.60 187	91.86 161	91.75 183	97.71 163	94.75 175	90.32 124	96.54 193	79.35 187	88.59 188	82.55 189	95.35 171	95.15 194	94.96 186	99.63 124	99.72 81

IterMVS		94.81 130	97.71 85	91.42 169	94.83 135	97.63 174	97.38 105	85.08 184	98.93 77	75.67 206	94.02 141	97.64 66	96.66 124	98.45 74	97.60 105	98.90 194	99.72 81
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
MCST-MVS		99.11 17	99.27 26	98.93 17	99.67 12	99.33 77	99.51 17	98.31 5	99.28 34	96.57 32	99.10 24	99.90 28	99.71 2	99.19 25	98.35 66	99.82 13	99.71 85

ACMP	96.25 10	96.62 95	96.72 110	96.50 82	96.96 78	98.75 116	97.80 95	94.30 60	98.85 82	93.12 89	98.78 42	86.61 146	97.23 108	97.73 121	96.61 131	99.62 130	99.71 85
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
tfpn111		96.96 79	96.91 107	97.03 60	96.31 84	99.67 13	98.41 64	93.99 68	97.35 154	94.50 61	98.65 46	86.93 138	99.14 45	98.26 84	97.80 94	99.82 13	99.70 87

tfpn1000		97.60 60	98.21 69	96.89 73	96.32 83	99.60 42	97.99 90	93.85 78	99.21 46	95.03 50	98.49 51	93.69 110	98.31 79	98.50 73	98.31 72	99.86 2	99.70 87

conf200view11		96.75 83	96.51 117	97.03 60	96.31 84	99.67 13	98.41 64	93.99 68	97.35 154	94.50 61	95.90 119	86.93 138	99.14 45	98.26 84	97.80 94	99.82 13	99.70 87

tfpn200view9		96.75 83	96.51 117	97.03 60	96.31 84	99.67 13	98.41 64	93.99 68	97.35 154	94.52 59	95.90 119	86.93 138	99.14 45	98.26 84	97.80 94	99.82 13	99.70 87

v11		92.43 183	93.77 181	90.85 187	91.72 187	95.58 213	94.87 163	84.07 199	96.98 172	79.28 188	88.03 198	84.22 169	95.53 155	96.55 154	95.36 163	99.65 107	99.70 87

thres400		96.71 87	96.45 124	97.02 64	96.28 94	99.63 30	98.41 64	94.00 67	97.82 144	94.42 66	95.74 124	86.26 149	99.18 40	98.20 91	97.79 98	99.81 26	99.70 87

conf0.01		96.35 99	95.71 138	97.10 55	96.30 90	99.65 20	98.41 64	94.10 64	97.35 154	94.82 55	95.44 132	81.88 199	99.14 45	98.16 93	97.80 94	99.82 13	99.69 93

conf0.002		96.31 101	95.63 140	97.11 54	96.29 91	99.64 25	98.41 64	94.11 63	97.35 154	94.86 53	95.49 131	81.06 204	99.14 45	98.14 94	98.02 84	99.82 13	99.69 93

v148		92.36 188	92.88 204	91.75 163	91.63 193	97.66 172	92.64 202	90.55 122	96.09 202	83.34 148	88.19 194	80.00 209	92.74 205	93.98 207	94.58 198	99.58 150	99.69 93

v2v482		92.77 173	93.52 192	91.90 159	91.59 195	97.63 174	94.57 187	90.31 126	96.80 182	79.22 189	88.74 185	81.55 202	96.04 140	95.26 179	94.97 185	99.66 101	99.69 93

CPTT-MVS		99.14 16	99.20 29	99.06 10	99.58 21	99.53 52	99.45 22	97.80 31	99.19 49	98.32 8	98.58 48	99.95 15	99.60 7	99.28 23	98.20 76	99.64 118	99.69 93

FMVSNet1		95.77 112	96.41 129	95.03 104	93.42 151	97.86 159	97.11 118	89.89 136	98.53 109	92.00 98	89.17 178	93.23 114	98.15 84	98.07 100	98.34 68	99.61 133	99.69 93

Vis-MVSNet		96.16 106	98.22 68	93.75 120	95.33 126	99.70 10	97.27 110	90.85 116	98.30 117	85.51 138	95.72 126	96.45 76	93.69 201	98.70 57	99.00 26	99.84 5	99.69 93
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
v17		92.55 179	93.65 184	91.27 174	92.11 165	95.63 207	94.89 159	85.15 182	97.12 168	80.39 173	90.02 167	83.02 179	95.45 164	95.17 186	94.92 189	99.66 101	99.68 100

v16		92.66 177	93.80 180	91.32 172	92.13 163	95.62 208	94.89 159	85.12 183	97.20 163	80.66 165	89.96 172	83.93 170	95.49 156	95.17 186	95.04 174	99.63 124	99.68 100

v13		92.16 195	93.28 199	90.85 187	91.75 183	95.58 213	94.65 183	84.23 197	96.49 197	79.51 186	88.40 193	82.58 188	95.31 175	95.21 183	95.03 176	99.66 101	99.68 100

v12		92.18 194	93.29 198	90.88 185	91.70 188	95.59 211	94.61 184	84.36 194	96.65 187	79.59 184	88.85 182	82.03 197	95.35 171	95.22 180	95.04 174	99.65 107	99.68 100

MVSTER		97.16 73	97.71 85	96.52 80	95.97 107	98.48 132	98.63 57	92.10 94	98.68 101	95.96 37	99.23 16	91.79 121	96.87 117	98.76 52	97.37 117	99.57 154	99.68 100

ESAPD		99.23 11	99.41 16	99.01 14	99.70 7	99.69 11	99.40 27	98.31 5	98.94 75	97.70 17	99.40 10	99.97 4	99.17 42	99.54 8	98.67 43	99.78 37	99.67 105

v18		92.63 178	93.67 183	91.43 168	92.13 163	95.65 206	95.09 153	85.44 181	97.06 169	80.78 164	90.06 166	83.06 178	95.47 162	95.16 190	95.01 179	99.64 118	99.67 105

v15		92.27 191	93.33 195	91.04 179	91.83 176	95.60 209	94.79 170	84.88 187	96.66 186	79.66 183	88.72 186	82.45 192	95.40 167	95.19 185	95.00 183	99.65 107	99.67 105

V14		92.31 190	93.41 193	91.03 180	91.80 179	95.59 211	94.79 170	84.70 188	96.58 191	79.83 178	88.79 184	82.98 182	95.41 166	95.22 180	95.02 178	99.65 107	99.67 105

V9		92.24 192	93.32 197	90.98 182	91.76 182	95.58 213	94.83 168	84.50 192	96.68 185	79.73 181	88.66 187	82.39 193	95.39 168	95.22 180	95.03 176	99.65 107	99.67 105

GBi-Net		96.98 77	98.00 79	95.78 95	93.81 144	97.98 152	98.09 83	91.32 109	98.80 90	93.92 76	97.21 83	95.94 85	97.89 92	98.07 100	98.34 68	99.68 89	99.67 105

test1		96.98 77	98.00 79	95.78 95	93.81 144	97.98 152	98.09 83	91.32 109	98.80 90	93.92 76	97.21 83	95.94 85	97.89 92	98.07 100	98.34 68	99.68 89	99.67 105

FMVSNet2		96.64 93	97.50 90	95.63 101	93.81 144	97.98 152	98.09 83	90.87 115	98.99 71	93.48 84	93.17 151	95.25 90	97.89 92	98.63 63	98.80 40	99.68 89	99.67 105

3Dnovator+	96.92 7	98.71 33	99.05 37	98.32 30	99.53 25	99.34 75	99.06 43	94.61 54	99.65 4	97.49 19	96.75 95	99.86 33	99.44 22	98.78 50	99.30 12	99.81 26	99.67 105

HPM-MVS++		99.10 18	99.30 24	98.86 20	99.69 8	99.48 57	99.59 13	98.34 2	99.26 39	96.55 33	99.10 24	99.96 10	99.36 26	99.25 24	98.37 64	99.64 118	99.66 114

thres200		96.76 82	96.53 115	97.03 60	96.31 84	99.67 13	98.37 72	93.99 68	97.68 149	94.49 63	95.83 123	86.77 143	99.18 40	98.26 84	97.82 93	99.82 13	99.66 114

3Dnovator	96.92 7	98.67 34	99.05 37	98.23 34	99.57 22	99.45 61	99.11 39	94.66 53	99.69 3	96.80 28	96.55 105	99.61 46	99.40 24	98.87 45	99.49 3	99.85 4	99.66 114

TSAR-MVS + GP.		98.66 36	99.36 20	97.85 41	97.16 75	99.46 59	99.03 45	94.59 56	99.09 58	97.19 24	99.73 3	99.95 15	99.39 25	98.95 37	98.69 42	99.75 45	99.65 117

FMVSNet3		97.02 76	98.12 74	95.73 99	93.59 150	97.98 152	98.34 75	91.32 109	98.80 90	93.92 76	97.21 83	95.94 85	97.63 99	98.61 64	98.62 46	99.61 133	99.65 117

CDS-MVSNet		96.59 96	98.02 78	94.92 106	94.45 137	98.96 103	97.46 104	91.75 100	97.86 141	90.07 113	96.02 115	97.25 72	96.21 133	98.04 106	98.38 62	99.60 140	99.65 117
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
ACMH+	95.51 13	95.40 119	96.00 131	94.70 108	96.33 82	98.79 110	96.79 124	91.32 109	98.77 96	87.18 128	95.60 129	85.46 157	96.97 113	97.15 141	96.59 132	99.59 146	99.65 117

QAPM		98.62 37	99.04 40	98.13 35	99.57 22	99.48 57	99.17 36	94.78 50	99.57 8	96.16 35	96.73 97	99.80 38	99.33 28	98.79 49	99.29 13	99.75 45	99.64 121

DeepC-MVS_fast	98.34 1	99.17 14	99.45 9	98.85 21	99.55 24	99.37 70	99.64 8	98.05 26	99.53 12	96.58 31	98.93 31	99.92 25	99.49 18	99.46 12	99.32 11	99.80 33	99.64 121
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 86	96.47 121	97.00 68	96.31 84	99.52 55	98.28 77	94.01 66	97.35 154	94.52 59	95.90 119	86.93 138	99.09 55	98.07 100	97.87 90	99.81 26	99.63 123

Effi-MVS+-dtu		95.74 113	98.04 76	93.06 137	93.92 140	99.16 95	97.90 91	88.16 159	99.07 65	82.02 158	98.02 67	94.32 101	96.74 121	98.53 71	97.56 106	99.61 133	99.62 124

HQP-MVS		96.37 98	96.58 112	96.13 89	97.31 71	98.44 136	98.45 62	95.22 44	98.86 80	88.58 119	98.33 58	87.00 137	97.67 98	97.23 138	96.56 133	99.56 157	99.62 124

tfpn_n400		97.32 67	98.38 60	96.09 90	96.07 100	99.30 81	98.00 88	93.84 79	99.35 26	90.50 109	98.93 31	94.24 103	98.30 80	98.65 59	98.60 48	99.83 9	99.60 126

tfpnconf		97.32 67	98.38 60	96.09 90	96.07 100	99.30 81	98.00 88	93.84 79	99.35 26	90.50 109	98.93 31	94.24 103	98.30 80	98.65 59	98.60 48	99.83 9	99.60 126

tfpnview11		97.32 67	98.33 63	96.14 88	96.07 100	99.31 80	98.08 86	93.96 74	99.25 41	90.50 109	98.93 31	94.24 103	98.38 76	98.61 64	98.36 65	99.84 5	99.59 128

IB-MVS	93.96 15	95.02 127	96.44 126	93.36 133	97.05 77	99.28 84	90.43 211	93.39 86	98.02 128	96.02 36	94.92 135	92.07 120	83.52 223	95.38 177	95.82 153	99.72 60	99.59 128
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 32	99.11 32	98.28 32	99.36 34	99.35 73	99.48 20	97.96 28	98.83 85	93.86 79	98.70 45	99.86 33	99.44 22	99.08 32	98.38 62	99.61 133	99.58 130

CDPH-MVS		98.41 40	99.10 33	97.61 45	99.32 38	99.36 71	99.49 18	96.15 39	98.82 87	91.82 99	98.41 55	99.66 45	99.10 53	98.93 39	98.97 28	99.75 45	99.58 130

APD-MVS		99.25 9	99.38 18	99.09 7	99.69 8	99.58 46	99.56 14	98.32 4	98.85 82	97.87 14	98.91 36	99.92 25	99.30 31	99.45 13	99.38 8	99.79 34	99.58 130
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
MVS_111021_LR		98.67 34	99.41 16	97.81 42	99.37 32	99.53 52	98.51 60	95.52 42	99.27 36	94.85 54	99.56 5	99.69 44	99.04 57	99.36 18	98.88 34	99.60 140	99.58 130

PHI-MVS		99.08 19	99.43 14	98.67 25	99.15 41	99.59 44	99.11 39	97.35 34	99.14 53	97.30 22	99.44 8	99.96 10	99.32 29	98.89 43	99.39 7	99.79 34	99.58 130

MVS_111021_HR		98.59 38	99.36 20	97.68 43	99.42 30	99.61 38	98.14 82	94.81 49	99.31 31	95.00 51	99.51 6	99.79 40	99.00 60	98.94 38	98.83 38	99.69 80	99.57 135

tfpn_ndepth		97.71 56	98.30 64	97.02 64	96.31 84	99.56 49	98.05 87	93.94 76	98.95 72	95.59 40	98.40 56	94.79 96	98.39 75	98.40 78	98.42 57	99.86 2	99.56 136

DeepPCF-MVS	97.74 3	98.34 42	99.46 8	97.04 59	98.82 47	99.33 77	96.28 135	97.47 33	99.58 7	94.70 57	98.99 28	99.85 36	97.24 107	99.55 7	99.34 10	97.73 211	99.56 136

CLD-MVS		96.74 85	96.51 117	97.01 66	96.71 80	98.62 125	98.73 54	94.38 58	98.94 75	94.46 64	97.33 79	87.03 136	98.07 87	97.20 140	96.87 125	99.72 60	99.54 138
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 93	99.08 34	93.81 119	97.10 76	99.42 65	98.85 50	90.01 133	99.31 31	79.98 177	99.78 2	99.10 53	97.42 104	98.35 79	98.05 82	99.47 169	99.53 139

pmmvs6		91.90 200	92.53 210	91.17 176	91.81 178	97.63 174	93.23 197	88.37 155	93.43 221	80.61 166	77.32 225	87.47 133	94.12 193	96.58 152	95.72 156	98.88 195	99.53 139

FC-MVSNet-test		96.07 108	97.94 81	93.89 117	93.60 149	98.67 122	96.62 128	90.30 128	98.76 97	88.62 118	95.57 130	97.63 67	94.48 188	97.97 108	97.48 112	99.71 69	99.52 141

CNVR-MVS		99.23 11	99.28 25	99.17 2	99.65 16	99.34 75	99.46 21	98.21 14	99.28 34	98.47 5	98.89 38	99.94 23	99.50 16	99.42 15	98.61 47	99.73 55	99.52 141

PLC	97.93 2	99.02 25	98.94 45	99.11 6	99.46 29	99.24 89	99.06 43	97.96 28	99.31 31	99.16 1	97.90 69	99.79 40	99.36 26	98.71 56	98.12 79	99.65 107	99.52 141
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
MAR-MVS		97.71 56	98.04 76	97.32 48	99.35 36	98.91 105	97.65 99	91.68 101	98.00 129	97.01 26	97.72 75	94.83 94	98.85 62	98.44 76	98.86 36	99.41 177	99.52 141
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 22	99.08 34	99.02 13	99.62 18	99.38 68	99.43 26	98.21 14	99.36 24	97.66 18	97.79 71	99.90 28	99.45 21	99.17 26	98.43 56	99.77 42	99.51 145

OpenMVS	96.23 11	97.95 50	98.45 58	97.35 47	99.52 27	99.42 65	98.91 49	94.61 54	98.87 79	92.24 97	94.61 138	99.05 54	99.10 53	98.64 62	99.05 24	99.74 49	99.51 145

thresconf0.02		97.18 72	97.81 83	96.45 83	96.11 99	99.20 94	98.21 78	94.26 61	99.14 53	91.72 101	98.65 46	91.51 123	98.57 70	98.22 90	98.47 54	99.82 13	99.50 147

Fast-Effi-MVS+-dtu		95.38 120	98.20 70	92.09 150	93.91 141	98.87 107	97.35 107	85.01 186	99.08 60	81.09 162	98.10 63	96.36 79	95.62 150	98.43 77	97.03 121	99.55 158	99.50 147

ACMM	96.26 9	96.67 92	96.69 111	96.66 76	97.29 72	98.46 133	96.48 132	95.09 45	99.21 46	93.19 88	98.78 42	86.73 144	98.17 83	97.84 115	96.32 139	99.74 49	99.49 149
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
CHOSEN 280x420		97.99 49	99.24 27	96.53 79	98.34 54	99.61 38	98.36 73	89.80 139	99.27 36	95.08 49	99.81 1	98.58 57	98.64 68	99.02 34	98.92 31	98.93 193	99.48 150

TAPA-MVS	97.53 5	98.41 40	98.84 50	97.91 40	99.08 43	99.33 77	99.15 37	97.13 35	99.34 29	93.20 87	97.75 73	99.19 52	99.20 36	98.66 58	98.13 78	99.66 101	99.48 150
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
GA-MVS		93.93 147	96.31 130	91.16 177	93.61 148	98.79 110	95.39 150	90.69 121	98.25 119	73.28 214	96.15 113	88.42 131	94.39 190	97.76 119	95.35 164	99.58 150	99.45 152

CVMVSNet		95.33 123	97.09 103	93.27 135	95.23 127	98.39 141	95.49 147	92.58 92	97.71 148	83.00 152	94.44 140	93.28 113	93.92 198	97.79 116	98.54 53	99.41 177	99.45 152

LS3D		97.79 52	98.25 66	97.26 52	98.40 53	99.63 30	99.53 15	98.63 1	99.25 41	88.13 121	96.93 93	94.14 106	99.19 38	99.14 28	99.23 17	99.69 80	99.42 154

abl_6				98.09 36	99.33 37	99.22 91	98.79 53	94.96 48	98.52 111	97.00 27	97.30 81	99.86 33	98.76 63			99.69 80	99.41 155

test0.0.03 1		96.69 90	98.12 74	95.01 105	95.49 120	98.99 100	95.86 140	90.82 117	98.38 115	92.54 95	96.66 99	97.33 69	95.75 144	97.75 120	98.34 68	99.60 140	99.40 156

testgi		95.67 114	97.48 92	93.56 126	95.07 130	99.00 98	95.33 151	88.47 153	98.80 90	86.90 130	97.30 81	92.33 118	95.97 141	97.66 123	97.91 88	99.60 140	99.38 157

TAMVS		95.53 116	96.50 120	94.39 112	93.86 143	99.03 97	96.67 126	89.55 142	97.33 160	90.64 107	93.02 155	91.58 122	96.21 133	97.72 122	97.43 115	99.43 174	99.36 158

AdaColmap		99.06 21	98.98 44	99.15 5	99.60 20	99.30 81	99.38 28	98.16 17	99.02 69	98.55 4	98.71 44	99.57 49	99.58 12	99.09 30	97.84 92	99.64 118	99.36 158

PM-MVS		89.55 210	90.30 215	88.67 206	87.06 218	95.60 209	90.88 209	84.51 191	96.14 201	75.75 205	86.89 209	63.47 230	94.64 184	96.85 148	93.89 203	99.17 190	99.29 160

pmmvs4		95.09 125	95.90 134	94.14 114	92.29 160	97.70 166	95.45 148	90.31 126	98.60 103	90.70 105	93.25 149	89.90 128	96.67 123	97.13 142	95.42 161	99.44 173	99.28 161

EG-PatchMatch MVS		92.45 181	93.92 177	90.72 189	92.56 157	98.43 138	94.88 162	84.54 190	97.18 164	79.55 185	86.12 213	83.23 177	93.15 204	97.22 139	96.00 147	99.67 96	99.27 162

UA-Net		97.13 74	99.14 31	94.78 107	97.21 73	99.38 68	97.56 100	92.04 95	98.48 112	88.03 122	98.39 57	99.91 27	94.03 195	99.33 21	99.23 17	99.81 26	99.25 163

pmmvs-eth3d		89.81 209	89.65 216	90.00 198	86.94 219	95.38 216	91.08 207	86.39 175	94.57 217	82.27 157	83.03 219	64.94 227	93.96 196	96.57 153	93.82 204	99.35 181	99.24 164

gg-mvs-nofinetune		90.85 204	94.14 164	87.02 210	94.89 134	99.25 87	98.64 56	76.29 227	88.24 228	57.50 232	79.93 223	95.45 88	95.18 179	98.77 51	98.07 80	99.62 130	99.24 164

PMMVS		97.52 61	98.39 59	96.51 81	95.82 112	98.73 119	97.80 95	93.05 91	98.76 97	94.39 68	99.07 27	97.03 74	98.55 71	98.31 81	97.61 104	99.43 174	99.21 166

CNLPA		99.03 24	99.05 37	99.01 14	99.27 39	99.22 91	99.03 45	97.98 27	99.34 29	99.00 2	98.25 60	99.71 43	99.31 30	98.80 48	98.82 39	99.48 167	99.17 167

CR-MVSNet		94.57 137	97.34 96	91.33 171	94.90 133	98.59 127	97.15 115	79.14 216	97.98 130	80.42 170	96.59 104	93.50 112	96.85 118	98.10 95	97.49 110	99.50 166	99.15 168

PatchT		93.96 146	97.36 95	90.00 198	94.76 136	98.65 123	90.11 214	78.57 221	97.96 133	80.42 170	96.07 114	94.10 107	96.85 118	98.10 95	97.49 110	99.26 186	99.15 168

COLMAP_ROB	96.15 12	97.78 53	98.17 71	97.32 48	98.84 46	99.45 61	99.28 31	95.43 43	99.48 16	91.80 100	94.83 136	98.36 61	98.90 61	98.09 97	97.85 91	99.68 89	99.15 168
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
MSDG		98.27 44	98.29 65	98.24 33	99.20 40	99.22 91	99.20 34	97.82 30	99.37 22	94.43 65	95.90 119	97.31 70	99.12 50	98.76 52	98.35 66	99.67 96	99.14 171

test-mter		94.86 129	97.32 97	92.00 154	92.41 159	98.82 109	96.18 137	86.35 176	98.05 127	82.28 156	96.48 106	94.39 100	95.46 163	98.17 92	96.20 143	99.32 183	99.13 172

RPMNet		94.66 132	97.16 102	91.75 163	94.98 131	98.59 127	97.00 122	78.37 222	97.98 130	83.78 143	96.27 110	94.09 108	96.91 115	97.36 134	96.73 127	99.48 167	99.09 173

OMC-MVS		98.84 29	99.01 43	98.65 26	99.39 31	99.23 90	99.22 33	96.70 36	99.40 19	97.77 16	97.89 70	99.80 38	99.21 35	99.02 34	98.65 45	99.57 154	99.07 174

TSAR-MVS + COLMAP		96.79 81	96.55 114	97.06 58	97.70 64	98.46 133	99.07 42	96.23 38	99.38 20	91.32 104	98.80 40	85.61 155	98.69 66	97.64 126	96.92 124	99.37 180	99.06 175

tpm		92.38 186	94.79 152	89.56 201	94.30 138	97.50 185	94.24 192	78.97 219	97.72 147	74.93 210	97.97 68	82.91 183	96.60 126	93.65 210	94.81 193	98.33 202	98.98 176

PatchMatch-RL		97.77 54	98.25 66	97.21 53	99.11 42	99.25 87	97.06 121	94.09 65	98.72 100	95.14 48	98.47 53	96.29 80	98.43 74	98.65 59	97.44 114	99.45 171	98.94 177

pmmvs5		92.71 176	94.27 163	90.90 184	91.42 199	97.74 162	93.23 197	86.66 173	95.99 206	78.96 193	91.45 157	83.44 174	95.55 152	97.30 136	95.05 173	99.58 150	98.93 178

test-LLR		95.50 117	97.32 97	93.37 132	95.49 120	98.74 117	96.44 133	90.82 117	98.18 121	82.75 153	96.60 102	94.67 98	95.54 153	98.09 97	96.00 147	99.20 188	98.93 178

TESTMET0.1,1		94.95 128	97.32 97	92.20 147	92.62 155	98.74 117	96.44 133	86.67 172	98.18 121	82.75 153	96.60 102	94.67 98	95.54 153	98.09 97	96.00 147	99.20 188	98.93 178

EU-MVSNet		92.80 169	94.76 153	90.51 192	91.88 173	96.74 202	92.48 203	88.69 150	96.21 199	79.00 192	91.51 156	87.82 132	91.83 209	95.87 173	96.27 140	99.21 187	98.92 181

Anonymous20231211		83.86 218	83.39 224	84.40 217	85.29 222	93.44 224	86.29 222	84.24 195	85.55 231	68.63 220	61.25 231	59.57 233	84.33 222	92.50 216	92.52 214	97.65 213	98.89 182

PCF-MVS	97.50 6	98.18 46	98.35 62	97.99 38	98.65 50	99.36 71	98.94 48	98.14 21	98.59 104	93.62 83	96.61 101	99.76 42	99.03 58	97.77 118	97.45 113	99.57 154	98.89 182
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
EPNet_dtu		96.30 102	98.53 56	93.70 123	98.97 45	98.24 147	97.36 106	94.23 62	98.85 82	79.18 191	99.19 17	98.47 59	94.09 194	97.89 112	98.21 75	98.39 201	98.85 184
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
OPM-MVS		96.22 104	95.85 137	96.65 77	97.75 62	98.54 130	99.00 47	95.53 41	96.88 178	89.88 115	95.95 118	86.46 148	98.07 87	97.65 125	96.63 130	99.67 96	98.83 185

GG-mvs-BLEND		69.11 228	98.13 73	35.26 234	3.49 240	98.20 149	94.89 159	2.38 238	98.42 114	5.82 243	96.37 109	98.60 56	5.97 238	98.75 54	97.98 85	99.01 192	98.61 186

ambc			80.99 226		80.04 232	90.84 225	90.91 208		96.09 202	74.18 211	62.81 230	30.59 241	82.44 224	96.25 167	91.77 221	95.91 229	98.56 187

MDTV_nov1_ep13_2view		92.44 182	95.66 139	88.68 205	91.05 206	97.92 156	92.17 204	79.64 211	98.83 85	76.20 204	91.45 157	93.51 111	95.04 181	95.68 175	93.70 205	97.96 207	98.53 188

USDC		94.26 140	94.83 151	93.59 125	96.02 103	98.44 136	97.84 92	88.65 151	98.86 80	82.73 155	94.02 141	80.56 205	96.76 120	97.28 137	96.15 146	99.55 158	98.50 189

MDA-MVSNet-bldmvs		87.84 216	89.22 217	86.23 212	81.74 229	96.77 201	83.74 224	89.57 141	94.50 218	72.83 216	96.64 100	64.47 229	92.71 206	81.43 230	92.28 218	96.81 225	98.47 190

gm-plane-assit		89.44 211	92.82 208	85.49 214	91.37 201	95.34 217	79.55 228	82.12 202	91.68 224	64.79 227	87.98 199	80.26 208	95.66 147	98.51 72	97.56 106	99.45 171	98.41 191

MS-PatchMatch		95.99 109	97.26 101	94.51 110	97.46 66	98.76 115	97.27 110	86.97 169	99.09 58	89.83 116	93.51 146	97.78 65	96.18 135	97.53 130	95.71 157	99.35 181	98.41 191

TransMVSNet (Re)		93.45 154	94.08 168	92.72 142	92.83 153	97.62 177	94.94 157	91.54 106	95.65 212	83.06 151	88.93 181	83.53 173	94.25 191	97.41 133	97.03 121	99.67 96	98.40 193

TinyColmap		94.00 144	94.35 162	93.60 124	95.89 108	98.26 145	97.49 103	88.82 148	98.56 107	83.21 149	91.28 159	80.48 207	96.68 122	97.34 135	96.26 142	99.53 163	98.24 194

TDRefinement		93.04 163	93.57 188	92.41 143	96.58 81	98.77 113	97.78 97	91.96 98	98.12 124	80.84 163	89.13 180	79.87 211	87.78 214	96.44 156	94.50 199	99.54 162	98.15 195

MDTV_nov1_ep13		95.57 115	97.48 92	93.35 134	95.43 122	98.97 102	97.19 114	83.72 200	98.92 78	87.91 124	97.75 73	96.12 83	97.88 95	96.84 149	95.64 158	97.96 207	98.10 196

tpmp4_e23		93.84 151	94.58 157	92.98 139	95.41 125	98.29 144	96.81 123	80.57 206	98.15 123	90.53 108	97.00 89	84.39 168	96.91 115	93.69 208	92.45 215	97.67 212	98.06 197

MIMVSNet		94.49 138	97.59 89	90.87 186	91.74 186	98.70 121	94.68 180	78.73 220	97.98 130	83.71 146	97.71 76	94.81 95	96.96 114	97.97 108	97.92 87	99.40 179	98.04 198

CostFormer		94.25 141	94.88 150	93.51 129	95.43 122	98.34 143	96.21 136	80.64 205	97.94 135	94.01 73	98.30 59	86.20 151	97.52 100	92.71 212	92.69 212	97.23 223	98.02 199

RPSCF		97.61 59	98.16 72	96.96 71	98.10 56	99.00 98	98.84 51	93.76 81	99.45 17	94.78 56	99.39 11	99.31 51	98.53 72	96.61 150	95.43 160	97.74 209	97.93 200

testus		88.77 213	92.77 209	84.10 218	88.24 216	93.95 221	87.16 220	84.24 195	97.37 153	61.54 231	95.70 127	73.10 223	84.90 220	95.56 176	95.82 153	98.51 197	97.88 201

Anonymous20231206		90.70 206	93.93 176	86.92 211	90.21 214	96.79 200	90.30 213	86.61 174	96.05 204	69.25 219	88.46 192	84.86 164	85.86 218	97.11 143	96.47 136	99.30 184	97.80 202

DWT-MVSNet_training		95.38 120	95.05 147	95.78 95	95.86 110	98.88 106	97.55 101	90.09 132	98.23 120	96.49 34	97.62 78	86.92 142	97.16 109	92.03 220	94.12 201	97.52 215	97.50 203

pmmvs3		88.19 215	91.27 212	84.60 216	85.60 221	93.66 222	85.68 223	81.13 203	92.36 223	63.66 229	89.51 176	77.10 219	93.22 203	96.37 159	92.40 216	98.30 203	97.46 204

test2356		88.81 212	92.86 205	84.09 219	87.85 217	93.46 223	87.07 221	83.60 201	96.50 196	62.08 230	97.06 88	75.04 221	85.17 219	95.08 199	95.42 161	98.75 196	97.46 204

N_pmnet		92.21 193	94.60 155	89.42 202	91.88 173	97.38 192	89.15 216	89.74 140	97.89 138	73.75 212	87.94 200	92.23 119	93.85 199	96.10 169	93.20 208	98.15 205	97.43 206

PatchmatchNet		94.70 131	97.08 104	91.92 157	95.53 118	98.85 108	95.77 141	79.54 213	98.95 72	85.98 134	98.52 49	96.45 76	97.39 105	95.32 178	94.09 202	97.32 219	97.38 207
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
ADS-MVSNet		94.65 133	97.04 105	91.88 160	95.68 115	98.99 100	95.89 139	79.03 218	99.15 51	85.81 136	96.96 91	98.21 63	97.10 110	94.48 205	94.24 200	97.74 209	97.21 208

MVS-HIRNet		92.51 180	95.97 132	88.48 207	93.73 147	98.37 142	90.33 212	75.36 230	98.32 116	77.78 197	89.15 179	94.87 93	95.14 180	97.62 127	96.39 137	98.51 197	97.11 209

dps		94.63 134	95.31 146	93.84 118	95.53 118	98.71 120	96.54 129	80.12 208	97.81 146	97.21 23	96.98 90	92.37 117	96.34 132	92.46 217	91.77 221	97.26 221	97.08 210

test20.03		90.65 207	93.71 182	87.09 209	90.44 212	96.24 203	89.74 215	85.46 180	95.59 213	72.99 215	90.68 162	85.33 158	84.41 221	95.94 172	95.10 172	99.52 164	97.06 211

EPMVS		95.05 126	96.86 109	92.94 140	95.84 111	98.96 103	96.68 125	79.87 209	99.05 66	90.15 112	97.12 87	95.99 84	97.49 102	95.17 186	94.75 195	97.59 214	96.96 212

testpf		91.80 201	94.43 161	88.74 204	93.89 142	95.30 218	92.05 205	71.77 231	97.52 151	87.24 127	94.77 137	92.68 116	91.48 210	91.75 223	92.11 220	96.02 228	96.89 213

LP		92.12 196	94.60 155	89.22 203	94.96 132	98.45 135	93.01 199	77.58 223	97.85 142	77.26 200	89.80 174	93.00 115	94.54 185	93.69 208	92.58 213	98.00 206	96.83 214

tpmrst		93.86 149	95.88 135	91.50 167	95.69 114	98.62 125	95.64 144	79.41 214	98.80 90	83.76 145	95.63 128	96.13 82	97.25 106	92.92 211	92.31 217	97.27 220	96.74 215

new-patchmatchnet		86.12 217	87.30 218	84.74 215	86.92 220	95.19 220	83.57 225	84.42 193	92.67 222	65.66 224	80.32 222	64.72 228	89.41 213	92.33 219	89.21 225	98.43 200	96.69 216

tpm cat1		94.06 142	94.90 149	93.06 137	95.42 124	98.52 131	96.64 127	80.67 204	97.82 144	92.63 94	93.39 148	95.00 92	96.06 139	91.36 224	91.58 223	96.98 224	96.66 217

FMVSNet5		95.42 118	96.47 121	94.20 113	92.26 161	95.99 205	95.66 143	87.15 166	97.87 139	93.46 85	96.68 98	93.79 109	97.52 100	97.10 144	97.21 119	99.11 191	96.62 218

DeepMVS_CX						96.85 199	87.43 219	89.27 143	98.30 117	75.55 207	95.05 133	79.47 212	92.62 207	89.48 226		95.18 230	95.96 219

MIMVSNet1		88.61 214	90.68 214	86.19 213	81.56 230	95.30 218	87.78 218	85.98 178	94.19 219	72.30 217	78.84 224	78.90 216	90.06 212	96.59 151	95.47 159	99.46 170	95.49 220

CMPMVS	70.31 18	90.74 205	91.06 213	90.36 195	97.32 69	97.43 189	92.97 200	87.82 163	93.50 220	75.34 209	83.27 218	84.90 163	92.19 208	92.64 215	91.21 224	96.50 226	94.46 221
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
new_pmnet		90.45 208	92.84 207	87.66 208	88.96 215	96.16 204	88.71 217	84.66 189	97.56 150	71.91 218	85.60 214	86.58 147	93.28 202	96.07 170	93.54 206	98.46 199	94.39 222

Gipuma		81.40 223	81.78 225	80.96 222	83.21 225	85.61 234	79.73 227	76.25 228	97.33 160	64.21 228	55.32 232	55.55 235	86.04 217	92.43 218	92.20 219	96.32 227	93.99 223
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
testmv		81.83 221	86.26 219	76.66 224	84.10 223	89.42 229	74.29 232	79.65 210	90.61 225	51.85 236	82.11 220	63.06 232	72.61 228	91.94 221	92.75 210	97.49 216	93.94 224

test1235678		81.83 221	86.26 219	76.66 224	84.10 223	89.41 230	74.29 232	79.64 211	90.60 226	51.84 237	82.11 220	63.07 231	72.61 228	91.94 221	92.75 210	97.49 216	93.94 224

PMMVS2		77.26 225	79.47 227	74.70 228	76.00 233	88.37 232	74.22 234	76.34 226	78.31 233	54.13 233	69.96 229	52.50 236	70.14 231	84.83 228	88.71 226	97.35 218	93.58 226

test12356		80.53 224	84.80 222	75.54 226	82.31 226	88.05 233	75.99 229	79.31 215	88.53 227	53.24 235	83.30 217	56.38 234	65.16 234	90.87 225	93.10 209	97.25 222	93.34 227

1111		82.87 220	85.67 221	79.62 223	81.86 227	89.62 227	74.44 230	68.81 233	87.44 229	66.59 222	76.83 226	70.33 225	87.71 215	92.65 213	93.37 207	98.28 204	89.42 228

MVE	67.97 19	65.53 232	67.43 233	63.31 233	59.33 237	74.20 236	53.09 239	70.43 232	66.27 236	43.13 238	45.98 237	30.62 240	70.65 230	79.34 232	86.30 227	83.25 237	89.33 229
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
FPMVS		83.82 219	84.61 223	82.90 220	90.39 213	90.71 226	90.85 210	84.10 198	95.47 214	65.15 225	83.44 216	74.46 222	75.48 225	81.63 229	79.42 231	91.42 232	87.14 230

no-one		66.79 231	67.62 232	65.81 232	73.06 236	81.79 235	51.90 240	76.20 229	61.07 237	54.05 234	51.62 236	41.72 238	49.18 235	67.26 234	82.83 229	90.47 233	87.07 231

EMVS		68.12 230	68.11 231	68.14 230	75.51 234	71.76 237	55.38 238	77.20 225	77.78 234	37.79 240	53.59 233	43.61 237	74.72 226	67.05 235	76.70 233	88.27 236	86.24 232

E-PMN		68.30 229	68.43 230	68.15 229	74.70 235	71.56 238	55.64 237	77.24 224	77.48 235	39.46 239	51.95 235	41.68 239	73.28 227	70.65 233	79.51 230	88.61 235	86.20 233

PMVS	72.60 17	76.39 226	77.66 228	74.92 227	81.04 231	69.37 239	68.47 235	80.54 207	85.39 232	65.07 226	73.52 228	72.91 224	65.67 233	80.35 231	76.81 232	88.71 234	85.25 234
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
.test1245		69.67 227	72.22 229	66.70 231	81.86 227	89.62 227	74.44 230	68.81 233	87.44 229	66.59 222	76.83 226	70.33 225	87.71 215	92.65 213	37.65 234	20.79 238	51.04 235

testmvs		31.24 233	40.15 234	20.86 235	12.61 238	17.99 240	25.16 241	13.30 236	48.42 238	24.82 241	53.07 234	30.13 242	28.47 236	42.73 236	37.65 234	20.79 238	51.04 235

test123		26.75 234	34.25 235	18.01 236	7.93 239	17.18 241	24.85 242	12.36 237	44.83 239	16.52 242	41.80 238	18.10 243	28.29 237	33.08 237	34.79 236	18.10 240	49.95 237

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										98.09 10		99.97 4

MTMP										98.46 7		99.96 10

Patchmatch-RL test							66.86 236

tmp_tt				82.25 221	97.73 63	88.71 231	80.18 226	68.65 235	99.15 51	86.98 129	99.47 7	85.31 159	68.35 232	87.51 227	83.81 228	91.64 231

XVS					97.42 67	99.62 34	98.59 58			93.81 80		99.95 15				99.69 80

X-MVStestdata					97.42 67	99.62 34	98.59 58			93.81 80		99.95 15				99.69 80

mPP-MVS					99.53 25							99.89 30

NP-MVS									98.57 106

Patchmtry						98.59 127	97.15 115	79.14 216		80.42 170

High-resolution multi-view benchmark