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.

CS-MVS		98.56 43	99.32 28	97.68 47	98.28 61	99.89 2	98.71 60	94.53 63	99.41 23	95.43 48	99.05 35	98.66 65	99.19 40	99.21 29	99.07 26	99.93 1	99.94 1

DROMVSNet		98.22 51	99.44 17	96.79 75	95.62 120	99.56 51	99.01 50	92.22 99	99.17 53	94.51 66	99.41 13	99.62 51	99.49 18	99.16 34	99.26 14	99.91 2	99.94 1

CS-MVS-test		98.58 42	99.42 20	97.60 51	98.52 56	99.91 1	98.60 63	94.60 60	99.37 27	94.62 62	99.40 14	99.16 60	99.39 26	99.36 20	98.85 47	99.90 3	99.92 3

EPP-MVSNet		97.75 63	98.71 60	96.63 81	95.68 118	99.56 51	97.51 108	93.10 95	99.22 46	94.99 57	97.18 93	97.30 83	98.65 75	98.83 58	98.93 38	99.84 12	99.92 3

LTVRE_ROB	93.20 16	92.84 174	94.92 161	90.43 187	92.83 162	98.63 132	97.08 127	87.87 165	97.91 154	68.42 213	93.54 149	79.46 209	96.62 135	97.55 140	97.40 125	99.74 49	99.92 3
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
DVP-MVS++		99.41 4	99.64 1	99.14 7	99.69 7	99.75 9	99.64 8	98.33 6	99.67 4	98.10 13	99.66 4	99.99 1	99.33 30	99.62 5	98.86 44	99.74 49	99.90 6

canonicalmvs		97.31 76	97.81 95	96.72 76	96.20 101	99.45 68	98.21 85	91.60 111	99.22 46	95.39 49	98.48 59	90.95 140	99.16 46	97.66 134	99.05 29	99.76 41	99.90 6

PVSNet_Blended_VisFu		97.41 74	98.49 66	96.15 92	97.49 71	99.76 6	96.02 148	93.75 80	99.26 42	93.38 90	93.73 147	99.35 56	96.47 140	98.96 46	98.46 65	99.77 39	99.90 6

CSCG		98.90 30	98.93 53	98.85 24	99.75 3	99.72 12	99.49 21	96.58 42	99.38 25	98.05 16	98.97 37	97.87 76	99.49 18	97.78 127	98.92 39	99.78 34	99.90 6

MVS_0304		98.14 54	99.03 48	97.10 63	98.05 65	99.63 29	99.27 34	94.33 68	99.63 7	93.06 94	97.32 88	99.05 63	98.09 94	98.82 59	98.87 43	99.81 22	99.89 10

PS-CasMVS		92.72 179	93.36 193	91.98 161	91.62 190	97.52 189	94.13 188	88.98 151	95.94 201	81.51 170	87.35 195	79.95 206	95.91 153	96.37 171	96.49 145	99.70 82	99.89 10

CP-MVSNet		93.25 168	94.00 181	92.38 152	91.65 188	97.56 187	94.38 184	89.20 149	96.05 198	83.16 159	89.51 178	81.97 194	96.16 148	96.43 169	96.56 143	99.71 74	99.89 10

WR-MVS_H		93.54 164	94.67 168	92.22 153	91.95 177	97.91 167	94.58 181	88.75 154	96.64 188	83.88 151	90.66 172	85.13 176	94.40 182	96.54 167	95.91 165	99.73 57	99.89 10

FC-MVSNet-train		97.04 87	97.91 92	96.03 96	96.00 104	98.41 149	96.53 139	93.42 85	99.04 78	93.02 95	98.03 74	94.32 120	97.47 114	97.93 119	97.77 108	99.75 44	99.88 14

IterMVS-LS		96.12 116	97.48 103	94.53 115	95.19 135	97.56 187	97.15 122	89.19 150	99.08 70	88.23 126	94.97 135	94.73 114	97.84 106	97.86 124	98.26 83	99.60 130	99.88 14
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
DCV-MVSNet		97.56 69	98.36 69	96.62 82	96.44 92	98.36 153	98.37 76	91.73 108	99.11 66	94.80 59	98.36 65	96.28 94	98.60 79	98.12 102	98.44 67	99.76 41	99.87 16

v7n		91.61 195	92.95 196	90.04 189	90.56 204	97.69 175	93.74 189	85.59 181	95.89 202	76.95 190	86.60 200	78.60 212	93.76 193	97.01 158	94.99 184	99.65 110	99.87 16

CHOSEN 1792x2688		96.41 108	96.99 124	95.74 102	98.01 66	99.72 12	97.70 104	90.78 128	99.13 65	90.03 120	87.35 195	95.36 105	98.33 87	98.59 81	98.91 41	99.59 136	99.87 16

CANet		98.46 44	99.16 37	97.64 49	98.48 57	99.64 26	99.35 31	94.71 56	99.53 14	95.17 53	97.63 85	99.59 53	98.38 86	98.88 56	98.99 34	99.74 49	99.86 19

baseline		97.45 73	98.70 61	95.99 98	95.89 107	99.36 81	98.29 81	91.37 117	99.21 48	92.99 96	98.40 63	96.87 88	97.96 99	98.60 79	98.60 60	99.42 168	99.86 19

HyFIR lowres test		95.99 118	96.56 134	95.32 107	97.99 67	99.65 22	96.54 137	88.86 152	98.44 129	89.77 123	84.14 205	97.05 86	99.03 55	98.55 83	98.19 87	99.73 57	99.86 19

GeoE		95.98 120	97.24 117	94.51 116	95.02 138	99.38 77	98.02 95	87.86 166	98.37 132	87.86 131	92.99 161	93.54 127	98.56 80	98.61 76	97.92 98	99.73 57	99.85 22

tfpnnormal		93.85 162	94.12 177	93.54 137	93.22 161	98.24 157	95.45 158	91.96 105	94.61 207	83.91 150	90.74 170	81.75 196	97.04 121	97.49 142	96.16 156	99.68 93	99.84 23

Effi-MVS+		95.81 121	97.31 115	94.06 124	95.09 136	99.35 84	97.24 118	88.22 161	98.54 124	85.38 147	98.52 57	88.68 151	98.70 72	98.32 93	97.93 97	99.74 49	99.84 23

SD-MVS		99.25 12	99.50 12	98.96 20	98.79 51	99.55 53	99.33 32	98.29 12	99.75 1	97.96 18	99.15 24	99.95 17	99.61 6	99.17 32	99.06 28	99.81 22	99.84 23
Zhenlong Yuan, Jiakai Cao, Zhaoxin Li, Hao Jiang and Zhaoqi Wang: SD-MVS: Segmentation-driven Deformation Multi-View Stereo with Spherical Refinement and EM optimization. AAAI2024
EPNet		98.05 55	98.86 55	97.10 63	99.02 47	99.43 72	98.47 69	94.73 55	99.05 76	95.62 44	98.93 40	97.62 80	95.48 165	98.59 81	98.55 61	99.29 178	99.84 23
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
SteuartSystems-ACMMP		99.20 15	99.51 11	98.83 26	99.66 16	99.66 21	99.71 3	98.12 27	99.14 61	96.62 33	99.16 23	99.98 2	99.12 48	99.63 3	99.19 21	99.78 34	99.83 27
Skip Steuart: Steuart Systems R&D Blog.
TSAR-MVS + MP.		99.27 10	99.57 5	98.92 22	98.78 52	99.53 55	99.72 2	98.11 28	99.73 2	97.43 25	99.15 24	99.96 12	99.59 9	99.73 1	99.07 26	99.88 4	99.82 28
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
anonymousdsp		93.12 170	95.86 153	89.93 192	91.09 201	98.25 156	95.12 162	85.08 183	97.44 167	73.30 203	90.89 169	90.78 141	95.25 173	97.91 120	95.96 164	99.71 74	99.82 28

TSAR-MVS + ACMM		98.77 33	99.45 14	97.98 42	99.37 36	99.46 66	99.44 27	98.13 26	99.65 5	92.30 106	98.91 42	99.95 17	99.05 53	99.42 17	98.95 37	99.58 140	99.82 28

PEN-MVS		92.72 179	93.20 195	92.15 156	91.29 198	97.31 197	94.67 178	89.81 141	96.19 194	81.83 168	88.58 186	79.06 210	95.61 161	95.21 192	96.27 151	99.72 64	99.82 28

WR-MVS		93.43 167	94.48 171	92.21 154	91.52 193	97.69 175	94.66 179	89.98 138	96.86 182	83.43 156	90.12 174	85.03 177	93.94 190	96.02 183	95.82 166	99.71 74	99.82 28

UniMVSNet_ETH3D		93.15 169	92.33 202	94.11 123	93.91 150	98.61 135	94.81 172	90.98 123	97.06 177	87.51 134	82.27 209	76.33 215	97.87 105	94.79 199	97.47 121	99.56 147	99.81 33

DeepC-MVS	97.63 4	98.33 48	98.57 62	98.04 40	98.62 55	99.65 22	99.45 25	98.15 23	99.51 17	92.80 98	95.74 126	96.44 91	99.46 21	99.37 19	99.50 2	99.78 34	99.81 33
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
EIA-MVS		97.70 65	98.78 58	96.44 88	95.72 115	99.65 22	98.14 88	93.72 81	98.30 136	92.31 105	98.63 55	97.90 75	98.97 58	98.92 51	98.30 81	99.78 34	99.80 35

v8		92.87 173	93.87 186	91.72 169	92.05 175	97.50 190	94.79 173	88.20 162	96.85 183	80.11 178	90.01 175	82.86 190	95.48 165	95.15 194	94.90 187	99.66 106	99.80 35

v10		92.79 177	94.06 179	91.31 175	91.78 183	97.29 199	94.87 170	86.10 179	96.97 180	79.82 180	88.16 189	84.56 180	95.63 159	96.33 174	95.31 175	99.65 110	99.80 35

UniMVSNet_NR-MVSNet		94.59 147	95.47 157	93.55 136	91.85 181	97.89 168	95.03 163	92.00 103	97.33 170	86.12 139	93.19 155	87.29 155	96.60 136	96.12 179	96.70 137	99.72 64	99.80 35

DU-MVS		93.98 157	94.44 172	93.44 139	91.66 186	97.77 170	95.03 163	91.57 112	97.17 174	86.12 139	93.13 157	81.13 198	96.60 136	95.10 195	97.01 132	99.67 101	99.80 35

casdiffmvs_mvg		97.27 78	97.97 90	96.46 87	95.83 111	99.51 61	98.42 72	93.32 89	98.34 134	92.38 104	95.64 129	95.35 106	98.91 61	98.73 68	98.45 66	99.86 9	99.80 35
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
UGNet		97.66 66	99.07 43	96.01 97	97.19 80	99.65 22	97.09 126	93.39 86	99.35 31	94.40 71	98.79 47	99.59 53	94.24 185	98.04 113	98.29 82	99.73 57	99.80 35
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 59	98.86 55	96.68 77	96.02 102	99.72 12	98.35 79	93.37 88	98.75 115	94.01 75	96.88 100	98.40 70	98.48 84	99.09 37	99.42 5	99.83 15	99.80 35

ETV-MVS		98.05 55	99.25 33	96.65 79	95.61 121	99.61 38	98.26 84	93.52 84	98.90 91	93.74 85	99.32 17	99.20 58	98.90 63	99.21 29	98.72 54	99.87 8	99.79 43

DVP-MVS		99.45 2	99.54 7	99.35 1	99.72 6	99.76 6	99.63 12	98.37 2	99.63 7	99.03 3	98.95 39	99.98 2	99.60 7	99.60 7	99.05 29	99.74 49	99.79 43
Zhenlong Yuan, Jinguo Luo, Fei Shen, Zhaoxin Li, Cong Liu, Tianlu Mao, Zhaoqi Wang: DVP-MVS: Synergize Depth-Edge and Visibility Prior for Multi-View Stereo. AAAI2025
thisisatest0515		94.61 146	96.89 126	91.95 162	92.00 176	98.47 143	92.01 197	90.73 129	98.18 141	83.96 149	94.51 140	95.13 109	93.38 195	97.38 145	94.74 192	99.61 122	99.79 43

MSP-MVS		99.34 7	99.52 10	99.14 7	99.68 12	99.75 9	99.64 8	98.31 9	99.44 21	98.10 13	99.28 18	99.98 2	99.30 35	99.34 23	99.05 29	99.81 22	99.79 43
Zhenlong Yuan, Cong Liu, Fei Shen, Zhaoxin Li, Jingguo luo, Tianlu Mao and Zhaoqi Wang: MSP-MVS: Multi-granularity Segmentation Prior Guided Multi-View Stereo. AAAI2025
UniMVSNet (Re)		94.58 148	95.34 158	93.71 131	92.25 173	98.08 161	94.97 165	91.29 122	97.03 179	87.94 129	93.97 146	86.25 168	96.07 149	96.27 176	95.97 163	99.72 64	99.79 43

DELS-MVS		98.19 52	98.77 59	97.52 52	98.29 60	99.71 15	99.12 41	94.58 62	98.80 105	95.38 50	96.24 116	98.24 73	97.92 100	99.06 40	99.52 1	99.82 16	99.79 43
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
tttt0517		97.23 80	98.24 76	96.04 95	95.60 123	99.60 43	96.94 131	93.23 90	99.15 58	92.56 102	98.74 52	96.12 98	98.17 89	98.21 98	96.10 158	99.73 57	99.78 49

v144192		92.38 188	93.55 191	91.00 180	91.44 194	97.47 192	94.27 185	87.41 169	96.52 191	78.03 187	87.50 194	82.65 192	95.32 170	95.82 187	95.15 180	99.55 149	99.78 49

V42		93.05 171	93.90 185	92.04 158	91.91 178	97.66 177	94.91 167	89.91 139	96.85 183	80.58 174	89.66 177	83.43 186	95.37 169	95.03 197	94.90 187	99.59 136	99.78 49

MVS_Test		97.30 77	98.54 63	95.87 99	95.74 114	99.28 93	98.19 86	91.40 116	99.18 52	91.59 112	98.17 70	96.18 96	98.63 77	98.61 76	98.55 61	99.66 106	99.78 49

TranMVSNet+NR-MVSNet		93.67 163	94.14 175	93.13 145	91.28 200	97.58 185	95.60 155	91.97 104	97.06 177	84.05 148	90.64 173	82.22 193	96.17 147	94.94 198	96.78 135	99.69 85	99.78 49

PVSNet_BlendedMVS		97.51 71	97.71 96	97.28 58	98.06 63	99.61 38	97.31 114	95.02 51	99.08 70	95.51 46	98.05 72	90.11 143	98.07 95	98.91 52	98.40 70	99.72 64	99.78 49

PVSNet_Blended		97.51 71	97.71 96	97.28 58	98.06 63	99.61 38	97.31 114	95.02 51	99.08 70	95.51 46	98.05 72	90.11 143	98.07 95	98.91 52	98.40 70	99.72 64	99.78 49

SED-MVS		99.44 3	99.58 4	99.28 3	99.69 7	99.76 6	99.62 14	98.35 3	99.51 17	99.05 2	99.60 6	99.98 2	99.28 37	99.61 6	98.83 49	99.70 82	99.77 56

thisisatest0530		97.23 80	98.25 73	96.05 94	95.60 123	99.59 45	96.96 130	93.23 90	99.17 53	92.60 101	98.75 51	96.19 95	98.17 89	98.19 100	96.10 158	99.72 64	99.77 56

Fast-Effi-MVS+		95.38 130	96.52 137	94.05 125	94.15 148	99.14 103	97.24 118	86.79 172	98.53 125	87.62 133	94.51 140	87.06 156	98.76 71	98.60 79	98.04 95	99.72 64	99.77 56

APDe-MVS		99.49 1	99.64 1	99.32 2	99.74 4	99.74 11	99.75 1	98.34 4	99.56 11	98.72 6	99.57 7	99.97 8	99.53 15	99.65 2	99.25 15	99.84 12	99.77 56

ACMMPR		99.30 9	99.54 7	99.03 16	99.66 16	99.64 26	99.68 4	98.25 14	99.56 11	97.12 30	99.19 21	99.95 17	99.72 1	99.43 16	99.25 15	99.72 64	99.77 56

Anonymous202405211			97.40 108		96.45 91	99.54 54	98.08 93	93.79 77	98.24 140		93.55 148	94.41 118	98.88 68	98.04 113	98.24 84	99.75 44	99.76 61

SMA-MVS		99.38 6	99.60 3	99.12 9	99.76 2	99.62 33	99.39 29	98.23 18	99.52 16	98.03 17	99.45 11	99.98 2	99.64 5	99.58 8	99.30 11	99.68 93	99.76 61
Yufeng Yin; Xiaoyan Liu; Zichao Zhang: SMA-MVS: Segmentation-Guided Multi-Scale Anchor Deformation Patch Multi-View Stereo. IEEE Transactions on Circuits and Systems for Video Technology
v1144		92.81 175	94.03 180	91.40 173	91.68 185	97.60 184	94.73 174	88.40 159	96.71 186	78.48 186	88.14 190	84.46 181	95.45 168	96.31 175	95.22 178	99.65 110	99.76 61

HFP-MVS		99.32 8	99.53 9	99.07 13	99.69 7	99.59 45	99.63 12	98.31 9	99.56 11	97.37 26	99.27 19	99.97 8	99.70 3	99.35 22	99.24 17	99.71 74	99.76 61

MSLP-MVS++		99.15 18	99.24 34	99.04 15	99.52 31	99.49 63	99.09 44	98.07 29	99.37 27	98.47 8	97.79 79	99.89 34	99.50 16	98.93 49	99.45 4	99.61 122	99.76 61

ACMMP		98.74 34	99.03 48	98.40 32	99.36 38	99.64 26	99.20 36	97.75 37	98.82 102	95.24 52	98.85 45	99.87 36	99.17 45	98.74 67	97.50 117	99.71 74	99.76 61
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	95.27 133	95.96 150	94.45 118	96.83 88	98.78 119	94.72 175	91.67 110	98.95 84	86.82 138	96.42 113	83.67 183	97.00 122	97.48 143	96.68 138	99.69 85	99.76 61
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
Anonymous20231211		97.10 85	97.06 122	97.14 62	96.32 94	99.52 58	98.16 87	93.76 78	98.84 99	95.98 40	90.92 168	94.58 117	98.90 63	97.72 132	98.10 92	99.71 74	99.75 68

X-MVS		98.93 29	99.37 23	98.42 31	99.67 13	99.62 33	99.60 15	98.15 23	99.08 70	93.81 81	98.46 61	99.95 17	99.59 9	99.49 13	99.21 20	99.68 93	99.75 68

NR-MVSNet		94.01 155	94.51 170	93.44 139	92.56 166	97.77 170	95.67 152	91.57 112	97.17 174	85.84 142	93.13 157	80.53 201	95.29 171	97.01 158	96.17 155	99.69 85	99.75 68

Vis-MVSNet (Re-imp)		97.40 75	98.89 54	95.66 104	95.99 105	99.62 33	97.82 98	93.22 92	98.82 102	91.40 113	96.94 97	98.56 68	95.70 157	99.14 35	99.41 6	99.79 31	99.75 68

ACMMP_NAP		99.05 25	99.45 14	98.58 30	99.73 5	99.60 43	99.64 8	98.28 13	99.23 45	94.57 63	99.35 16	99.97 8	99.55 13	99.63 3	98.66 56	99.70 82	99.74 72

v1192		92.43 186	93.61 188	91.05 179	91.53 192	97.43 193	94.61 180	87.99 164	96.60 189	76.72 191	87.11 197	82.74 191	95.85 154	96.35 173	95.30 176	99.60 130	99.74 72

PGM-MVS		98.86 31	99.35 27	98.29 34	99.77 1	99.63 29	99.67 5	95.63 45	98.66 118	95.27 51	99.11 28	99.82 41	99.67 4	99.33 24	99.19 21	99.73 57	99.74 72

CP-MVS		99.27 10	99.44 17	99.08 12	99.62 22	99.58 48	99.53 18	98.16 21	99.21 48	97.79 20	99.15 24	99.96 12	99.59 9	99.54 11	98.86 44	99.78 34	99.74 72

IterMVS-SCA-FT		94.89 139	97.87 93	91.42 171	94.86 142	97.70 173	97.24 118	84.88 186	98.93 88	75.74 195	94.26 143	98.25 72	96.69 131	98.52 85	97.68 110	99.10 185	99.73 76

v1921920		92.36 190	93.57 189	90.94 181	91.39 196	97.39 195	94.70 176	87.63 168	96.60 189	76.63 192	86.98 198	82.89 189	95.75 155	96.26 177	95.14 181	99.55 149	99.73 76

DI_MVS_plusplus_trai		96.90 92	97.49 102	96.21 91	95.61 121	99.40 76	98.72 59	92.11 100	99.14 61	92.98 97	93.08 159	95.14 108	98.13 93	98.05 112	97.91 100	99.74 49	99.73 76

v1240		91.99 193	93.33 194	90.44 186	91.29 198	97.30 198	94.25 186	86.79 172	96.43 192	75.49 198	86.34 201	81.85 195	95.29 171	96.42 170	95.22 178	99.52 156	99.73 76

thres600view7		96.69 100	96.43 145	97.00 72	96.28 98	99.67 18	98.41 73	93.99 74	97.85 158	94.29 73	95.96 120	85.91 170	99.19 40	98.26 95	97.63 111	99.82 16	99.73 76

MP-MVS		99.07 23	99.36 24	98.74 27	99.63 20	99.57 50	99.66 6	98.25 14	99.00 81	95.62 44	98.97 37	99.94 25	99.54 14	99.51 12	98.79 53	99.71 74	99.73 76
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
DTE-MVSNet		92.42 187	92.85 198	91.91 164	90.87 203	96.97 201	94.53 183	89.81 141	95.86 203	81.59 169	88.83 184	77.88 213	95.01 177	94.34 202	96.35 149	99.64 114	99.73 76

Baseline_NR-MVSNet		93.87 160	93.98 182	93.75 129	91.66 186	97.02 200	95.53 156	91.52 115	97.16 176	87.77 132	87.93 193	83.69 182	96.35 142	95.10 195	97.23 127	99.68 93	99.73 76

SixPastTwentyTwo		93.44 166	95.32 159	91.24 176	92.11 174	98.40 150	92.77 193	88.64 157	98.09 145	77.83 188	93.51 151	85.74 171	96.52 139	96.91 160	94.89 189	99.59 136	99.73 76

LGP-MVS_train		96.23 112	96.89 126	95.46 106	97.32 75	98.77 120	98.81 57	93.60 83	98.58 121	85.52 145	99.08 32	86.67 163	97.83 107	97.87 123	97.51 116	99.69 85	99.73 76

pm-mvs1		94.27 151	95.57 156	92.75 149	92.58 165	98.13 160	94.87 170	90.71 130	96.70 187	83.78 152	89.94 176	89.85 147	94.96 178	97.58 139	97.07 129	99.61 122	99.72 86

casdiffmvs		96.93 91	97.43 107	96.34 89	95.70 116	99.50 62	97.75 102	93.22 92	98.98 83	92.64 99	94.97 135	91.71 138	98.93 59	98.62 75	98.52 64	99.82 16	99.72 86
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
IterMVS		94.81 141	97.71 96	91.42 171	94.83 143	97.63 180	97.38 111	85.08 183	98.93 88	75.67 196	94.02 144	97.64 78	96.66 134	98.45 88	97.60 113	98.90 188	99.72 86
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
MCST-MVS		99.11 20	99.27 32	98.93 21	99.67 13	99.33 89	99.51 20	98.31 9	99.28 38	96.57 35	99.10 30	99.90 32	99.71 2	99.19 31	98.35 75	99.82 16	99.71 89

ACMP	96.25 10	96.62 105	96.72 130	96.50 86	96.96 84	98.75 124	97.80 99	94.30 69	98.85 95	93.12 93	98.78 48	86.61 164	97.23 119	97.73 131	96.61 141	99.62 120	99.71 89
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
DPE-MVS		99.39 5	99.55 6	99.20 4	99.63 20	99.71 15	99.66 6	98.33 6	99.29 37	98.40 11	99.64 5	99.98 2	99.31 33	99.56 9	98.96 36	99.85 10	99.70 91
Kehua Chen, Zhenlong Yuan, Tianlu Mao, Zhaoqi Wang: Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization. AAAI2025
tfpn200view9		96.75 96	96.51 138	97.03 67	96.31 95	99.67 18	98.41 73	93.99 74	97.35 168	94.52 64	95.90 122	86.93 159	99.14 47	98.26 95	97.80 106	99.82 16	99.70 91

thres400		96.71 99	96.45 143	97.02 69	96.28 98	99.63 29	98.41 73	94.00 73	97.82 159	94.42 70	95.74 126	86.26 167	99.18 43	98.20 99	97.79 107	99.81 22	99.70 91

diffmvs		96.83 93	97.33 111	96.25 90	95.76 113	99.34 86	98.06 94	93.22 92	99.43 22	92.30 106	96.90 99	89.83 148	98.55 81	98.00 116	98.14 88	99.64 114	99.70 91
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
v148		92.36 190	92.88 197	91.75 167	91.63 189	97.66 177	92.64 194	90.55 132	96.09 196	83.34 157	88.19 188	80.00 204	92.74 199	93.98 203	94.58 193	99.58 140	99.69 95

v2v482		92.77 178	93.52 192	91.90 165	91.59 191	97.63 180	94.57 182	90.31 134	96.80 185	79.22 182	88.74 185	81.55 197	96.04 151	95.26 191	94.97 185	99.66 106	99.69 95

CPTT-MVS		99.14 19	99.20 36	99.06 14	99.58 25	99.53 55	99.45 25	97.80 36	99.19 51	98.32 12	98.58 56	99.95 17	99.60 7	99.28 26	98.20 86	99.64 114	99.69 95

FMVSNet1		95.77 122	96.41 146	95.03 109	93.42 160	97.86 169	97.11 125	89.89 140	98.53 125	92.00 109	89.17 180	93.23 131	98.15 92	98.07 108	98.34 77	99.61 122	99.69 95

Vis-MVSNet		96.16 115	98.22 77	93.75 129	95.33 133	99.70 17	97.27 116	90.85 125	98.30 136	85.51 146	95.72 128	96.45 89	93.69 194	98.70 70	99.00 33	99.84 12	99.69 95
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
MVSTER		97.16 82	97.71 96	96.52 84	95.97 106	98.48 142	98.63 62	92.10 101	98.68 117	95.96 41	99.23 20	91.79 137	96.87 126	98.76 64	97.37 126	99.57 144	99.68 100

GBi-Net		96.98 89	98.00 88	95.78 100	93.81 153	97.98 162	98.09 90	91.32 118	98.80 105	93.92 77	97.21 90	95.94 101	97.89 101	98.07 108	98.34 77	99.68 93	99.67 101

test1		96.98 89	98.00 88	95.78 100	93.81 153	97.98 162	98.09 90	91.32 118	98.80 105	93.92 77	97.21 90	95.94 101	97.89 101	98.07 108	98.34 77	99.68 93	99.67 101

FMVSNet2		96.64 103	97.50 101	95.63 105	93.81 153	97.98 162	98.09 90	90.87 124	98.99 82	93.48 88	93.17 156	95.25 107	97.89 101	98.63 74	98.80 52	99.68 93	99.67 101

3Dnovator+	96.92 7	98.71 36	99.05 44	98.32 33	99.53 29	99.34 86	99.06 46	94.61 58	99.65 5	97.49 24	96.75 101	99.86 37	99.44 23	98.78 62	99.30 11	99.81 22	99.67 101

HPM-MVS++		99.10 21	99.30 30	98.86 23	99.69 7	99.48 64	99.59 16	98.34 4	99.26 42	96.55 36	99.10 30	99.96 12	99.36 28	99.25 27	98.37 74	99.64 114	99.66 105

thres200		96.76 95	96.53 136	97.03 67	96.31 95	99.67 18	98.37 76	93.99 74	97.68 164	94.49 67	95.83 125	86.77 161	99.18 43	98.26 95	97.82 105	99.82 16	99.66 105

3Dnovator	96.92 7	98.67 37	99.05 44	98.23 37	99.57 26	99.45 68	99.11 42	94.66 57	99.69 3	96.80 32	96.55 111	99.61 52	99.40 25	98.87 57	99.49 3	99.85 10	99.66 105

TSAR-MVS + GP.		98.66 39	99.36 24	97.85 44	97.16 81	99.46 66	99.03 48	94.59 61	99.09 68	97.19 29	99.73 3	99.95 17	99.39 26	98.95 47	98.69 55	99.75 44	99.65 108

FMVSNet3		97.02 88	98.12 82	95.73 103	93.59 159	97.98 162	98.34 80	91.32 118	98.80 105	93.92 77	97.21 90	95.94 101	97.63 110	98.61 76	98.62 58	99.61 122	99.65 108

CDS-MVSNet		96.59 106	98.02 87	94.92 111	94.45 146	98.96 111	97.46 110	91.75 107	97.86 157	90.07 119	96.02 119	97.25 84	96.21 144	98.04 113	98.38 72	99.60 130	99.65 108
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
ACMH+	95.51 13	95.40 129	96.00 148	94.70 113	96.33 93	98.79 117	96.79 132	91.32 118	98.77 111	87.18 135	95.60 131	85.46 173	96.97 123	97.15 154	96.59 142	99.59 136	99.65 108

QAPM		98.62 40	99.04 47	98.13 38	99.57 26	99.48 64	99.17 38	94.78 54	99.57 10	96.16 38	96.73 102	99.80 42	99.33 30	98.79 61	99.29 13	99.75 44	99.64 112

DeepC-MVS_fast	98.34 1	99.17 17	99.45 14	98.85 24	99.55 28	99.37 80	99.64 8	98.05 31	99.53 14	96.58 34	98.93 40	99.92 28	99.49 18	99.46 14	99.32 10	99.80 30	99.64 112
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
test1111		97.09 86	96.83 129	97.39 54	96.92 87	99.81 3	98.44 71	94.45 64	99.17 53	95.85 42	92.10 162	88.97 150	98.78 70	99.02 43	99.11 23	99.88 4	99.63 114

thres100view900		96.72 98	96.47 141	97.00 72	96.31 95	99.52 58	98.28 82	94.01 72	97.35 168	94.52 64	95.90 122	86.93 159	99.09 52	98.07 108	97.87 102	99.81 22	99.63 114

test2506		97.16 82	96.68 132	97.73 46	96.95 85	99.79 4	98.48 67	94.42 65	99.17 53	97.74 22	99.15 24	80.93 199	98.89 66	99.03 41	99.09 24	99.88 4	99.62 116

Effi-MVS+-dtu		95.74 123	98.04 85	93.06 146	93.92 149	99.16 101	97.90 96	88.16 163	99.07 75	82.02 167	98.02 75	94.32 120	96.74 130	98.53 84	97.56 114	99.61 122	99.62 116

HQP-MVS		96.37 109	96.58 133	96.13 93	97.31 77	98.44 146	98.45 70	95.22 49	98.86 93	88.58 125	98.33 66	87.00 158	97.67 109	97.23 151	96.56 143	99.56 147	99.62 116

ECVR-MVS		97.27 78	97.09 119	97.48 53	96.95 85	99.79 4	98.48 67	94.42 65	99.17 53	96.28 37	93.54 149	89.39 149	98.89 66	99.03 41	99.09 24	99.88 4	99.61 119

IB-MVS	93.96 15	95.02 136	96.44 144	93.36 142	97.05 83	99.28 93	90.43 202	93.39 86	98.02 147	96.02 39	94.92 137	92.07 136	83.52 211	95.38 189	95.82 166	99.72 64	99.59 120
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 35	99.11 39	98.28 35	99.36 38	99.35 84	99.48 23	97.96 33	98.83 100	93.86 80	98.70 54	99.86 37	99.44 23	99.08 39	98.38 72	99.61 122	99.58 121

CDPH-MVS		98.41 45	99.10 40	97.61 50	99.32 41	99.36 81	99.49 21	96.15 44	98.82 102	91.82 110	98.41 62	99.66 50	99.10 50	98.93 49	98.97 35	99.75 44	99.58 121

APD-MVS		99.25 12	99.38 22	99.09 11	99.69 7	99.58 48	99.56 17	98.32 8	98.85 95	97.87 19	98.91 42	99.92 28	99.30 35	99.45 15	99.38 8	99.79 31	99.58 121
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
MVS_111021_LR		98.67 37	99.41 21	97.81 45	99.37 36	99.53 55	98.51 66	95.52 47	99.27 40	94.85 58	99.56 8	99.69 49	99.04 54	99.36 20	98.88 42	99.60 130	99.58 121

PHI-MVS		99.08 22	99.43 19	98.67 28	99.15 44	99.59 45	99.11 42	97.35 39	99.14 61	97.30 27	99.44 12	99.96 12	99.32 32	98.89 54	99.39 7	99.79 31	99.58 121

MVS_111021_HR		98.59 41	99.36 24	97.68 47	99.42 34	99.61 38	98.14 88	94.81 53	99.31 34	95.00 56	99.51 9	99.79 44	99.00 57	98.94 48	98.83 49	99.69 85	99.57 126

SF-MVS		99.18 16	99.32 28	99.03 16	99.65 18	99.41 75	98.87 54	98.24 17	99.14 61	98.73 5	99.11 28	99.92 28	98.92 60	99.22 28	98.84 48	99.76 41	99.56 127

DeepPCF-MVS	97.74 3	98.34 47	99.46 13	97.04 66	98.82 50	99.33 89	96.28 144	97.47 38	99.58 9	94.70 61	98.99 36	99.85 39	97.24 118	99.55 10	99.34 9	97.73 202	99.56 127

CLD-MVS		96.74 97	96.51 138	97.01 71	96.71 89	98.62 133	98.73 58	94.38 67	98.94 86	94.46 68	97.33 87	87.03 157	98.07 95	97.20 153	96.87 134	99.72 64	99.54 129
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
CANet_DTU		96.64 103	99.08 41	93.81 128	97.10 82	99.42 73	98.85 55	90.01 137	99.31 34	79.98 179	99.78 2	99.10 62	97.42 115	98.35 92	98.05 94	99.47 161	99.53 130

pmmvs6		91.90 194	92.53 201	91.17 177	91.81 182	97.63 180	93.23 190	88.37 160	93.43 212	80.61 173	77.32 213	87.47 154	94.12 186	96.58 165	95.72 168	98.88 189	99.53 130

baseline1		97.58 68	98.05 84	97.02 69	96.21 100	99.45 68	97.71 103	93.71 82	98.47 128	95.75 43	98.78 48	93.20 132	98.91 61	98.52 85	98.44 67	99.81 22	99.53 130

FA-MVS(training)		96.52 107	98.29 71	94.45 118	95.88 109	99.52 58	97.66 105	81.47 194	98.94 86	93.79 84	95.54 133	99.11 61	98.29 88	98.89 54	96.49 145	99.63 119	99.52 133

FC-MVSNet-test		96.07 117	97.94 91	93.89 126	93.60 158	98.67 130	96.62 136	90.30 136	98.76 112	88.62 124	95.57 132	97.63 79	94.48 181	97.97 117	97.48 120	99.71 74	99.52 133

CNVR-MVS		99.23 14	99.28 31	99.17 5	99.65 18	99.34 86	99.46 24	98.21 19	99.28 38	98.47 8	98.89 44	99.94 25	99.50 16	99.42 17	98.61 59	99.73 57	99.52 133

PLC	97.93 2	99.02 28	98.94 52	99.11 10	99.46 33	99.24 97	99.06 46	97.96 33	99.31 34	99.16 1	97.90 77	99.79 44	99.36 28	98.71 69	98.12 90	99.65 110	99.52 133
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
MAR-MVS		97.71 64	98.04 85	97.32 56	99.35 40	98.91 113	97.65 106	91.68 109	98.00 148	97.01 31	97.72 83	94.83 112	98.85 69	98.44 90	98.86 44	99.41 169	99.52 133
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 25	99.08 41	99.02 18	99.62 22	99.38 77	99.43 28	98.21 19	99.36 30	97.66 23	97.79 79	99.90 32	99.45 22	99.17 32	98.43 69	99.77 39	99.51 138

OpenMVS	96.23 11	97.95 58	98.45 67	97.35 55	99.52 31	99.42 73	98.91 53	94.61 58	98.87 92	92.24 108	94.61 139	99.05 63	99.10 50	98.64 73	99.05 29	99.74 49	99.51 138

Fast-Effi-MVS+-dtu		95.38 130	98.20 78	92.09 157	93.91 150	98.87 114	97.35 113	85.01 185	99.08 70	81.09 171	98.10 71	96.36 92	95.62 160	98.43 91	97.03 130	99.55 149	99.50 140

ACMM	96.26 9	96.67 102	96.69 131	96.66 78	97.29 78	98.46 144	96.48 140	95.09 50	99.21 48	93.19 92	98.78 48	86.73 162	98.17 89	97.84 125	96.32 150	99.74 49	99.49 141
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
CHOSEN 280x420		97.99 57	99.24 34	96.53 83	98.34 59	99.61 38	98.36 78	89.80 143	99.27 40	95.08 55	99.81 1	98.58 67	98.64 76	99.02 43	98.92 39	98.93 187	99.48 142

TAPA-MVS	97.53 5	98.41 45	98.84 57	97.91 43	99.08 46	99.33 89	99.15 39	97.13 40	99.34 32	93.20 91	97.75 81	99.19 59	99.20 39	98.66 71	98.13 89	99.66 106	99.48 142
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
GA-MVS		93.93 159	96.31 147	91.16 178	93.61 157	98.79 117	95.39 160	90.69 131	98.25 139	73.28 204	96.15 117	88.42 152	94.39 183	97.76 129	95.35 174	99.58 140	99.45 144

CVMVSNet		95.33 132	97.09 119	93.27 144	95.23 134	98.39 151	95.49 157	92.58 98	97.71 163	83.00 161	94.44 142	93.28 130	93.92 191	97.79 126	98.54 63	99.41 169	99.45 144

LS3D		97.79 60	98.25 73	97.26 60	98.40 58	99.63 29	99.53 18	98.63 1	99.25 44	88.13 127	96.93 98	94.14 122	99.19 40	99.14 35	99.23 18	99.69 85	99.42 146

ET-MVSNet_ETH3D		96.17 114	96.99 124	95.21 108	88.53 209	98.54 139	98.28 82	92.61 97	98.85 95	93.60 87	99.06 34	90.39 142	98.63 77	95.98 184	96.68 138	99.61 122	99.41 147

baseline2		96.36 110	97.82 94	94.65 114	94.60 145	99.09 104	96.45 141	89.63 145	98.36 133	91.29 115	97.60 86	94.13 123	96.37 141	98.45 88	97.70 109	99.54 153	99.41 147

test0.0.03 1		96.69 100	98.12 82	95.01 110	95.49 128	98.99 108	95.86 150	90.82 126	98.38 131	92.54 103	96.66 105	97.33 81	95.75 155	97.75 130	98.34 77	99.60 130	99.40 149

testgi		95.67 124	97.48 103	93.56 135	95.07 137	99.00 106	95.33 161	88.47 158	98.80 105	86.90 137	97.30 89	92.33 134	95.97 152	97.66 134	97.91 100	99.60 130	99.38 150

TAMVS		95.53 126	96.50 140	94.39 120	93.86 152	99.03 105	96.67 134	89.55 147	97.33 170	90.64 117	93.02 160	91.58 139	96.21 144	97.72 132	97.43 124	99.43 166	99.36 151

AdaColmap		99.06 24	98.98 51	99.15 6	99.60 24	99.30 92	99.38 30	98.16 21	99.02 79	98.55 7	98.71 53	99.57 55	99.58 12	99.09 37	97.84 104	99.64 114	99.36 151

PM-MVS		89.55 202	90.30 207	88.67 197	87.06 210	95.60 209	90.88 200	84.51 189	96.14 195	75.75 194	86.89 199	63.47 221	94.64 180	96.85 161	93.89 197	99.17 183	99.29 153

DPM-MVS		98.31 49	98.53 64	98.05 39	98.76 53	98.77 120	99.13 40	98.07 29	99.10 67	94.27 74	96.70 103	99.84 40	98.70 72	97.90 121	98.11 91	99.40 171	99.28 154

pmmvs4		95.09 134	95.90 151	94.14 122	92.29 171	97.70 173	95.45 158	90.31 134	98.60 119	90.70 116	93.25 154	89.90 146	96.67 133	97.13 155	95.42 173	99.44 165	99.28 154

EG-PatchMatch MVS		92.45 183	93.92 184	90.72 184	92.56 166	98.43 148	94.88 169	84.54 188	97.18 173	79.55 181	86.12 202	83.23 187	93.15 198	97.22 152	96.00 160	99.67 101	99.27 156

UA-Net		97.13 84	99.14 38	94.78 112	97.21 79	99.38 77	97.56 107	92.04 102	98.48 127	88.03 128	98.39 64	99.91 31	94.03 188	99.33 24	99.23 18	99.81 22	99.25 157

pmmvs-eth3d		89.81 201	89.65 208	90.00 190	86.94 211	95.38 210	91.08 198	86.39 177	94.57 208	82.27 166	83.03 208	64.94 218	93.96 189	96.57 166	93.82 198	99.35 174	99.24 158

gg-mvs-nofinetune		90.85 196	94.14 175	87.02 201	94.89 141	99.25 95	98.64 61	76.29 215	88.24 216	57.50 220	79.93 211	95.45 104	95.18 174	98.77 63	98.07 93	99.62 120	99.24 158

PMMVS		97.52 70	98.39 68	96.51 85	95.82 112	98.73 127	97.80 99	93.05 96	98.76 112	94.39 72	99.07 33	97.03 87	98.55 81	98.31 94	97.61 112	99.43 166	99.21 160

CNLPA		99.03 27	99.05 44	99.01 19	99.27 42	99.22 99	99.03 48	97.98 32	99.34 32	99.00 4	98.25 68	99.71 48	99.31 33	98.80 60	98.82 51	99.48 159	99.17 161

CR-MVSNet		94.57 149	97.34 110	91.33 174	94.90 140	98.59 136	97.15 122	79.14 205	97.98 149	80.42 175	96.59 110	93.50 129	96.85 127	98.10 103	97.49 118	99.50 158	99.15 162

PatchT		93.96 158	97.36 109	90.00 190	94.76 144	98.65 131	90.11 205	78.57 210	97.96 152	80.42 175	96.07 118	94.10 124	96.85 127	98.10 103	97.49 118	99.26 179	99.15 162

COLMAP_ROB	96.15 12	97.78 61	98.17 79	97.32 56	98.84 49	99.45 68	99.28 33	95.43 48	99.48 19	91.80 111	94.83 138	98.36 71	98.90 63	98.09 105	97.85 103	99.68 93	99.15 162
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
MSDG		98.27 50	98.29 71	98.24 36	99.20 43	99.22 99	99.20 36	97.82 35	99.37 27	94.43 69	95.90 122	97.31 82	99.12 48	98.76 64	98.35 75	99.67 101	99.14 165

test-mter		94.86 140	97.32 112	92.00 160	92.41 169	98.82 116	96.18 147	86.35 178	98.05 146	82.28 165	96.48 112	94.39 119	95.46 167	98.17 101	96.20 154	99.32 176	99.13 166

RPMNet		94.66 143	97.16 118	91.75 167	94.98 139	98.59 136	97.00 129	78.37 211	97.98 149	83.78 152	96.27 115	94.09 125	96.91 125	97.36 146	96.73 136	99.48 159	99.09 167

OMC-MVS		98.84 32	99.01 50	98.65 29	99.39 35	99.23 98	99.22 35	96.70 41	99.40 24	97.77 21	97.89 78	99.80 42	99.21 38	99.02 43	98.65 57	99.57 144	99.07 168

TSAR-MVS + COLMAP		96.79 94	96.55 135	97.06 65	97.70 70	98.46 144	99.07 45	96.23 43	99.38 25	91.32 114	98.80 46	85.61 172	98.69 74	97.64 137	96.92 133	99.37 173	99.06 169

tpm		92.38 188	94.79 165	89.56 194	94.30 147	97.50 190	94.24 187	78.97 208	97.72 162	74.93 200	97.97 76	82.91 188	96.60 136	93.65 204	94.81 190	98.33 194	98.98 170

PatchMatch-RL		97.77 62	98.25 73	97.21 61	99.11 45	99.25 95	97.06 128	94.09 71	98.72 116	95.14 54	98.47 60	96.29 93	98.43 85	98.65 72	97.44 123	99.45 163	98.94 171

pmmvs5		92.71 181	94.27 174	90.90 182	91.42 195	97.74 172	93.23 190	86.66 175	95.99 200	78.96 185	91.45 165	83.44 185	95.55 162	97.30 149	95.05 183	99.58 140	98.93 172

test-LLR		95.50 127	97.32 112	93.37 141	95.49 128	98.74 125	96.44 142	90.82 126	98.18 141	82.75 162	96.60 108	94.67 115	95.54 163	98.09 105	96.00 160	99.20 181	98.93 172

TESTMET0.1,1		94.95 137	97.32 112	92.20 155	92.62 164	98.74 125	96.44 142	86.67 174	98.18 141	82.75 162	96.60 108	94.67 115	95.54 163	98.09 105	96.00 160	99.20 181	98.93 172

EU-MVSNet		92.80 176	94.76 166	90.51 185	91.88 179	96.74 205	92.48 195	88.69 155	96.21 193	79.00 184	91.51 164	87.82 153	91.83 203	95.87 186	96.27 151	99.21 180	98.92 175

PCF-MVS	97.50 6	98.18 53	98.35 70	97.99 41	98.65 54	99.36 81	98.94 52	98.14 25	98.59 120	93.62 86	96.61 107	99.76 47	99.03 55	97.77 128	97.45 122	99.57 144	98.89 176
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
EPNet_dtu		96.30 111	98.53 64	93.70 132	98.97 48	98.24 157	97.36 112	94.23 70	98.85 95	79.18 183	99.19 21	98.47 69	94.09 187	97.89 122	98.21 85	98.39 193	98.85 177
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
OPM-MVS		96.22 113	95.85 154	96.65 79	97.75 68	98.54 139	99.00 51	95.53 46	96.88 181	89.88 121	95.95 121	86.46 166	98.07 95	97.65 136	96.63 140	99.67 101	98.83 178
Ray L. Khuboni, Hongjun Xu: Octagram Propagation Matching for Multi-Scale View Stereopsis (OPM-MVS).
GG-mvs-BLEND		69.11 213	98.13 81	35.26 217	3.49 226	98.20 159	94.89 168	2.38 223	98.42 130	5.82 227	96.37 114	98.60 66	5.97 222	98.75 66	97.98 96	99.01 186	98.61 179

ambc			80.99 213		80.04 218	90.84 215	90.91 199		96.09 196	74.18 201	62.81 216	30.59 227	82.44 212	96.25 178	91.77 208	95.91 215	98.56 180

MDTV_nov1_ep13_2view		92.44 184	95.66 155	88.68 196	91.05 202	97.92 166	92.17 196	79.64 201	98.83 100	76.20 193	91.45 165	93.51 128	95.04 176	95.68 188	93.70 199	97.96 198	98.53 181

USDC		94.26 152	94.83 164	93.59 134	96.02 102	98.44 146	97.84 97	88.65 156	98.86 93	82.73 164	94.02 144	80.56 200	96.76 129	97.28 150	96.15 157	99.55 149	98.50 182

MDA-MVSNet-bldmvs		87.84 206	89.22 209	86.23 203	81.74 215	96.77 204	83.74 215	89.57 146	94.50 209	72.83 208	96.64 106	64.47 220	92.71 200	81.43 216	92.28 205	96.81 211	98.47 183

test_method		87.27 207	91.58 203	82.25 209	75.65 220	87.52 219	86.81 213	72.60 218	97.51 166	73.20 205	85.07 204	79.97 205	88.69 206	97.31 148	95.24 177	96.53 212	98.41 184

gm-plane-assit		89.44 203	92.82 200	85.49 205	91.37 197	95.34 211	79.55 219	82.12 193	91.68 215	64.79 217	87.98 191	80.26 203	95.66 158	98.51 87	97.56 114	99.45 163	98.41 184

MS-PatchMatch		95.99 118	97.26 116	94.51 116	97.46 72	98.76 123	97.27 116	86.97 171	99.09 68	89.83 122	93.51 151	97.78 77	96.18 146	97.53 141	95.71 169	99.35 174	98.41 184

TransMVSNet (Re)		93.45 165	94.08 178	92.72 150	92.83 162	97.62 183	94.94 166	91.54 114	95.65 204	83.06 160	88.93 183	83.53 184	94.25 184	97.41 144	97.03 130	99.67 101	98.40 187

TinyColmap		94.00 156	94.35 173	93.60 133	95.89 107	98.26 155	97.49 109	88.82 153	98.56 123	83.21 158	91.28 167	80.48 202	96.68 132	97.34 147	96.26 153	99.53 155	98.24 188

TDRefinement		93.04 172	93.57 189	92.41 151	96.58 90	98.77 120	97.78 101	91.96 105	98.12 144	80.84 172	89.13 182	79.87 207	87.78 207	96.44 168	94.50 194	99.54 153	98.15 189

MDTV_nov1_ep13		95.57 125	97.48 103	93.35 143	95.43 130	98.97 110	97.19 121	83.72 192	98.92 90	87.91 130	97.75 81	96.12 98	97.88 104	96.84 162	95.64 170	97.96 198	98.10 190

MIMVSNet		94.49 150	97.59 100	90.87 183	91.74 184	98.70 129	94.68 177	78.73 209	97.98 149	83.71 155	97.71 84	94.81 113	96.96 124	97.97 117	97.92 98	99.40 171	98.04 191

CostFormer		94.25 153	94.88 163	93.51 138	95.43 130	98.34 154	96.21 146	80.64 197	97.94 153	94.01 75	98.30 67	86.20 169	97.52 111	92.71 206	92.69 202	97.23 209	98.02 192

pmnet_mix02		92.44 184	94.68 167	89.83 193	92.46 168	97.65 179	89.92 207	90.49 133	98.76 112	73.05 206	91.78 163	90.08 145	94.86 179	94.53 200	91.94 207	98.21 196	98.01 193

RPSCF		97.61 67	98.16 80	96.96 74	98.10 62	99.00 106	98.84 56	93.76 78	99.45 20	94.78 60	99.39 15	99.31 57	98.53 83	96.61 163	95.43 172	97.74 200	97.93 194

Anonymous20231206		90.70 198	93.93 183	86.92 202	90.21 207	96.79 203	90.30 204	86.61 176	96.05 198	69.25 211	88.46 187	84.86 179	85.86 209	97.11 156	96.47 147	99.30 177	97.80 195

SCA		94.95 137	97.44 106	92.04 158	95.55 125	99.16 101	96.26 145	79.30 204	99.02 79	85.73 144	98.18 69	97.13 85	97.69 108	96.03 182	94.91 186	97.69 203	97.65 196

pmmvs3		88.19 205	91.27 204	84.60 207	85.60 213	93.66 214	85.68 214	81.13 195	92.36 214	63.66 219	89.51 178	77.10 214	93.22 197	96.37 171	92.40 203	98.30 195	97.46 197

N_pmnet		92.21 192	94.60 169	89.42 195	91.88 179	97.38 196	89.15 209	89.74 144	97.89 155	73.75 202	87.94 192	92.23 135	93.85 192	96.10 180	93.20 201	98.15 197	97.43 198

PatchmatchNet		94.70 142	97.08 121	91.92 163	95.53 126	98.85 115	95.77 151	79.54 202	98.95 84	85.98 141	98.52 57	96.45 89	97.39 116	95.32 190	94.09 196	97.32 206	97.38 199
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
ADS-MVSNet		94.65 144	97.04 123	91.88 166	95.68 118	98.99 108	95.89 149	79.03 207	99.15 58	85.81 143	96.96 96	98.21 74	97.10 120	94.48 201	94.24 195	97.74 200	97.21 200

MVS-HIRNet		92.51 182	95.97 149	88.48 198	93.73 156	98.37 152	90.33 203	75.36 217	98.32 135	77.78 189	89.15 181	94.87 111	95.14 175	97.62 138	96.39 148	98.51 190	97.11 201

dps		94.63 145	95.31 160	93.84 127	95.53 126	98.71 128	96.54 137	80.12 199	97.81 161	97.21 28	96.98 95	92.37 133	96.34 143	92.46 208	91.77 208	97.26 208	97.08 202

test20.03		90.65 199	93.71 187	87.09 200	90.44 205	96.24 206	89.74 208	85.46 182	95.59 205	72.99 207	90.68 171	85.33 174	84.41 210	95.94 185	95.10 182	99.52 156	97.06 203

EPMVS		95.05 135	96.86 128	92.94 148	95.84 110	98.96 111	96.68 133	79.87 200	99.05 76	90.15 118	97.12 94	95.99 100	97.49 113	95.17 193	94.75 191	97.59 204	96.96 204

tpmrst		93.86 161	95.88 152	91.50 170	95.69 117	98.62 133	95.64 154	79.41 203	98.80 105	83.76 154	95.63 130	96.13 97	97.25 117	92.92 205	92.31 204	97.27 207	96.74 205

new-patchmatchnet		86.12 208	87.30 210	84.74 206	86.92 212	95.19 213	83.57 216	84.42 190	92.67 213	65.66 214	80.32 210	64.72 219	89.41 205	92.33 210	89.21 212	98.43 192	96.69 206

tpm cat1		94.06 154	94.90 162	93.06 146	95.42 132	98.52 141	96.64 135	80.67 196	97.82 159	92.63 100	93.39 153	95.00 110	96.06 150	91.36 211	91.58 210	96.98 210	96.66 207

FMVSNet5		95.42 128	96.47 141	94.20 121	92.26 172	95.99 208	95.66 153	87.15 170	97.87 156	93.46 89	96.68 104	93.79 126	97.52 111	97.10 157	97.21 128	99.11 184	96.62 208

DeepMVS_CX						96.85 202	87.43 212	89.27 148	98.30 136	75.55 197	95.05 134	79.47 208	92.62 201	89.48 212		95.18 216	95.96 209

MIMVSNet1		88.61 204	90.68 206	86.19 204	81.56 216	95.30 212	87.78 211	85.98 180	94.19 210	72.30 209	78.84 212	78.90 211	90.06 204	96.59 164	95.47 171	99.46 162	95.49 210

CMPMVS	70.31 18	90.74 197	91.06 205	90.36 188	97.32 75	97.43 193	92.97 192	87.82 167	93.50 211	75.34 199	83.27 207	84.90 178	92.19 202	92.64 207	91.21 211	96.50 213	94.46 211
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
new_pmnet		90.45 200	92.84 199	87.66 199	88.96 208	96.16 207	88.71 210	84.66 187	97.56 165	71.91 210	85.60 203	86.58 165	93.28 196	96.07 181	93.54 200	98.46 191	94.39 212

Gipuma		81.40 210	81.78 212	80.96 211	83.21 214	85.61 220	79.73 218	76.25 216	97.33 170	64.21 218	55.32 217	55.55 222	86.04 208	92.43 209	92.20 206	96.32 214	93.99 213
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
PMMVS2		77.26 211	79.47 214	74.70 213	76.00 219	88.37 218	74.22 220	76.34 214	78.31 218	54.13 221	69.96 215	52.50 223	70.14 217	84.83 214	88.71 213	97.35 205	93.58 214

MVE	67.97 19	65.53 216	67.43 218	63.31 216	59.33 223	74.20 221	53.09 225	70.43 219	66.27 221	43.13 222	45.98 221	30.62 226	70.65 216	79.34 218	86.30 214	83.25 222	89.33 215
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
FPMVS		83.82 209	84.61 211	82.90 208	90.39 206	90.71 216	90.85 201	84.10 191	95.47 206	65.15 215	83.44 206	74.46 216	75.48 213	81.63 215	79.42 217	91.42 218	87.14 216

EMVS		68.12 215	68.11 217	68.14 215	75.51 221	71.76 222	55.38 224	77.20 213	77.78 219	37.79 224	53.59 218	43.61 224	74.72 214	67.05 220	76.70 219	88.27 221	86.24 217

E-PMN		68.30 214	68.43 216	68.15 214	74.70 222	71.56 223	55.64 223	77.24 212	77.48 220	39.46 223	51.95 220	41.68 225	73.28 215	70.65 219	79.51 216	88.61 220	86.20 218

PMVS	72.60 17	76.39 212	77.66 215	74.92 212	81.04 217	69.37 224	68.47 221	80.54 198	85.39 217	65.07 216	73.52 214	72.91 217	65.67 219	80.35 217	76.81 218	88.71 219	85.25 219
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
testmvs		31.24 217	40.15 219	20.86 218	12.61 224	17.99 225	25.16 226	13.30 221	48.42 222	24.82 225	53.07 219	30.13 228	28.47 220	42.73 221	37.65 220	20.79 223	51.04 220

test123		26.75 218	34.25 220	18.01 219	7.93 225	17.18 226	24.85 227	12.36 222	44.83 223	16.52 226	41.80 222	18.10 229	28.29 221	33.08 222	34.79 221	18.10 224	49.95 221

uanet_test		0.00 219	0.00 221	0.00 220	0.00 227	0.00 227	0.00 228	0.00 224	0.00 224	0.00 228	0.00 223	0.00 230	0.00 223	0.00 223	0.00 222	0.00 225	0.00 222

sosnet-low-res		0.00 219	0.00 221	0.00 220	0.00 227	0.00 227	0.00 228	0.00 224	0.00 224	0.00 228	0.00 223	0.00 230	0.00 223	0.00 223	0.00 222	0.00 225	0.00 222

sosnet		0.00 219	0.00 221	0.00 220	0.00 227	0.00 227	0.00 228	0.00 224	0.00 224	0.00 228	0.00 223	0.00 230	0.00 223	0.00 223	0.00 222	0.00 225	0.00 222

RE-MVS-def										69.05 212

9.14												99.79 44

SR-MVS					99.67 13			98.25 14				99.94 25

our_test_3					92.30 170	97.58 185	90.09 206

MTAPA										98.09 15		99.97 8

MTMP										98.46 10		99.96 12

Patchmatch-RL test							66.86 222

tmp_tt				82.25 209	97.73 69	88.71 217	80.18 217	68.65 220	99.15 58	86.98 136	99.47 10	85.31 175	68.35 218	87.51 213	83.81 215	91.64 217

XVS					97.42 73	99.62 33	98.59 64			93.81 81		99.95 17				99.69 85

X-MVStestdata					97.42 73	99.62 33	98.59 64			93.81 81		99.95 17				99.69 85

mPP-MVS					99.53 29							99.89 34

NP-MVS									98.57 122

Patchmtry						98.59 136	97.15 122	79.14 205		80.42 175

High-resolution multi-view benchmark