High-res multi-view results

This table lists the benchmark results for the high-res multi-view scenario. The following metrics are evaluated:

Accuracy [%]: The fraction of the reconstruction which is closer to the ground truth than the evaluation threshold distance (*). Larger is better.
Completeness [%]: The fraction of the ground truth which is closer to the reconstruction than the evaluation threshold distance (*). Larger is better.
F₁ score [%]: The harmonic mean of accuracy and completeness, used to rank methods based on both metrics. Larger is better.
Time [s]: The runtime of the method. Smaller is better.

(*) For exact definitions, detailing how potentially incomplete ground truth is taken into account, see our paper.

The datasets are grouped into different categories, and result averages are computed for a category and method if results of the method are available for all datasets within the category. Note that the category "all" includes both the high-res multi-view and the low-res many-view scenarios.

Methods with suffix _ROB may participate in the Robust Vision Challenge.

Click a dataset result cell to show a visualization of the reconstruction. For training datasets, ground truth and accuracy / completeness visualizations are also available. The visualizations may not work with mobile browsers.

Method	all	high-res multi-view	indoor	outdoor	botani.	boulde.	bridge	door	exhibi.	lectur.	living.	lounge	observ.	old co.	statue	terrac.

APDe-MVS		97.82 4	97.73 3	98.08 15	99.15 33	94.82 25	98.81 2	98.30 23	94.76 34	98.30 13	98.90 3	93.77 14	99.68 47	97.93 1	99.69 3	99.75 3

SED-MVS		98.05 1	97.99 1	98.24 7	99.42 6	95.30 15	98.25 29	98.27 28	95.13 17	99.19 1	98.89 4	95.54 3	99.85 14	97.52 2	99.66 8	99.56 22

test_241102_TWO								98.27 28	95.13 17	98.93 6	98.89 4	94.99 8	99.85 14	97.52 2	99.65 10	99.74 5

DVP-MVS		97.91 2	97.81 2	98.22 9	99.45 2	95.36 10	98.21 36	97.85 111	94.92 24	98.73 8	98.87 6	95.08 5	99.84 19	97.52 2	99.67 6	99.48 41
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
test_0728_SECOND				98.51 2	99.45 2	95.93 3	98.21 36	98.28 26					99.86 8	97.52 2	99.67 6	99.75 3

test_0728_THIRD									94.78 33	98.73 8	98.87 6	95.87 2	99.84 19	97.45 6	99.72 2	99.77 1

IU-MVS					99.42 6	95.39 9		97.94 102	90.40 174	98.94 5				97.41 7	99.66 8	99.74 5

DROMVSNet		96.25 59	96.29 53	96.13 112	96.87 159	91.35 127	98.66 6	97.74 118	93.91 53	96.29 72	97.43 112	89.36 87	98.59 183	97.23 8	99.07 81	98.45 133

TSAR-MVS + MP.		97.42 8	97.33 9	97.69 42	99.25 27	94.24 39	98.07 44	97.85 111	93.72 60	98.57 11	98.35 38	93.69 15	99.40 109	97.06 9	99.46 38	99.44 47
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
CNVR-MVS		97.68 5	97.44 8	98.37 5	98.90 51	95.86 4	97.27 125	98.08 64	95.81 3	97.87 23	98.31 47	94.26 10	99.68 47	97.02 10	99.49 34	99.57 19

SD-MVS		97.41 9	97.53 6	97.06 71	98.57 72	94.46 30	97.92 57	98.14 53	94.82 30	99.01 3	98.55 19	94.18 11	97.41 302	96.94 11	99.64 11	99.32 60
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
Regformer-4		96.97 29	96.87 21	97.25 61	98.34 83	92.66 85	96.96 155	98.01 91	95.12 19	97.14 41	98.42 31	91.82 46	99.61 62	96.90 12	99.13 75	99.50 37

CANet		96.39 55	96.02 60	97.50 50	97.62 129	93.38 67	97.02 147	97.96 100	95.42 7	94.86 114	97.81 82	87.38 116	99.82 25	96.88 13	99.20 70	99.29 62

TSAR-MVS + GP.		96.69 45	96.49 46	97.27 60	98.31 88	93.39 66	96.79 171	96.72 228	94.17 47	97.44 29	97.66 93	92.76 23	99.33 114	96.86 14	97.76 120	99.08 80

CS-MVS		95.88 70	95.98 61	95.58 142	96.44 184	90.56 157	97.78 69	97.73 119	93.01 86	96.07 81	96.77 140	90.13 80	98.57 184	96.83 15	99.10 80	97.60 179

Regformer-3		96.85 38	96.80 29	97.01 72	98.34 83	92.02 108	96.96 155	97.76 115	95.01 23	97.08 46	98.42 31	91.71 49	99.54 86	96.80 16	99.13 75	99.48 41

Regformer-2		97.16 18	96.99 16	97.67 43	98.32 86	93.84 53	96.83 167	98.10 61	95.24 11	97.49 26	98.25 54	92.57 30	99.61 62	96.80 16	99.29 57	99.56 22

Regformer-1		97.10 20	96.96 18	97.54 49	98.32 86	93.48 64	96.83 167	97.99 97	95.20 13	97.46 27	98.25 54	92.48 34	99.58 71	96.79 18	99.29 57	99.55 26

DeepPCF-MVS	93.97 1	96.61 48	97.09 12	95.15 162	98.09 105	86.63 266	96.00 236	98.15 51	95.43 6	97.95 19	98.56 17	93.40 16	99.36 113	96.77 19	99.48 35	99.45 45

SMA-MVS		97.35 12	97.03 14	98.30 6	99.06 40	95.42 8	97.94 55	98.18 46	90.57 170	98.85 7	98.94 1	93.33 17	99.83 22	96.72 20	99.68 4	99.63 11
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
DPE-MVS		97.86 3	97.65 4	98.47 3	99.17 32	95.78 5	97.21 134	98.35 19	95.16 16	98.71 10	98.80 9	95.05 7	99.89 3	96.70 21	99.73 1	99.73 7
Kehua Chen, Zhenlong Yuan, Tianlu Mao, Zhaoqi Wang: Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization. AAAI2025
MSP-MVS		97.59 7	97.54 5	97.73 38	99.40 11	93.77 58	98.53 10	98.29 24	95.55 5	98.56 12	97.81 82	93.90 12	99.65 53	96.62 22	99.21 69	99.77 1
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
MSLP-MVS++		96.94 32	97.06 13	96.59 83	98.72 59	91.86 112	97.67 83	98.49 12	94.66 37	97.24 36	98.41 34	92.31 37	98.94 150	96.61 23	99.46 38	98.96 92

MP-MVS-pluss		96.70 44	96.27 54	97.98 21	99.23 30	94.71 26	96.96 155	98.06 73	90.67 161	95.55 103	98.78 10	91.07 65	99.86 8	96.58 24	99.55 21	99.38 56
MP-MVS-pluss: MP-MVS-pluss. MP-MVS-pluss
SteuartSystems-ACMMP		97.62 6	97.53 6	97.87 24	98.39 80	94.25 38	98.43 18	98.27 28	95.34 10	98.11 16	98.56 17	94.53 9	99.71 38	96.57 25	99.62 13	99.65 9
Skip Steuart: Steuart Systems R&D Blog.
MCST-MVS		97.18 16	96.84 25	98.20 10	99.30 24	95.35 12	97.12 142	98.07 70	93.54 68	96.08 80	97.69 90	93.86 13	99.71 38	96.50 26	99.39 47	99.55 26

xxxxxxxxxxxxxcwj		97.36 11	97.20 10	97.83 26	98.91 49	94.28 35	97.02 147	97.22 185	95.35 8	98.27 14	98.65 13	93.33 17	99.72 35	96.49 27	99.52 25	99.51 34

SF-MVS		97.39 10	97.13 11	98.17 11	99.02 43	95.28 17	98.23 33	98.27 28	92.37 110	98.27 14	98.65 13	93.33 17	99.72 35	96.49 27	99.52 25	99.51 34

EI-MVSNet-Vis-set		96.51 51	96.47 47	96.63 80	98.24 93	91.20 135	96.89 162	97.73 119	94.74 35	96.49 64	98.49 24	90.88 70	99.58 71	96.44 29	98.32 104	99.13 74

VDD-MVS		93.82 127	93.08 133	96.02 119	97.88 116	89.96 176	97.72 78	95.85 269	92.43 108	95.86 90	98.44 28	68.42 328	99.39 110	96.31 30	94.85 177	98.71 115

test1172		96.93 33	96.86 22	97.15 67	99.10 34	92.34 94	97.96 54	98.04 81	93.79 58	97.35 33	98.53 21	91.40 57	99.56 81	96.30 31	99.30 56	99.55 26

ACMMP_NAP		97.20 15	96.86 22	98.23 8	99.09 36	95.16 20	97.60 94	98.19 44	92.82 97	97.93 20	98.74 11	91.60 53	99.86 8	96.26 32	99.52 25	99.67 8

diffmvs		95.25 85	95.13 83	95.63 138	96.43 186	89.34 197	95.99 237	97.35 176	92.83 96	96.31 71	97.37 114	86.44 127	98.67 173	96.26 32	97.19 137	98.87 103

CS-MVS-test		95.61 75	95.62 68	95.58 142	96.33 192	91.02 142	97.64 90	97.68 128	92.69 102	95.18 110	95.91 190	89.95 84	98.61 179	96.24 34	98.92 90	97.12 190

EI-MVSNet-UG-set		96.34 56	96.30 52	96.47 91	98.20 98	90.93 147	96.86 163	97.72 123	94.67 36	96.16 77	98.46 26	90.43 76	99.58 71	96.23 35	97.96 114	98.90 99

SR-MVS		97.01 28	96.86 22	97.47 51	99.09 36	93.27 71	97.98 49	98.07 70	93.75 59	97.45 28	98.48 25	91.43 56	99.59 68	96.22 36	99.27 61	99.54 29

xiu_mvs_v1_base_debu		95.01 91	94.76 89	95.75 130	96.58 173	91.71 113	96.25 221	97.35 176	92.99 87	96.70 52	96.63 154	82.67 186	99.44 104	96.22 36	97.46 124	96.11 220

xiu_mvs_v1_base		95.01 91	94.76 89	95.75 130	96.58 173	91.71 113	96.25 221	97.35 176	92.99 87	96.70 52	96.63 154	82.67 186	99.44 104	96.22 36	97.46 124	96.11 220

xiu_mvs_v1_base_debi		95.01 91	94.76 89	95.75 130	96.58 173	91.71 113	96.25 221	97.35 176	92.99 87	96.70 52	96.63 154	82.67 186	99.44 104	96.22 36	97.46 124	96.11 220

alignmvs		95.87 71	95.23 80	97.78 33	97.56 134	95.19 18	97.86 60	97.17 188	94.39 43	96.47 66	96.40 168	85.89 135	99.20 123	96.21 40	95.11 175	98.95 94

canonicalmvs		96.02 65	95.45 73	97.75 37	97.59 132	95.15 21	98.28 26	97.60 137	94.52 40	96.27 74	96.12 180	87.65 109	99.18 126	96.20 41	94.82 179	98.91 98

zzz-MVS		97.07 22	96.77 32	97.97 22	99.37 16	94.42 32	97.15 140	98.08 64	95.07 21	96.11 78	98.59 15	90.88 70	99.90 1	96.18 42	99.50 32	99.58 17

MTAPA		97.08 21	96.78 31	97.97 22	99.37 16	94.42 32	97.24 127	98.08 64	95.07 21	96.11 78	98.59 15	90.88 70	99.90 1	96.18 42	99.50 32	99.58 17

APD-MVS_3200maxsize		96.81 40	96.71 36	97.12 69	99.01 46	92.31 97	97.98 49	98.06 73	93.11 83	97.44 29	98.55 19	90.93 68	99.55 84	96.06 44	99.25 65	99.51 34

SR-MVS-dyc-post		96.88 36	96.80 29	97.11 70	99.02 43	92.34 94	97.98 49	98.03 84	93.52 69	97.43 31	98.51 22	91.40 57	99.56 81	96.05 45	99.26 63	99.43 49

RE-MVS-def			96.72 35		99.02 43	92.34 94	97.98 49	98.03 84	93.52 69	97.43 31	98.51 22	90.71 73		96.05 45	99.26 63	99.43 49

MVS_111021_HR		96.68 47	96.58 42	96.99 73	98.46 74	92.31 97	96.20 226	98.90 2	94.30 46	95.86 90	97.74 87	92.33 35	99.38 112	96.04 47	99.42 43	99.28 65

PHI-MVS		96.77 42	96.46 48	97.71 41	98.40 78	94.07 48	98.21 36	98.45 15	89.86 182	97.11 44	98.01 68	92.52 32	99.69 44	96.03 48	99.53 24	99.36 58

HPM-MVS++		97.34 13	96.97 17	98.47 3	99.08 38	96.16 2	97.55 98	97.97 99	95.59 4	96.61 58	97.89 72	92.57 30	99.84 19	95.95 49	99.51 29	99.40 53

DELS-MVS		96.61 48	96.38 51	97.30 57	97.79 121	93.19 72	95.96 238	98.18 46	95.23 12	95.87 89	97.65 94	91.45 55	99.70 43	95.87 50	99.44 42	99.00 90
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
MVS_111021_LR		96.24 60	96.19 58	96.39 98	98.23 97	91.35 127	96.24 224	98.79 4	93.99 51	95.80 92	97.65 94	89.92 85	99.24 121	95.87 50	99.20 70	98.58 119

hse-mvs3		94.15 112	93.52 120	96.04 118	97.81 119	90.22 167	97.62 93	97.58 140	95.19 14	96.74 50	97.45 109	83.67 164	99.61 62	95.85 52	79.73 332	98.29 147

hse-mvs2		93.45 139	92.99 135	94.81 179	97.02 154	88.59 218	96.69 181	96.47 246	95.19 14	96.74 50	96.16 179	83.67 164	98.48 193	95.85 52	79.13 336	97.35 187

NCCC		97.30 14	97.03 14	98.11 14	98.77 57	95.06 22	97.34 117	98.04 81	95.96 2	97.09 45	97.88 74	93.18 20	99.71 38	95.84 54	99.17 72	99.56 22

VNet		95.89 69	95.45 73	97.21 65	98.07 107	92.94 79	97.50 101	98.15 51	93.87 54	97.52 25	97.61 100	85.29 142	99.53 89	95.81 55	95.27 171	99.16 70

9.14			96.75 33		98.93 47		97.73 75	98.23 38	91.28 146	97.88 22	98.44 28	93.00 21	99.65 53	95.76 56	99.47 36

XVS		97.18 16	96.96 18	97.81 30	99.38 14	94.03 50	98.59 8	98.20 42	94.85 26	96.59 60	98.29 50	91.70 50	99.80 27	95.66 57	99.40 45	99.62 13

X-MVStestdata		91.71 199	89.67 258	97.81 30	99.38 14	94.03 50	98.59 8	98.20 42	94.85 26	96.59 60	32.69 364	91.70 50	99.80 27	95.66 57	99.40 45	99.62 13

baseline		95.58 77	95.42 75	96.08 114	96.78 165	90.41 164	97.16 138	97.45 160	93.69 63	95.65 101	97.85 78	87.29 117	98.68 172	95.66 57	97.25 135	99.13 74

ETV-MVS		96.02 65	95.89 64	96.40 96	97.16 142	92.44 92	97.47 106	97.77 114	94.55 39	96.48 65	94.51 255	91.23 62	98.92 151	95.65 60	98.19 107	97.82 168

casdiffmvs		95.64 74	95.49 71	96.08 114	96.76 168	90.45 162	97.29 124	97.44 164	94.00 50	95.46 107	97.98 70	87.52 113	98.73 167	95.64 61	97.33 132	99.08 80

HFP-MVS		97.14 19	96.92 20	97.83 26	99.42 6	94.12 45	98.52 11	98.32 20	93.21 77	97.18 38	98.29 50	92.08 39	99.83 22	95.63 62	99.59 15	99.54 29

ACMMPR		97.07 22	96.84 25	97.79 32	99.44 5	93.88 52	98.52 11	98.31 22	93.21 77	97.15 40	98.33 44	91.35 59	99.86 8	95.63 62	99.59 15	99.62 13

HPM-MVS		96.69 45	96.45 49	97.40 53	99.36 19	93.11 74	98.87 1	98.06 73	91.17 150	96.40 69	97.99 69	90.99 67	99.58 71	95.61 64	99.61 14	99.49 39
Chunlin Ren, Qingshan Xu, Shikun Zhang, Jiaqi Yang: Hierarchical Prior Mining for Non-local Multi-View Stereo. ICCV 2023
CP-MVS		97.02 26	96.81 28	97.64 46	99.33 22	93.54 62	98.80 3	98.28 26	92.99 87	96.45 68	98.30 49	91.90 45	99.85 14	95.61 64	99.68 4	99.54 29

DeepC-MVS	93.07 3	96.06 63	95.66 67	97.29 58	97.96 109	93.17 73	97.30 123	98.06 73	93.92 52	93.38 143	98.66 12	86.83 122	99.73 32	95.60 66	99.22 68	98.96 92
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
ZNCC-MVS		96.96 30	96.67 38	97.85 25	99.37 16	94.12 45	98.49 15	98.18 46	92.64 105	96.39 70	98.18 58	91.61 52	99.88 4	95.59 67	99.55 21	99.57 19

ETH3D-3000-0.1		97.07 22	96.71 36	98.14 13	98.90 51	95.33 14	97.68 82	98.24 34	91.57 132	97.90 21	98.37 36	92.61 29	99.66 52	95.59 67	99.51 29	99.43 49

region2R		97.07 22	96.84 25	97.77 35	99.46 1	93.79 55	98.52 11	98.24 34	93.19 80	97.14 41	98.34 41	91.59 54	99.87 7	95.46 69	99.59 15	99.64 10

OPU-MVS				98.55 1	98.82 56	96.86 1	98.25 29				98.26 53	96.04 1	99.24 121	95.36 70	99.59 15	99.56 22

lupinMVS		94.99 95	94.56 95	96.29 106	96.34 190	91.21 133	95.83 244	96.27 254	88.93 210	96.22 75	96.88 137	86.20 132	98.85 157	95.27 71	99.05 83	98.82 107

mPP-MVS		96.86 37	96.60 40	97.64 46	99.40 11	93.44 65	98.50 14	98.09 63	93.27 76	95.95 88	98.33 44	91.04 66	99.88 4	95.20 72	99.57 20	99.60 16

DeepC-MVS_fast	93.89 2	96.93 33	96.64 39	97.78 33	98.64 67	94.30 34	97.41 109	98.04 81	94.81 31	96.59 60	98.37 36	91.24 61	99.64 61	95.16 73	99.52 25	99.42 52
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
jason		94.84 100	94.39 103	96.18 111	95.52 224	90.93 147	96.09 230	96.52 244	89.28 198	96.01 86	97.32 115	84.70 149	98.77 164	95.15 74	98.91 91	98.85 104
jason: jason.
#test#		97.02 26	96.75 33	97.83 26	99.42 6	94.12 45	98.15 39	98.32 20	92.57 106	97.18 38	98.29 50	92.08 39	99.83 22	95.12 75	99.59 15	99.54 29

abl_6		96.40 54	96.21 56	96.98 74	98.89 54	92.20 102	97.89 58	98.03 84	93.34 75	97.22 37	98.42 31	87.93 105	99.72 35	95.10 76	99.07 81	99.02 83

train_agg		96.30 57	95.83 65	97.72 39	98.70 60	94.19 40	96.41 203	98.02 88	88.58 222	96.03 82	97.56 105	92.73 25	99.59 68	95.04 77	99.37 52	99.39 54

agg_prior1		96.22 61	95.77 66	97.56 48	98.67 62	93.79 55	96.28 219	98.00 93	88.76 219	95.68 97	97.55 107	92.70 27	99.57 79	95.01 78	99.32 53	99.32 60

test_prior3		96.46 53	96.20 57	97.23 62	98.67 62	92.99 76	96.35 211	98.00 93	92.80 98	96.03 82	97.59 101	92.01 41	99.41 107	95.01 78	99.38 48	99.29 62

test_prior2							96.35 211		92.80 98	96.03 82	97.59 101	92.01 41		95.01 78	99.38 48

nrg030		94.05 119	93.31 129	96.27 107	95.22 246	94.59 28	98.34 21	97.46 154	92.93 94	91.21 193	96.64 150	87.23 119	98.22 208	94.99 81	85.80 284	95.98 224

VDDNet		93.05 152	92.07 164	96.02 119	96.84 161	90.39 165	98.08 43	95.85 269	86.22 278	95.79 93	98.46 26	67.59 331	99.19 124	94.92 82	94.85 177	98.47 131

APD-MVS		96.95 31	96.60 40	98.01 19	99.03 42	94.93 24	97.72 78	98.10 61	91.50 134	98.01 18	98.32 46	92.33 35	99.58 71	94.85 83	99.51 29	99.53 33
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
ETH3D cwj APD-0.16		96.56 50	96.06 59	98.05 17	98.26 92	95.19 18	96.99 152	98.05 80	89.85 184	97.26 35	98.22 56	91.80 47	99.69 44	94.84 84	99.28 59	99.27 66

GST-MVS		96.85 38	96.52 45	97.82 29	99.36 19	94.14 44	98.29 25	98.13 54	92.72 101	96.70 52	98.06 64	91.35 59	99.86 8	94.83 85	99.28 59	99.47 44

MP-MVS		96.77 42	96.45 49	97.72 39	99.39 13	93.80 54	98.41 19	98.06 73	93.37 72	95.54 105	98.34 41	90.59 75	99.88 4	94.83 85	99.54 23	99.49 39
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
test9_res														94.81 87	99.38 48	99.45 45

PS-MVSNAJ		95.37 81	95.33 78	95.49 151	97.35 136	90.66 156	95.31 265	97.48 149	93.85 55	96.51 63	95.70 207	88.65 96	99.65 53	94.80 88	98.27 105	96.17 215

HPM-MVS_fast		96.51 51	96.27 54	97.22 64	99.32 23	92.74 82	98.74 4	98.06 73	90.57 170	96.77 49	98.35 38	90.21 79	99.53 89	94.80 88	99.63 12	99.38 56

xiu_mvs_v2_base		95.32 83	95.29 79	95.40 156	97.22 138	90.50 160	95.44 259	97.44 164	93.70 62	96.46 67	96.18 176	88.59 99	99.53 89	94.79 90	97.81 117	96.17 215

CSCG		96.05 64	95.91 63	96.46 93	99.24 28	90.47 161	98.30 24	98.57 11	89.01 205	93.97 130	97.57 103	92.62 28	99.76 30	94.66 91	99.27 61	99.15 72

EIA-MVS		95.53 79	95.47 72	95.71 135	97.06 150	89.63 181	97.82 65	97.87 107	93.57 64	93.92 131	95.04 232	90.61 74	98.95 149	94.62 92	98.68 96	98.54 121

ZD-MVS					99.05 41	94.59 28		98.08 64	89.22 200	97.03 47	98.10 60	92.52 32	99.65 53	94.58 93	99.31 55

ACMMP		96.27 58	95.93 62	97.28 59	99.24 28	92.62 87	98.25 29	98.81 3	92.99 87	94.56 118	98.39 35	88.96 91	99.85 14	94.57 94	97.63 121	99.36 58
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
PGM-MVS		96.81 40	96.53 44	97.65 44	99.35 21	93.53 63	97.65 86	98.98 1	92.22 113	97.14 41	98.44 28	91.17 64	99.85 14	94.35 95	99.46 38	99.57 19

ET-MVSNet_ETH3D		91.49 211	90.11 240	95.63 138	96.40 187	91.57 121	95.34 262	93.48 335	90.60 169	75.58 349	95.49 218	80.08 232	96.79 322	94.25 96	89.76 249	98.52 123

testtj		96.93 33	96.56 43	98.05 17	99.10 34	94.66 27	97.78 69	98.22 39	92.74 100	97.59 24	98.20 57	91.96 44	99.86 8	94.21 97	99.25 65	99.63 11

bset_n11_16_dypcd		91.55 207	90.59 219	94.44 195	91.51 336	90.25 166	92.70 327	93.42 336	92.27 112	90.22 207	94.74 246	78.42 262	97.80 266	94.19 98	87.86 265	95.29 267

LFMVS		93.60 134	92.63 147	96.52 85	98.13 104	91.27 130	97.94 55	93.39 337	90.57 170	96.29 72	98.31 47	69.00 324	99.16 128	94.18 99	95.87 160	99.12 77

MVSFormer		95.37 81	95.16 82	95.99 121	96.34 190	91.21 133	98.22 34	97.57 141	91.42 138	96.22 75	97.32 115	86.20 132	97.92 255	94.07 100	99.05 83	98.85 104

test_djsdf		93.07 151	92.76 141	94.00 212	93.49 311	88.70 216	98.22 34	97.57 141	91.42 138	90.08 218	95.55 215	82.85 183	97.92 255	94.07 100	91.58 223	95.40 254

mvs_anonymous		93.82 127	93.74 111	94.06 209	96.44 184	85.41 284	95.81 245	97.05 201	89.85 184	90.09 217	96.36 170	87.44 115	97.75 272	93.97 102	96.69 147	99.02 83

VPA-MVSNet		93.24 145	92.48 156	95.51 148	95.70 218	92.39 93	97.86 60	98.66 9	92.30 111	92.09 173	95.37 221	80.49 224	98.40 196	93.95 103	85.86 283	95.75 237

agg_prior2														93.94 104	99.38 48	99.50 37

mvs_tets		92.31 179	91.76 174	93.94 219	93.41 313	88.29 225	97.63 92	97.53 145	92.04 122	88.76 256	96.45 164	74.62 294	98.09 227	93.91 105	91.48 225	95.45 250

Effi-MVS+		94.93 96	94.45 101	96.36 101	96.61 170	91.47 123	96.41 203	97.41 169	91.02 155	94.50 119	95.92 189	87.53 112	98.78 162	93.89 106	96.81 142	98.84 106

jajsoiax		92.42 174	91.89 172	94.03 211	93.33 316	88.50 222	97.73 75	97.53 145	92.00 124	88.85 252	96.50 162	75.62 290	98.11 222	93.88 107	91.56 224	95.48 245

XVG-OURS-SEG-HR		93.86 126	93.55 117	94.81 179	97.06 150	88.53 221	95.28 266	97.45 160	91.68 130	94.08 127	97.68 91	82.41 194	98.90 154	93.84 108	92.47 208	96.98 193

PS-MVSNAJss		93.74 130	93.51 121	94.44 195	93.91 298	89.28 202	97.75 72	97.56 144	92.50 107	89.94 220	96.54 160	88.65 96	98.18 214	93.83 109	90.90 236	95.86 226

EPNet		95.20 88	94.56 95	97.14 68	92.80 324	92.68 84	97.85 63	94.87 314	96.64 1	92.46 159	97.80 84	86.23 129	99.65 53	93.72 110	98.62 98	99.10 79
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
RRT_test8_iter05		91.19 230	90.78 211	92.41 275	95.76 217	83.14 311	97.32 120	97.46 154	91.37 142	89.07 248	95.57 212	70.33 317	98.21 209	93.56 111	86.62 278	95.89 225

PVSNet_Blended_VisFu		95.27 84	94.91 87	96.38 99	98.20 98	90.86 149	97.27 125	98.25 33	90.21 175	94.18 125	97.27 117	87.48 114	99.73 32	93.53 112	97.77 119	98.55 120

CPTT-MVS		95.57 78	95.19 81	96.70 77	99.27 26	91.48 122	98.33 22	98.11 59	87.79 247	95.17 111	98.03 66	87.09 120	99.61 62	93.51 113	99.42 43	99.02 83

MVSTER		93.20 147	92.81 140	94.37 199	96.56 176	89.59 184	97.06 143	97.12 192	91.24 147	91.30 187	95.96 187	82.02 201	98.05 234	93.48 114	90.55 240	95.47 247

PVSNet_BlendedMVS		94.06 118	93.92 107	94.47 194	98.27 89	89.46 192	96.73 175	98.36 16	90.17 176	94.36 121	95.24 226	88.02 102	99.58 71	93.44 115	90.72 238	94.36 308

PVSNet_Blended		94.87 99	94.56 95	95.81 127	98.27 89	89.46 192	95.47 258	98.36 16	88.84 213	94.36 121	96.09 184	88.02 102	99.58 71	93.44 115	98.18 108	98.40 140

3Dnovator	91.36 5	95.19 89	94.44 102	97.44 52	96.56 176	93.36 69	98.65 7	98.36 16	94.12 48	89.25 245	98.06 64	82.20 198	99.77 29	93.41 117	99.32 53	99.18 69

EPP-MVSNet		95.22 87	95.04 85	95.76 128	97.49 135	89.56 185	98.67 5	97.00 207	90.69 160	94.24 124	97.62 99	89.79 86	98.81 160	93.39 118	96.49 152	98.92 97

RRT_MVS		93.21 146	92.32 160	95.91 123	94.92 261	94.15 43	96.92 159	96.86 222	91.42 138	91.28 190	96.43 165	79.66 241	98.10 223	93.29 119	90.06 245	95.46 248

CHOSEN 280x420		93.12 149	92.72 145	94.34 201	96.71 169	87.27 248	90.29 343	97.72 123	86.61 273	91.34 184	95.29 223	84.29 157	98.41 195	93.25 120	98.94 88	97.35 187

3Dnovator+	91.43 4	95.40 80	94.48 100	98.16 12	96.90 158	95.34 13	98.48 16	97.87 107	94.65 38	88.53 261	98.02 67	83.69 163	99.71 38	93.18 121	98.96 87	99.44 47

test_yl		94.78 102	94.23 104	96.43 94	97.74 123	91.22 131	96.85 164	97.10 194	91.23 148	95.71 95	96.93 132	84.30 155	99.31 116	93.10 122	95.12 173	98.75 109

DCV-MVSNet		94.78 102	94.23 104	96.43 94	97.74 123	91.22 131	96.85 164	97.10 194	91.23 148	95.71 95	96.93 132	84.30 155	99.31 116	93.10 122	95.12 173	98.75 109

HQP_MVS		93.78 129	93.43 125	94.82 177	96.21 195	89.99 172	97.74 73	97.51 147	94.85 26	91.34 184	96.64 150	81.32 212	98.60 180	93.02 124	92.23 211	95.86 226

plane_prior5								97.51 147					98.60 180	93.02 124	92.23 211	95.86 226

MVS_Test		94.89 98	94.62 93	95.68 136	96.83 163	89.55 186	96.70 179	97.17 188	91.17 150	95.60 102	96.11 183	87.87 106	98.76 165	93.01 126	97.17 138	98.72 113

CLD-MVS		92.98 155	92.53 153	94.32 202	96.12 204	89.20 204	95.28 266	97.47 152	92.66 103	89.90 221	95.62 210	80.58 222	98.40 196	92.73 127	92.40 209	95.38 256
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
XVG-OURS		93.72 131	93.35 128	94.80 182	97.07 147	88.61 217	94.79 276	97.46 154	91.97 125	93.99 128	97.86 77	81.74 207	98.88 156	92.64 128	92.67 206	96.92 197

ETH3 D test6400		96.16 62	95.52 70	98.07 16	98.90 51	95.06 22	97.03 144	98.21 40	88.16 236	96.64 57	97.70 89	91.18 63	99.67 49	92.44 129	99.47 36	99.48 41

旧先验2							95.94 239		81.66 327	97.34 34			98.82 159	92.26 130

CDPH-MVS		95.97 67	95.38 76	97.77 35	98.93 47	94.44 31	96.35 211	97.88 105	86.98 266	96.65 56	97.89 72	91.99 43	99.47 100	92.26 130	99.46 38	99.39 54

FIs		94.09 117	93.70 112	95.27 158	95.70 218	92.03 107	98.10 41	98.68 7	93.36 74	90.39 203	96.70 145	87.63 110	97.94 252	92.25 132	90.50 242	95.84 229

LPG-MVS_test		92.94 158	92.56 150	94.10 207	96.16 200	88.26 227	97.65 86	97.46 154	91.29 143	90.12 214	97.16 123	79.05 249	98.73 167	92.25 132	91.89 219	95.31 260

LGP-MVS_train				94.10 207	96.16 200	88.26 227		97.46 154	91.29 143	90.12 214	97.16 123	79.05 249	98.73 167	92.25 132	91.89 219	95.31 260

cascas		91.20 227	90.08 241	94.58 192	94.97 257	89.16 207	93.65 311	97.59 139	79.90 338	89.40 237	92.92 308	75.36 291	98.36 200	92.14 135	94.75 181	96.23 212

OPM-MVS		93.28 144	92.76 141	94.82 177	94.63 277	90.77 153	96.65 185	97.18 186	93.72 60	91.68 178	97.26 118	79.33 246	98.63 176	92.13 136	92.28 210	95.07 272

BP-MVS														92.13 136

HQP-MVS		93.19 148	92.74 144	94.54 193	95.86 210	89.33 198	96.65 185	97.39 170	93.55 65	90.14 208	95.87 192	80.95 215	98.50 189	92.13 136	92.10 216	95.78 233

DP-MVS Recon		95.68 73	95.12 84	97.37 54	99.19 31	94.19 40	97.03 144	98.08 64	88.35 229	95.09 112	97.65 94	89.97 83	99.48 99	92.08 139	98.59 99	98.44 137

VPNet		92.23 185	91.31 191	94.99 168	95.56 222	90.96 145	97.22 133	97.86 110	92.96 93	90.96 195	96.62 157	75.06 292	98.20 211	91.90 140	83.65 317	95.80 232

sss		94.51 106	93.80 110	96.64 78	97.07 147	91.97 110	96.32 215	98.06 73	88.94 209	94.50 119	96.78 139	84.60 150	99.27 119	91.90 140	96.02 156	98.68 117

anonymousdsp		92.16 188	91.55 182	93.97 215	92.58 328	89.55 186	97.51 100	97.42 168	89.42 195	88.40 262	94.84 240	80.66 221	97.88 260	91.87 142	91.28 229	94.48 304

ACMP	89.59 10	92.62 169	92.14 163	94.05 210	96.40 187	88.20 230	97.36 116	97.25 184	91.52 133	88.30 265	96.64 150	78.46 261	98.72 170	91.86 143	91.48 225	95.23 268
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
HyFIR lowres test		93.66 132	92.92 138	95.87 125	98.24 93	89.88 177	94.58 280	98.49 12	85.06 295	93.78 133	95.78 201	82.86 182	98.67 173	91.77 144	95.71 165	99.07 82

UGNet		94.04 120	93.28 130	96.31 103	96.85 160	91.19 136	97.88 59	97.68 128	94.40 42	93.00 151	96.18 176	73.39 304	99.61 62	91.72 145	98.46 101	98.13 151
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
UniMVSNet_NR-MVSNet		93.37 141	92.67 146	95.47 154	95.34 235	92.83 80	97.17 137	98.58 10	92.98 92	90.13 212	95.80 197	88.37 101	97.85 261	91.71 146	83.93 312	95.73 239

DU-MVS		92.90 160	92.04 165	95.49 151	94.95 259	92.83 80	97.16 138	98.24 34	93.02 85	90.13 212	95.71 205	83.47 167	97.85 261	91.71 146	83.93 312	95.78 233

Effi-MVS+-dtu		93.08 150	93.21 132	92.68 270	96.02 207	83.25 310	97.14 141	96.72 228	93.85 55	91.20 194	93.44 302	83.08 175	98.30 204	91.69 148	95.73 164	96.50 208

mvs-test1		93.63 133	93.69 113	93.46 242	96.02 207	84.61 296	97.24 127	96.72 228	93.85 55	92.30 166	95.76 202	83.08 175	98.89 155	91.69 148	96.54 150	96.87 199

UniMVSNet (Re)		93.31 143	92.55 151	95.61 140	95.39 229	93.34 70	97.39 113	98.71 5	93.14 82	90.10 216	94.83 241	87.71 107	98.03 238	91.67 150	83.99 311	95.46 248

LCM-MVSNet-Re		92.50 170	92.52 154	92.44 273	96.82 164	81.89 319	96.92 159	93.71 333	92.41 109	84.30 317	94.60 253	85.08 145	97.03 313	91.51 151	97.36 130	98.40 140

FC-MVSNet-test		93.94 123	93.57 116	95.04 166	95.48 226	91.45 125	98.12 40	98.71 5	93.37 72	90.23 206	96.70 145	87.66 108	97.85 261	91.49 152	90.39 243	95.83 230

PMMVS		92.86 162	92.34 158	94.42 198	94.92 261	86.73 262	94.53 282	96.38 250	84.78 300	94.27 123	95.12 231	83.13 174	98.40 196	91.47 153	96.49 152	98.12 152

Vis-MVSNet		95.23 86	94.81 88	96.51 88	97.18 141	91.58 120	98.26 28	98.12 56	94.38 44	94.90 113	98.15 59	82.28 196	98.92 151	91.45 154	98.58 100	99.01 87
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
CHOSEN 1792x2688		94.15 112	93.51 121	96.06 116	98.27 89	89.38 195	95.18 272	98.48 14	85.60 286	93.76 134	97.11 126	83.15 173	99.61 62	91.33 155	98.72 95	99.19 68

OMC-MVS		95.09 90	94.70 92	96.25 109	98.46 74	91.28 129	96.43 201	97.57 141	92.04 122	94.77 116	97.96 71	87.01 121	99.09 137	91.31 156	96.77 143	98.36 144

MG-MVS		95.61 75	95.38 76	96.31 103	98.42 77	90.53 159	96.04 232	97.48 149	93.47 71	95.67 100	98.10 60	89.17 89	99.25 120	91.27 157	98.77 93	99.13 74

ACMM	89.79 8	92.96 156	92.50 155	94.35 200	96.30 193	88.71 215	97.58 95	97.36 175	91.40 141	90.53 199	96.65 149	79.77 238	98.75 166	91.24 158	91.64 221	95.59 243
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
WTY-MVS		94.71 104	94.02 106	96.79 76	97.71 125	92.05 106	96.59 194	97.35 176	90.61 167	94.64 117	96.93 132	86.41 128	99.39 110	91.20 159	94.71 183	98.94 95

Anonymous20240529		91.98 193	90.73 214	95.73 133	98.14 103	89.40 194	97.99 48	97.72 123	79.63 339	93.54 138	97.41 113	69.94 322	99.56 81	91.04 160	91.11 231	98.22 148

test_part1		92.21 187	91.10 201	95.51 148	97.80 120	92.66 85	98.02 47	97.68 128	89.79 187	88.80 255	96.02 185	76.85 279	98.18 214	90.86 161	84.11 310	95.69 240

AUN-MVS		91.76 198	90.75 213	94.81 179	97.00 155	88.57 219	96.65 185	96.49 245	89.63 189	92.15 169	96.12 180	78.66 258	98.50 189	90.83 162	79.18 335	97.36 186

CANet_DTU		94.37 107	93.65 115	96.55 84	96.46 183	92.13 104	96.21 225	96.67 236	94.38 44	93.53 139	97.03 130	79.34 245	99.71 38	90.76 163	98.45 102	97.82 168

ab-mvs		93.57 136	92.55 151	96.64 78	97.28 137	91.96 111	95.40 260	97.45 160	89.81 186	93.22 149	96.28 173	79.62 242	99.46 101	90.74 164	93.11 200	98.50 126

CostFormer		91.18 231	90.70 215	92.62 271	94.84 267	81.76 320	94.09 299	94.43 321	84.15 306	92.72 158	93.77 291	79.43 244	98.20 211	90.70 165	92.18 214	97.90 161

Anonymous202405211		92.07 191	90.83 210	95.76 128	98.19 100	88.75 214	97.58 95	95.00 305	86.00 281	93.64 135	97.45 109	66.24 340	99.53 89	90.68 166	92.71 204	99.01 87

tpmrst		91.44 213	91.32 190	91.79 290	95.15 249	79.20 342	93.42 315	95.37 287	88.55 225	93.49 140	93.67 296	82.49 192	98.27 205	90.41 167	89.34 252	97.90 161

thisisatest0530		93.03 153	92.21 162	95.49 151	97.07 147	89.11 208	97.49 105	92.19 345	90.16 177	94.09 126	96.41 167	76.43 284	99.05 143	90.38 168	95.68 166	98.31 146

UA-Net		95.95 68	95.53 69	97.20 66	97.67 126	92.98 78	97.65 86	98.13 54	94.81 31	96.61 58	98.35 38	88.87 92	99.51 94	90.36 169	97.35 131	99.11 78

UniMVSNet_ETH3D		91.34 221	90.22 237	94.68 187	94.86 266	87.86 240	97.23 132	97.46 154	87.99 239	89.90 221	96.92 135	66.35 338	98.23 207	90.30 170	90.99 234	97.96 157

tttt0517		92.96 156	92.33 159	94.87 176	97.11 145	87.16 254	97.97 53	92.09 346	90.63 165	93.88 132	97.01 131	76.50 281	99.06 142	90.29 171	95.45 168	98.38 142

IS-MVSNet		94.90 97	94.52 98	96.05 117	97.67 126	90.56 157	98.44 17	96.22 257	93.21 77	93.99 128	97.74 87	85.55 140	98.45 194	89.98 172	97.86 115	99.14 73

miper_enhance_ethall		91.54 209	91.01 202	93.15 254	95.35 234	87.07 256	93.97 301	96.90 216	86.79 270	89.17 246	93.43 304	86.55 125	97.64 280	89.97 173	86.93 273	94.74 298

EI-MVSNet		93.03 153	92.88 139	93.48 240	95.77 215	86.98 257	96.44 199	97.12 192	90.66 163	91.30 187	97.64 97	86.56 124	98.05 234	89.91 174	90.55 240	95.41 251

IterMVS-LS		92.29 181	91.94 170	93.34 247	96.25 194	86.97 258	96.57 197	97.05 201	90.67 161	89.50 236	94.80 243	86.59 123	97.64 280	89.91 174	86.11 282	95.40 254
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
cl-mvsnet2		91.21 226	90.56 222	93.14 255	96.09 206	86.80 260	94.41 287	96.58 243	87.80 246	88.58 260	93.99 284	80.85 220	97.62 283	89.87 176	86.93 273	94.99 275

CDS-MVSNet		94.14 115	93.54 118	95.93 122	96.18 198	91.46 124	96.33 214	97.04 203	88.97 208	93.56 136	96.51 161	87.55 111	97.89 259	89.80 177	95.95 158	98.44 137
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
WR-MVS		92.34 177	91.53 183	94.77 184	95.13 251	90.83 150	96.40 206	97.98 98	91.88 126	89.29 242	95.54 216	82.50 191	97.80 266	89.79 178	85.27 292	95.69 240

NR-MVSNet		92.34 177	91.27 194	95.53 147	94.95 259	93.05 75	97.39 113	98.07 70	92.65 104	84.46 315	95.71 205	85.00 146	97.77 271	89.71 179	83.52 318	95.78 233

Anonymous20231211		90.63 251	89.42 262	94.27 203	98.24 93	89.19 206	98.05 45	97.89 103	79.95 337	88.25 268	94.96 233	72.56 305	98.13 218	89.70 180	85.14 294	95.49 244

testdata				95.46 155	98.18 102	88.90 212		97.66 131	82.73 321	97.03 47	98.07 63	90.06 81	98.85 157	89.67 181	98.98 86	98.64 118

Baseline_NR-MVSNet		91.20 227	90.62 217	92.95 261	93.83 301	88.03 235	97.01 151	95.12 301	88.42 227	89.70 227	95.13 230	83.47 167	97.44 299	89.66 182	83.24 320	93.37 326

DPM-MVS		95.69 72	94.92 86	98.01 19	98.08 106	95.71 7	95.27 268	97.62 136	90.43 173	95.55 103	97.07 128	91.72 48	99.50 97	89.62 183	98.94 88	98.82 107

XXY-MVS		92.16 188	91.23 196	94.95 173	94.75 271	90.94 146	97.47 106	97.43 167	89.14 202	88.90 249	96.43 165	79.71 239	98.24 206	89.56 184	87.68 266	95.67 242

miper_ehance_all_eth		91.59 203	91.13 200	92.97 260	95.55 223	86.57 267	94.47 283	96.88 219	87.77 248	88.88 251	94.01 282	86.22 130	97.54 289	89.49 185	86.93 273	94.79 294

XVG-ACMP-BASELINE		90.93 240	90.21 238	93.09 256	94.31 289	85.89 277	95.33 263	97.26 182	91.06 154	89.38 238	95.44 220	68.61 326	98.60 180	89.46 186	91.05 232	94.79 294

thisisatest0515		92.29 181	91.30 192	95.25 159	96.60 171	88.90 212	94.36 289	92.32 344	87.92 241	93.43 142	94.57 254	77.28 277	99.00 146	89.42 187	95.86 161	97.86 164

cl_fuxian		91.38 216	90.89 204	92.88 263	95.58 221	86.30 270	94.68 278	96.84 224	88.17 234	88.83 254	94.23 274	85.65 139	97.47 296	89.36 188	84.63 302	94.89 284

AdaColmap		94.34 108	93.68 114	96.31 103	98.59 69	91.68 116	96.59 194	97.81 113	89.87 181	92.15 169	97.06 129	83.62 166	99.54 86	89.34 189	98.07 111	97.70 172

TranMVSNet+NR-MVSNet		92.50 170	91.63 179	95.14 163	94.76 270	92.07 105	97.53 99	98.11 59	92.90 95	89.56 233	96.12 180	83.16 172	97.60 285	89.30 190	83.20 321	95.75 237

D2MVS		91.30 223	90.95 203	92.35 276	94.71 273	85.52 282	96.18 227	98.21 40	88.89 211	86.60 298	93.82 289	79.92 236	97.95 251	89.29 191	90.95 235	93.56 322

1314		92.81 166	92.03 166	95.14 163	95.33 238	89.52 189	96.04 232	97.44 164	87.72 251	86.25 301	95.33 222	83.84 161	98.79 161	89.26 192	97.05 140	97.11 191

v2v482		91.59 203	90.85 208	93.80 225	93.87 300	88.17 232	96.94 158	96.88 219	89.54 190	89.53 234	94.90 237	81.70 208	98.02 239	89.25 193	85.04 298	95.20 269

114514_t		93.95 122	93.06 134	96.63 80	99.07 39	91.61 117	97.46 108	97.96 100	77.99 345	93.00 151	97.57 103	86.14 134	99.33 114	89.22 194	99.15 73	98.94 95

PAPM_NR		95.01 91	94.59 94	96.26 108	98.89 54	90.68 155	97.24 127	97.73 119	91.80 127	92.93 156	96.62 157	89.13 90	99.14 131	89.21 195	97.78 118	98.97 91

baseline1		92.82 165	91.90 171	95.55 146	97.20 140	90.77 153	97.19 135	94.58 319	92.20 115	92.36 163	96.34 171	84.16 158	98.21 209	89.20 196	83.90 315	97.68 173

IB-MVS	87.33 17	89.91 266	88.28 278	94.79 183	95.26 245	87.70 243	95.12 274	93.95 332	89.35 197	87.03 292	92.49 314	70.74 315	99.19 124	89.18 197	81.37 328	97.49 184
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
HY-MVS	89.66 9	93.87 125	92.95 137	96.63 80	97.10 146	92.49 91	95.64 252	96.64 237	89.05 204	93.00 151	95.79 200	85.77 138	99.45 103	89.16 198	94.35 185	97.96 157

V42		91.58 205	90.87 205	93.73 227	94.05 295	88.50 222	97.32 120	96.97 208	88.80 218	89.71 226	94.33 266	82.54 190	98.05 234	89.01 199	85.07 296	94.64 302

OurMVSNet-221017-0		90.51 254	90.19 239	91.44 299	93.41 313	81.25 323	96.98 154	96.28 253	91.68 130	86.55 299	96.30 172	74.20 297	97.98 242	88.96 200	87.40 271	95.09 271

API-MVS		94.84 100	94.49 99	95.90 124	97.90 115	92.00 109	97.80 67	97.48 149	89.19 201	94.81 115	96.71 143	88.84 93	99.17 127	88.91 201	98.76 94	96.53 206

test-LLR		91.42 214	91.19 198	92.12 280	94.59 278	80.66 326	94.29 293	92.98 339	91.11 152	90.76 197	92.37 316	79.02 251	98.07 231	88.81 202	96.74 144	97.63 174

test-mter		90.19 262	89.54 261	92.12 280	94.59 278	80.66 326	94.29 293	92.98 339	87.68 252	90.76 197	92.37 316	67.67 330	98.07 231	88.81 202	96.74 144	97.63 174

eth_miper_zixun_eth		91.02 235	90.59 219	92.34 277	95.33 238	84.35 297	94.10 298	96.90 216	88.56 224	88.84 253	94.33 266	84.08 159	97.60 285	88.77 204	84.37 307	95.06 273

TAMVS		94.01 121	93.46 123	95.64 137	96.16 200	90.45 162	96.71 178	96.89 218	89.27 199	93.46 141	96.92 135	87.29 117	97.94 252	88.70 205	95.74 163	98.53 122

Patchmatch-RL test		87.38 294	86.24 295	90.81 309	88.74 352	78.40 346	88.12 352	93.17 338	87.11 265	82.17 332	89.29 341	81.95 203	95.60 338	88.64 206	77.02 339	98.41 139

baseline2		91.63 202	90.86 206	93.94 219	94.33 287	86.32 269	95.92 240	91.64 350	89.37 196	86.94 294	94.69 248	81.62 209	98.69 171	88.64 206	94.57 184	96.81 201

TESTMET0.1,1		90.06 264	89.42 262	91.97 283	94.41 285	80.62 328	94.29 293	91.97 348	87.28 262	90.44 202	92.47 315	68.79 325	97.67 277	88.50 208	96.60 149	97.61 178

Vis-MVSNet (Re-imp)		94.15 112	93.88 108	94.95 173	97.61 130	87.92 237	98.10 41	95.80 271	92.22 113	93.02 150	97.45 109	84.53 152	97.91 258	88.24 209	97.97 113	99.02 83

DWT-MVSNet_test		90.76 244	89.89 248	93.38 245	95.04 255	83.70 306	95.85 243	94.30 327	88.19 232	90.46 201	92.80 309	73.61 302	98.50 189	88.16 210	90.58 239	97.95 159

1112_ss		93.37 141	92.42 157	96.21 110	97.05 152	90.99 143	96.31 216	96.72 228	86.87 269	89.83 224	96.69 147	86.51 126	99.14 131	88.12 211	93.67 194	98.50 126

CVMVSNet		91.23 225	91.75 175	89.67 323	95.77 215	74.69 351	96.44 199	94.88 311	85.81 283	92.18 168	97.64 97	79.07 248	95.58 339	88.06 212	95.86 161	98.74 111

MAR-MVS		94.22 110	93.46 123	96.51 88	98.00 108	92.19 103	97.67 83	97.47 152	88.13 238	93.00 151	95.84 194	84.86 148	99.51 94	87.99 213	98.17 109	97.83 167
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
原ACMM1				96.38 99	98.59 69	91.09 141		97.89 103	87.41 258	95.22 109	97.68 91	90.25 77	99.54 86	87.95 214	99.12 78	98.49 128

CP-MVSNet		91.89 195	91.24 195	93.82 224	95.05 254	88.57 219	97.82 65	98.19 44	91.70 129	88.21 269	95.76 202	81.96 202	97.52 293	87.86 215	84.65 301	95.37 257

v148		90.99 236	90.38 226	92.81 266	93.83 301	85.80 278	96.78 173	96.68 234	89.45 194	88.75 257	93.93 286	82.96 181	97.82 265	87.83 216	83.25 319	94.80 292

v1144		91.37 218	90.60 218	93.68 232	93.89 299	88.23 229	96.84 166	97.03 205	88.37 228	89.69 228	94.39 262	82.04 200	97.98 242	87.80 217	85.37 289	94.84 286

cl-mvsnet1		90.97 238	90.33 227	92.88 263	95.36 233	86.19 274	94.46 285	96.63 240	87.82 244	88.18 270	94.23 274	82.99 178	97.53 291	87.72 218	85.57 286	94.93 280

gm-plane-assit					93.22 317	78.89 345			84.82 299		93.52 299		98.64 175	87.72 218

GeoE		93.89 124	93.28 130	95.72 134	96.96 157	89.75 180	98.24 32	96.92 215	89.47 193	92.12 171	97.21 121	84.42 153	98.39 199	87.71 220	96.50 151	99.01 87

cl-mvsnet____		90.96 239	90.32 228	92.89 262	95.37 232	86.21 273	94.46 285	96.64 237	87.82 244	88.15 271	94.18 277	82.98 179	97.54 289	87.70 221	85.59 285	94.92 282

pmmvs4		90.93 240	89.85 250	94.17 205	93.34 315	90.79 152	94.60 279	96.02 263	84.62 301	87.45 282	95.15 228	81.88 205	97.45 298	87.70 221	87.87 264	94.27 313

Test_1112_low_res		92.84 164	91.84 173	95.85 126	97.04 153	89.97 175	95.53 256	96.64 237	85.38 289	89.65 230	95.18 227	85.86 136	99.10 134	87.70 221	93.58 199	98.49 128

无先验							95.79 246	97.87 107	83.87 311				99.65 53	87.68 224		98.89 101

1121		94.71 104	93.83 109	97.34 55	98.57 72	93.64 60	96.04 232	97.73 119	81.56 329	95.68 97	97.85 78	90.23 78	99.65 53	87.68 224	99.12 78	98.73 112

Fast-Effi-MVS+		93.46 138	92.75 143	95.59 141	96.77 166	90.03 169	96.81 170	97.13 191	88.19 232	91.30 187	94.27 271	86.21 131	98.63 176	87.66 226	96.46 154	98.12 152

CNLPA		94.28 109	93.53 119	96.52 85	98.38 81	92.55 89	96.59 194	96.88 219	90.13 178	91.91 175	97.24 119	85.21 143	99.09 137	87.64 227	97.83 116	97.92 160

v8		91.29 224	90.53 223	93.57 237	94.15 291	88.12 234	97.34 117	97.06 200	88.99 206	88.32 264	94.26 273	83.08 175	98.01 240	87.62 228	83.92 314	94.57 303

pmmvs5		89.86 269	88.87 271	92.82 265	92.86 322	86.23 272	96.26 220	95.39 285	84.24 305	87.12 289	94.51 255	74.27 296	97.36 305	87.61 229	87.57 267	94.86 285

Fast-Effi-MVS+-dtu		92.29 181	91.99 168	93.21 253	95.27 242	85.52 282	97.03 144	96.63 240	92.09 120	89.11 247	95.14 229	80.33 228	98.08 228	87.54 230	94.74 182	96.03 223

OpenMVS	89.19 12	92.86 162	91.68 178	96.40 96	95.34 235	92.73 83	98.27 27	98.12 56	84.86 298	85.78 304	97.75 86	78.89 256	99.74 31	87.50 231	98.65 97	96.73 203

miper_lstm_enhance		90.50 255	90.06 244	91.83 287	95.33 238	83.74 303	93.86 304	96.70 233	87.56 255	87.79 277	93.81 290	83.45 169	96.92 319	87.39 232	84.62 303	94.82 289

IterMVS-SCA-FT		90.31 257	89.81 252	91.82 288	95.52 224	84.20 300	94.30 292	96.15 260	90.61 167	87.39 285	94.27 271	75.80 287	96.44 325	87.34 233	86.88 277	94.82 289

PLC	91.00 6	94.11 116	93.43 125	96.13 112	98.58 71	91.15 140	96.69 181	97.39 170	87.29 261	91.37 183	96.71 143	88.39 100	99.52 93	87.33 234	97.13 139	97.73 170
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
tpm		90.25 259	89.74 257	91.76 293	93.92 297	79.73 338	93.98 300	93.54 334	88.28 230	91.99 174	93.25 305	77.51 276	97.44 299	87.30 235	87.94 263	98.12 152

GA-MVS		91.38 216	90.31 229	94.59 188	94.65 275	87.62 244	94.34 290	96.19 259	90.73 159	90.35 204	93.83 287	71.84 307	97.96 249	87.22 236	93.61 197	98.21 149

BH-untuned		92.94 158	92.62 148	93.92 221	97.22 138	86.16 275	96.40 206	96.25 256	90.06 179	89.79 225	96.17 178	83.19 171	98.35 201	87.19 237	97.27 134	97.24 189

v144192		91.06 233	90.28 231	93.39 244	93.66 306	87.23 251	96.83 167	97.07 198	87.43 257	89.69 228	94.28 270	81.48 210	98.00 241	87.18 238	84.92 300	94.93 280

RPSCF		90.75 246	90.86 206	90.42 316	96.84 161	76.29 349	95.61 253	96.34 251	83.89 309	91.38 182	97.87 75	76.45 282	98.78 162	87.16 239	92.23 211	96.20 213

PS-CasMVS		91.55 207	90.84 209	93.69 231	94.96 258	88.28 226	97.84 64	98.24 34	91.46 136	88.04 273	95.80 197	79.67 240	97.48 295	87.02 240	84.54 305	95.31 260

pm-mvs1		90.72 248	89.65 260	93.96 216	94.29 290	89.63 181	97.79 68	96.82 225	89.07 203	86.12 303	95.48 219	78.61 259	97.78 269	86.97 241	81.67 326	94.46 305

IterMVS		90.15 263	89.67 258	91.61 295	95.48 226	83.72 304	94.33 291	96.12 261	89.99 180	87.31 288	94.15 279	75.78 289	96.27 328	86.97 241	86.89 276	94.83 287
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
F-COLMAP		93.58 135	92.98 136	95.37 157	98.40 78	88.98 210	97.18 136	97.29 181	87.75 250	90.49 200	97.10 127	85.21 143	99.50 97	86.70 243	96.72 146	97.63 174

PVSNet	86.66 18	92.24 184	91.74 177	93.73 227	97.77 122	83.69 307	92.88 324	96.72 228	87.91 242	93.00 151	94.86 239	78.51 260	99.05 143	86.53 244	97.45 128	98.47 131

v1192		91.07 232	90.23 235	93.58 236	93.70 304	87.82 241	96.73 175	97.07 198	87.77 248	89.58 231	94.32 268	80.90 219	97.97 245	86.52 245	85.48 287	94.95 276

新几何1				97.32 56	98.60 68	93.59 61		97.75 116	81.58 328	95.75 94	97.85 78	90.04 82	99.67 49	86.50 246	99.13 75	98.69 116

v10		91.04 234	90.23 235	93.49 239	94.12 292	88.16 233	97.32 120	97.08 197	88.26 231	88.29 266	94.22 276	82.17 199	97.97 245	86.45 247	84.12 309	94.33 309

v1921920		90.85 242	90.03 245	93.29 249	93.55 307	86.96 259	96.74 174	97.04 203	87.36 259	89.52 235	94.34 265	80.23 230	97.97 245	86.27 248	85.21 293	94.94 278

MDTV_nov1_ep13_2view						70.35 356	93.10 322		83.88 310	93.55 137		82.47 193		86.25 249		98.38 142

test_post1							92.81 326				16.58 368	80.53 223	97.68 276	86.20 250

SCA		91.84 196	91.18 199	93.83 223	95.59 220	84.95 292	94.72 277	95.58 281	90.82 156	92.25 167	93.69 293	75.80 287	98.10 223	86.20 250	95.98 157	98.45 133

PAPR		94.18 111	93.42 127	96.48 90	97.64 128	91.42 126	95.55 254	97.71 127	88.99 206	92.34 165	95.82 196	89.19 88	99.11 133	86.14 252	97.38 129	98.90 99

GBi-Net		91.35 219	90.27 232	94.59 188	96.51 179	91.18 137	97.50 101	96.93 211	88.82 215	89.35 239	94.51 255	73.87 298	97.29 308	86.12 253	88.82 255	95.31 260

test1		91.35 219	90.27 232	94.59 188	96.51 179	91.18 137	97.50 101	96.93 211	88.82 215	89.35 239	94.51 255	73.87 298	97.29 308	86.12 253	88.82 255	95.31 260

FMVSNet3		91.78 197	90.69 216	95.03 167	96.53 178	92.27 99	97.02 147	96.93 211	89.79 187	89.35 239	94.65 251	77.01 278	97.47 296	86.12 253	88.82 255	95.35 258

EPMVS		90.70 249	89.81 252	93.37 246	94.73 272	84.21 299	93.67 310	88.02 357	89.50 192	92.38 162	93.49 300	77.82 274	97.78 269	86.03 256	92.68 205	98.11 155

MVS		91.71 199	90.44 224	95.51 148	95.20 248	91.59 119	96.04 232	97.45 160	73.44 352	87.36 286	95.60 211	85.42 141	99.10 134	85.97 257	97.46 124	95.83 230

testdata2													99.67 49	85.96 258

K. test v3		87.64 293	86.75 294	90.32 317	93.02 321	79.48 340	96.61 191	92.08 347	90.66 163	80.25 341	94.09 280	67.21 334	96.65 324	85.96 258	80.83 330	94.83 287

WR-MVS_H		92.00 192	91.35 188	93.95 217	95.09 253	89.47 190	98.04 46	98.68 7	91.46 136	88.34 263	94.68 249	85.86 136	97.56 287	85.77 260	84.24 308	94.82 289

gg-mvs-nofinetune		87.82 291	85.61 300	94.44 195	94.46 282	89.27 203	91.21 338	84.61 362	80.88 332	89.89 223	74.98 356	71.50 309	97.53 291	85.75 261	97.21 136	96.51 207

tpm2		89.96 265	89.21 266	92.23 279	94.91 264	81.25 323	93.78 306	94.42 322	80.62 335	91.56 179	93.44 302	76.44 283	97.94 252	85.60 262	92.08 218	97.49 184

v1240		90.70 249	89.85 250	93.23 251	93.51 310	86.80 260	96.61 191	97.02 206	87.16 264	89.58 231	94.31 269	79.55 243	97.98 242	85.52 263	85.44 288	94.90 283

PEN-MVS		91.20 227	90.44 224	93.48 240	94.49 281	87.91 239	97.76 71	98.18 46	91.29 143	87.78 278	95.74 204	80.35 227	97.33 306	85.46 264	82.96 322	95.19 270

QAPM		93.45 139	92.27 161	96.98 74	96.77 166	92.62 87	98.39 20	98.12 56	84.50 303	88.27 267	97.77 85	82.39 195	99.81 26	85.40 265	98.81 92	98.51 125

EU-MVSNet		88.72 283	88.90 270	88.20 328	93.15 319	74.21 352	96.63 190	94.22 328	85.18 292	87.32 287	95.97 186	76.16 285	94.98 343	85.27 266	86.17 280	95.41 251

BH-w/o		92.14 190	91.75 175	93.31 248	96.99 156	85.73 279	95.67 249	95.69 275	88.73 220	89.26 244	94.82 242	82.97 180	98.07 231	85.26 267	96.32 155	96.13 219

FMVSNet2		91.31 222	90.08 241	94.99 168	96.51 179	92.21 100	97.41 109	96.95 209	88.82 215	88.62 258	94.75 245	73.87 298	97.42 301	85.20 268	88.55 260	95.35 258

PM-MVS		83.48 317	81.86 321	88.31 327	87.83 355	77.59 347	93.43 314	91.75 349	86.91 267	80.63 337	89.91 338	44.42 360	95.84 334	85.17 269	76.73 341	91.50 346

LF4IMVS		87.94 290	87.25 287	89.98 320	92.38 332	80.05 336	94.38 288	95.25 295	87.59 254	84.34 316	94.74 246	64.31 345	97.66 279	84.83 270	87.45 268	92.23 339

PatchmatchNet		91.91 194	91.35 188	93.59 235	95.38 230	84.11 301	93.15 320	95.39 285	89.54 190	92.10 172	93.68 295	82.82 184	98.13 218	84.81 271	95.32 170	98.52 123
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
pmmvs6		87.81 292	86.19 296	92.69 269	91.32 337	86.30 270	97.34 117	96.41 249	80.59 336	84.05 323	94.37 264	67.37 333	97.67 277	84.75 272	79.51 334	94.09 317

v7n		90.76 244	89.86 249	93.45 243	93.54 308	87.60 245	97.70 81	97.37 173	88.85 212	87.65 280	94.08 281	81.08 214	98.10 223	84.68 273	83.79 316	94.66 301

SixPastTwentyTwo		89.15 275	88.54 275	90.98 306	93.49 311	80.28 333	96.70 179	94.70 315	90.78 157	84.15 320	95.57 212	71.78 308	97.71 275	84.63 274	85.07 296	94.94 278

TDRefinement		86.53 299	84.76 309	91.85 286	82.23 359	84.25 298	96.38 209	95.35 288	84.97 297	84.09 321	94.94 234	65.76 343	98.34 203	84.60 275	74.52 344	92.97 328

MVS_0304		88.79 281	87.57 283	92.46 272	94.65 275	86.15 276	96.40 206	97.17 188	86.44 274	88.02 274	91.71 328	56.68 355	97.03 313	84.47 276	92.58 207	94.19 314

ACMH	87.59 16	90.53 253	89.42 262	93.87 222	96.21 195	87.92 237	97.24 127	96.94 210	88.45 226	83.91 324	96.27 174	71.92 306	98.62 178	84.43 277	89.43 251	95.05 274
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
ACMH+	87.92 14	90.20 261	89.18 267	93.25 250	96.48 182	86.45 268	96.99 152	96.68 234	88.83 214	84.79 314	96.22 175	70.16 320	98.53 187	84.42 278	88.04 262	94.77 297

MS-PatchMatch		90.27 258	89.77 254	91.78 291	94.33 287	84.72 295	95.55 254	96.73 227	86.17 279	86.36 300	95.28 225	71.28 311	97.80 266	84.09 279	98.14 110	92.81 331

PatchMatch-RL		92.90 160	92.02 167	95.56 144	98.19 100	90.80 151	95.27 268	97.18 186	87.96 240	91.86 177	95.68 208	80.44 225	98.99 147	84.01 280	97.54 123	96.89 198

lessismore_v0				90.45 315	91.96 335	79.09 344		87.19 360		80.32 340	94.39 262	66.31 339	97.55 288	84.00 281	76.84 340	94.70 299

CMPMVS	62.92 21	85.62 310	84.92 307	87.74 330	89.14 350	73.12 354	94.17 296	96.80 226	73.98 350	73.65 351	94.93 235	66.36 337	97.61 284	83.95 282	91.28 229	92.48 337
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
MVP-Stereo		90.74 247	90.08 241	92.71 268	93.19 318	88.20 230	95.86 242	96.27 254	86.07 280	84.86 313	94.76 244	77.84 273	97.75 272	83.88 283	98.01 112	92.17 342
Qingsong Yan: MVP-Stereo: A Parallel Multi-View Patchmatch Stereo Method with Dilation Matching for Photogrammetric Application.
LS3D		93.57 136	92.61 149	96.47 91	97.59 132	91.61 117	97.67 83	97.72 123	85.17 293	90.29 205	98.34 41	84.60 150	99.73 32	83.85 284	98.27 105	98.06 156

DTE-MVSNet		90.56 252	89.75 256	93.01 258	93.95 296	87.25 249	97.64 90	97.65 133	90.74 158	87.12 289	95.68 208	79.97 235	97.00 317	83.33 285	81.66 327	94.78 296

BH-RMVSNet		92.72 168	91.97 169	94.97 171	97.16 142	87.99 236	96.15 228	95.60 279	90.62 166	91.87 176	97.15 125	78.41 263	98.57 184	83.16 286	97.60 122	98.36 144

pmmvs-eth3d		86.22 304	84.45 310	91.53 296	88.34 353	87.25 249	94.47 283	95.01 304	83.47 316	79.51 344	89.61 340	69.75 323	95.71 336	83.13 287	76.73 341	91.64 343

FMVSNet1		89.88 268	88.31 277	94.59 188	95.41 228	91.18 137	97.50 101	96.93 211	86.62 272	87.41 284	94.51 255	65.94 342	97.29 308	83.04 288	87.43 269	95.31 260

MDTV_nov1_ep13			90.76 212		95.22 246	80.33 331	93.03 323	95.28 292	88.14 237	92.84 157	93.83 287	81.34 211	98.08 228	82.86 289	94.34 186

TR-MVS		91.48 212	90.59 219	94.16 206	96.40 187	87.33 246	95.67 249	95.34 291	87.68 252	91.46 181	95.52 217	76.77 280	98.35 201	82.85 290	93.61 197	96.79 202

JIA-IIPM		88.26 288	87.04 292	91.91 284	93.52 309	81.42 322	89.38 349	94.38 323	80.84 333	90.93 196	80.74 354	79.22 247	97.92 255	82.76 291	91.62 222	96.38 211

PVSNet_0	82.17 19	85.46 311	83.64 314	90.92 307	95.27 242	79.49 339	90.55 342	95.60 279	83.76 312	83.00 330	89.95 337	71.09 312	97.97 245	82.75 292	60.79 358	95.31 260

ambc				86.56 334	83.60 357	70.00 357	85.69 354	94.97 307		80.60 338	88.45 342	37.42 362	96.84 321	82.69 293	75.44 343	92.86 330

USDC		88.94 277	87.83 282	92.27 278	94.66 274	84.96 291	93.86 304	95.90 267	87.34 260	83.40 326	95.56 214	67.43 332	98.19 213	82.64 294	89.67 250	93.66 321

ITE_SJBPF				92.43 274	95.34 235	85.37 285		95.92 265	91.47 135	87.75 279	96.39 169	71.00 313	97.96 249	82.36 295	89.86 248	93.97 318

UnsupCasMVSNet_eth		85.99 306	84.45 310	90.62 313	89.97 345	82.40 317	93.62 312	97.37 173	89.86 182	78.59 346	92.37 316	65.25 344	95.35 342	82.27 296	70.75 350	94.10 315

GG-mvs-BLEND				93.62 233	93.69 305	89.20 204	92.39 332	83.33 363		87.98 276	89.84 339	71.00 313	96.87 320	82.08 297	95.40 169	94.80 292

thres600view7		92.49 172	91.60 180	95.18 161	97.91 114	89.47 190	97.65 86	94.66 316	92.18 119	93.33 144	94.91 236	78.06 270	99.10 134	81.61 298	94.06 191	96.98 193

LTVRE_ROB	88.41 13	90.99 236	89.92 247	94.19 204	96.18 198	89.55 186	96.31 216	97.09 196	87.88 243	85.67 305	95.91 190	78.79 257	98.57 184	81.50 299	89.98 246	94.44 306
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
tpmvs		89.83 270	89.15 268	91.89 285	94.92 261	80.30 332	93.11 321	95.46 284	86.28 276	88.08 272	92.65 311	80.44 225	98.52 188	81.47 300	89.92 247	96.84 200

thres100view900		92.43 173	91.58 181	94.98 170	97.92 113	89.37 196	97.71 80	94.66 316	92.20 115	93.31 145	94.90 237	78.06 270	99.08 139	81.40 301	94.08 188	96.48 209

tfpn200view9		92.38 176	91.52 184	94.95 173	97.85 117	89.29 200	97.41 109	94.88 311	92.19 117	93.27 147	94.46 260	78.17 266	99.08 139	81.40 301	94.08 188	96.48 209

thres400		92.42 174	91.52 184	95.12 165	97.85 117	89.29 200	97.41 109	94.88 311	92.19 117	93.27 147	94.46 260	78.17 266	99.08 139	81.40 301	94.08 188	96.98 193

DP-MVS		92.76 167	91.51 186	96.52 85	98.77 57	90.99 143	97.38 115	96.08 262	82.38 322	89.29 242	97.87 75	83.77 162	99.69 44	81.37 304	96.69 147	98.89 101

thres200		92.23 185	91.39 187	94.75 186	97.61 130	89.03 209	96.60 193	95.09 302	92.08 121	93.28 146	94.00 283	78.39 264	99.04 145	81.26 305	94.18 187	96.19 214

CR-MVSNet		90.82 243	89.77 254	93.95 217	94.45 283	87.19 252	90.23 344	95.68 277	86.89 268	92.40 160	92.36 319	80.91 217	97.05 312	81.09 306	93.95 192	97.60 179

MSDG		91.42 214	90.24 234	94.96 172	97.15 144	88.91 211	93.69 309	96.32 252	85.72 285	86.93 295	96.47 163	80.24 229	98.98 148	80.57 307	95.05 176	96.98 193

dp		88.90 279	88.26 279	90.81 309	94.58 280	76.62 348	92.85 325	94.93 309	85.12 294	90.07 219	93.07 306	75.81 286	98.12 221	80.53 308	87.42 270	97.71 171

tpm cat1		88.36 286	87.21 289	91.81 289	95.13 251	80.55 329	92.58 329	95.70 274	74.97 349	87.45 282	91.96 324	78.01 272	98.17 216	80.39 309	88.74 258	96.72 204

DIV-MVS_2432*1600		85.95 307	84.95 306	88.96 325	89.55 349	79.11 343	95.13 273	96.42 248	85.91 282	84.07 322	90.48 333	70.03 321	94.82 344	80.04 310	72.94 348	92.94 329

AllTest		90.23 260	88.98 269	93.98 213	97.94 111	86.64 263	96.51 198	95.54 282	85.38 289	85.49 307	96.77 140	70.28 318	99.15 129	80.02 311	92.87 201	96.15 217

TestCases				93.98 213	97.94 111	86.64 263		95.54 282	85.38 289	85.49 307	96.77 140	70.28 318	99.15 129	80.02 311	92.87 201	96.15 217

ADS-MVSNet2		89.45 272	88.59 274	92.03 282	95.86 210	82.26 318	90.93 339	94.32 326	83.23 318	91.28 190	91.81 326	79.01 253	95.99 330	79.52 313	91.39 227	97.84 165

ADS-MVSNet		89.89 267	88.68 273	93.53 238	95.86 210	84.89 293	90.93 339	95.07 303	83.23 318	91.28 190	91.81 326	79.01 253	97.85 261	79.52 313	91.39 227	97.84 165

our_test_3		88.78 282	87.98 281	91.20 304	92.45 330	82.53 314	93.61 313	95.69 275	85.77 284	84.88 312	93.71 292	79.99 234	96.78 323	79.47 315	86.24 279	94.28 312

EPNet_dtu		91.71 199	91.28 193	92.99 259	93.76 303	83.71 305	96.69 181	95.28 292	93.15 81	87.02 293	95.95 188	83.37 170	97.38 304	79.46 316	96.84 141	97.88 163
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
TransMVSNet (Re)		88.94 277	87.56 284	93.08 257	94.35 286	88.45 224	97.73 75	95.23 296	87.47 256	84.26 318	95.29 223	79.86 237	97.33 306	79.44 317	74.44 345	93.45 325

EG-PatchMatch MVS		87.02 297	85.44 301	91.76 293	92.67 326	85.00 290	96.08 231	96.45 247	83.41 317	79.52 343	93.49 300	57.10 354	97.72 274	79.34 318	90.87 237	92.56 335

Patchmtry		88.64 284	87.25 287	92.78 267	94.09 293	86.64 263	89.82 347	95.68 277	80.81 334	87.63 281	92.36 319	80.91 217	97.03 313	78.86 319	85.12 295	94.67 300

FMVSNet5		87.29 295	85.79 299	91.78 291	94.80 269	87.28 247	95.49 257	95.28 292	84.09 307	83.85 325	91.82 325	62.95 348	94.17 348	78.48 320	85.34 291	93.91 319

COLMAP_ROB	87.81 15	90.40 256	89.28 265	93.79 226	97.95 110	87.13 255	96.92 159	95.89 268	82.83 320	86.88 297	97.18 122	73.77 301	99.29 118	78.44 321	93.62 196	94.95 276
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
Anonymous20240521		86.42 301	85.44 301	89.34 324	90.33 342	79.79 337	96.73 175	95.92 265	83.71 313	83.25 327	91.36 331	63.92 346	96.01 329	78.39 322	85.36 290	92.22 340

test0.0.03 1		89.37 274	88.70 272	91.41 300	92.47 329	85.63 280	95.22 271	92.70 342	91.11 152	86.91 296	93.65 297	79.02 251	93.19 353	78.00 323	89.18 253	95.41 251

MIMVSNet		88.50 285	86.76 293	93.72 229	94.84 267	87.77 242	91.39 334	94.05 329	86.41 275	87.99 275	92.59 313	63.27 347	95.82 335	77.44 324	92.84 203	97.57 182

MDA-MVSNet_test_wron		85.87 308	84.23 312	90.80 311	92.38 332	82.57 313	93.17 318	95.15 299	82.15 323	67.65 353	92.33 322	78.20 265	95.51 340	77.33 325	79.74 331	94.31 311

YYNet1		85.87 308	84.23 312	90.78 312	92.38 332	82.46 316	93.17 318	95.14 300	82.12 324	67.69 352	92.36 319	78.16 268	95.50 341	77.31 326	79.73 332	94.39 307

UnsupCasMVSNet_bld		82.13 321	79.46 324	90.14 319	88.00 354	82.47 315	90.89 341	96.62 242	78.94 342	75.61 348	84.40 352	56.63 356	96.31 327	77.30 327	66.77 354	91.63 344

KD-MVS_2432*1600		84.81 314	82.64 317	91.31 301	91.07 339	85.34 286	91.22 336	95.75 272	85.56 287	83.09 328	90.21 335	67.21 334	95.89 331	77.18 328	62.48 356	92.69 332

miper_refine_blended		84.81 314	82.64 317	91.31 301	91.07 339	85.34 286	91.22 336	95.75 272	85.56 287	83.09 328	90.21 335	67.21 334	95.89 331	77.18 328	62.48 356	92.69 332

PCF-MVS	89.48 11	91.56 206	89.95 246	96.36 101	96.60 171	92.52 90	92.51 330	97.26 182	79.41 340	88.90 249	96.56 159	84.04 160	99.55 84	77.01 330	97.30 133	97.01 192
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
testgi		87.97 289	87.21 289	90.24 318	92.86 322	80.76 325	96.67 184	94.97 307	91.74 128	85.52 306	95.83 195	62.66 349	94.47 347	76.25 331	88.36 261	95.48 245

TinyColmap		86.82 298	85.35 304	91.21 303	94.91 264	82.99 312	93.94 302	94.02 331	83.58 314	81.56 333	94.68 249	62.34 350	98.13 218	75.78 332	87.35 272	92.52 336

ppachtmachnet_test		88.35 287	87.29 286	91.53 296	92.45 330	83.57 308	93.75 307	95.97 264	84.28 304	85.32 310	94.18 277	79.00 255	96.93 318	75.71 333	84.99 299	94.10 315

PAPM		91.52 210	90.30 230	95.20 160	95.30 241	89.83 178	93.38 316	96.85 223	86.26 277	88.59 259	95.80 197	84.88 147	98.15 217	75.67 334	95.93 159	97.63 174

CL-MVSNet_2432*1600		86.31 303	85.15 305	89.80 322	88.83 351	81.74 321	93.93 303	96.22 257	86.67 271	85.03 311	90.80 332	78.09 269	94.50 345	74.92 335	71.86 349	93.15 327

tfpnnormal		89.70 271	88.40 276	93.60 234	95.15 249	90.10 168	97.56 97	98.16 50	87.28 262	86.16 302	94.63 252	77.57 275	98.05 234	74.48 336	84.59 304	92.65 334

DSMNet-mixed		86.34 302	86.12 298	87.00 333	89.88 346	70.43 355	94.93 275	90.08 355	77.97 346	85.42 309	92.78 310	74.44 295	93.96 349	74.43 337	95.14 172	96.62 205

Patchmatch-test		89.42 273	87.99 280	93.70 230	95.27 242	85.11 288	88.98 350	94.37 324	81.11 330	87.10 291	93.69 293	82.28 196	97.50 294	74.37 338	94.76 180	98.48 130

LCM-MVSNet		72.55 324	69.39 328	82.03 336	70.81 366	65.42 361	90.12 346	94.36 325	55.02 358	65.88 355	81.72 353	24.16 369	89.96 355	74.32 339	68.10 353	90.71 350

new-patchmatchnet		83.18 318	81.87 320	87.11 332	86.88 356	75.99 350	93.70 308	95.18 298	85.02 296	77.30 347	88.40 343	65.99 341	93.88 350	74.19 340	70.18 351	91.47 347

MDA-MVSNet-bldmvs		85.00 312	82.95 316	91.17 305	93.13 320	83.33 309	94.56 281	95.00 305	84.57 302	65.13 357	92.65 311	70.45 316	95.85 333	73.57 341	77.49 338	94.33 309

pmmvs3		79.97 322	77.50 326	87.39 331	82.80 358	79.38 341	92.70 327	90.75 354	70.69 353	78.66 345	87.47 350	51.34 358	93.40 351	73.39 342	69.65 352	89.38 352

test_method		66.11 328	64.89 331	69.79 343	72.62 364	35.23 371	65.19 362	92.83 341	20.35 364	65.20 356	88.08 347	43.14 361	82.70 360	73.12 343	63.46 355	91.45 348

PatchT		88.87 280	87.42 285	93.22 252	94.08 294	85.10 289	89.51 348	94.64 318	81.92 325	92.36 163	88.15 346	80.05 233	97.01 316	72.43 344	93.65 195	97.54 183

Anonymous20231206		87.09 296	86.14 297	89.93 321	91.22 338	80.35 330	96.11 229	95.35 288	83.57 315	84.16 319	93.02 307	73.54 303	95.61 337	72.16 345	86.14 281	93.84 320

MVS-HIRNet		82.47 320	81.21 322	86.26 335	95.38 230	69.21 358	88.96 351	89.49 356	66.28 354	80.79 336	74.08 358	68.48 327	97.39 303	71.93 346	95.47 167	92.18 341

new_pmnet		82.89 319	81.12 323	88.18 329	89.63 347	80.18 334	91.77 333	92.57 343	76.79 348	75.56 350	88.23 345	61.22 351	94.48 346	71.43 347	82.92 323	89.87 351

TAPA-MVS	90.10 7	92.30 180	91.22 197	95.56 144	98.33 85	89.60 183	96.79 171	97.65 133	81.83 326	91.52 180	97.23 120	87.94 104	98.91 153	71.31 348	98.37 103	98.17 150
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
test20.03		86.14 305	85.40 303	88.35 326	90.12 343	80.06 335	95.90 241	95.20 297	88.59 221	81.29 334	93.62 298	71.43 310	92.65 354	71.26 349	81.17 329	92.34 338

tmp_tt		51.94 334	53.82 334	46.29 348	33.73 370	45.30 369	78.32 359	67.24 369	18.02 365	50.93 361	87.05 351	52.99 357	53.11 366	70.76 350	25.29 364	40.46 362

MIMVSNet1		84.93 313	83.05 315	90.56 314	89.56 348	84.84 294	95.40 260	95.35 288	83.91 308	80.38 339	92.21 323	57.23 353	93.34 352	70.69 351	82.75 325	93.50 323

RPMNet		88.98 276	87.05 291	94.77 184	94.45 283	87.19 252	90.23 344	98.03 84	77.87 347	92.40 160	87.55 349	80.17 231	99.51 94	68.84 352	93.95 192	97.60 179

N_pmnet		78.73 323	78.71 325	78.79 338	92.80 324	46.50 367	94.14 297	43.71 370	78.61 343	80.83 335	91.66 329	74.94 293	96.36 326	67.24 353	84.45 306	93.50 323

OpenMVS_ROB	81.14 20	84.42 316	82.28 319	90.83 308	90.06 344	84.05 302	95.73 248	94.04 330	73.89 351	80.17 342	91.53 330	59.15 352	97.64 280	66.92 354	89.05 254	90.80 349

PMMVS2		70.19 326	66.92 329	80.01 337	76.35 360	65.67 360	86.22 353	87.58 359	64.83 356	62.38 358	80.29 355	26.78 367	88.49 357	63.79 355	54.07 359	85.88 353

test_0402		86.46 300	84.79 308	91.45 298	95.02 256	85.55 281	96.29 218	94.89 310	80.90 331	82.21 331	93.97 285	68.21 329	97.29 308	62.98 356	88.68 259	91.51 345

DeepMVS_CX				74.68 342	90.84 341	64.34 362		81.61 365	65.34 355	67.47 354	88.01 348	48.60 359	80.13 362	62.33 357	73.68 347	79.58 356

FPMVS		71.27 325	69.85 327	75.50 340	74.64 361	59.03 363	91.30 335	91.50 351	58.80 357	57.92 359	88.28 344	29.98 365	85.53 359	53.43 358	82.84 324	81.95 355

ANet_high		63.94 329	59.58 332	77.02 339	61.24 368	66.06 359	85.66 355	87.93 358	78.53 344	42.94 362	71.04 359	25.42 368	80.71 361	52.60 359	30.83 362	84.28 354

Gipuma		67.86 327	65.41 330	75.18 341	92.66 327	73.45 353	66.50 361	94.52 320	53.33 359	57.80 360	66.07 360	30.81 363	89.20 356	48.15 360	78.88 337	62.90 359
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
PMVS	53.92 22	58.58 330	55.40 333	68.12 344	51.00 369	48.64 365	78.86 358	87.10 361	46.77 360	35.84 366	74.28 357	8.76 370	86.34 358	42.07 361	73.91 346	69.38 357
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
MVE	50.73 23	53.25 332	48.81 337	66.58 345	65.34 367	57.50 364	72.49 360	70.94 368	40.15 363	39.28 365	63.51 361	6.89 372	73.48 365	38.29 362	42.38 360	68.76 358
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
E-PMN		53.28 331	52.56 335	55.43 346	74.43 362	47.13 366	83.63 357	76.30 366	42.23 361	42.59 363	62.22 362	28.57 366	74.40 363	31.53 363	31.51 361	44.78 360

EMVS		52.08 333	51.31 336	54.39 347	72.62 364	45.39 368	83.84 356	75.51 367	41.13 362	40.77 364	59.65 363	30.08 364	73.60 364	28.31 364	29.90 363	44.18 361

wuyk23d		25.11 335	24.57 339	26.74 349	73.98 363	39.89 370	57.88 363	9.80 371	12.27 366	10.39 367	6.97 369	7.03 371	36.44 367	25.43 365	17.39 365	3.89 365

testmvs		13.36 337	16.33 340	4.48 351	5.04 371	2.26 373	93.18 317	3.28 372	2.70 367	8.24 368	21.66 365	2.29 374	2.19 368	7.58 366	2.96 366	9.00 364

test123		13.04 338	15.66 341	5.18 350	4.51 372	3.45 372	92.50 331	1.81 373	2.50 368	7.58 369	20.15 366	3.67 373	2.18 369	7.13 367	1.07 367	9.90 363

uanet_test		0.00 341	0.00 344	0.00 352	0.00 373	0.00 374	0.00 364	0.00 374	0.00 369	0.00 370	0.00 370	0.00 375	0.00 370	0.00 368	0.00 368	0.00 366

cdsmvs_eth3d_5k		23.24 336	30.99 338	0.00 352	0.00 373	0.00 374	0.00 364	97.63 135	0.00 369	0.00 370	96.88 137	84.38 154	0.00 370	0.00 368	0.00 368	0.00 366

pcd_1.5k_mvsjas		7.39 340	9.85 343	0.00 352	0.00 373	0.00 374	0.00 364	0.00 374	0.00 369	0.00 370	0.00 370	88.65 96	0.00 370	0.00 368	0.00 368	0.00 366

sosnet-low-res		0.00 341	0.00 344	0.00 352	0.00 373	0.00 374	0.00 364	0.00 374	0.00 369	0.00 370	0.00 370	0.00 375	0.00 370	0.00 368	0.00 368	0.00 366

sosnet		0.00 341	0.00 344	0.00 352	0.00 373	0.00 374	0.00 364	0.00 374	0.00 369	0.00 370	0.00 370	0.00 375	0.00 370	0.00 368	0.00 368	0.00 366

uncertanet		0.00 341	0.00 344	0.00 352	0.00 373	0.00 374	0.00 364	0.00 374	0.00 369	0.00 370	0.00 370	0.00 375	0.00 370	0.00 368	0.00 368	0.00 366

Regformer		0.00 341	0.00 344	0.00 352	0.00 373	0.00 374	0.00 364	0.00 374	0.00 369	0.00 370	0.00 370	0.00 375	0.00 370	0.00 368	0.00 368	0.00 366

ab-mvs-re		8.06 339	10.74 342	0.00 352	0.00 373	0.00 374	0.00 364	0.00 374	0.00 369	0.00 370	96.69 147	0.00 375	0.00 370	0.00 368	0.00 368	0.00 366

uanet		0.00 341	0.00 344	0.00 352	0.00 373	0.00 374	0.00 364	0.00 374	0.00 369	0.00 370	0.00 370	0.00 375	0.00 370	0.00 368	0.00 368	0.00 366

eth-test2					0.00 373

eth-test					0.00 373

test_241102_ONE					99.42 6	95.30 15		98.27 28	95.09 20	99.19 1	98.81 8	95.54 3	99.65 53

save fliter					98.91 49	94.28 35	97.02 147	98.02 88	95.35 8

test0726					99.45 2	95.36 10	98.31 23	98.29 24	94.92 24	98.99 4	98.92 2	95.08 5

GSMVS																98.45 133

test_part2					99.28 25	95.74 6				98.10 17

sam_mvs1												82.76 185				98.45 133

sam_mvs												81.94 204

MTGPA								98.08 64

test_post											17.58 367	81.76 206	98.08 228

patchmatchnet-post											90.45 334	82.65 189	98.10 223

MTMP							97.86 60	82.03 364

TEST9					98.70 60	94.19 40	96.41 203	98.02 88	88.17 234	96.03 82	97.56 105	92.74 24	99.59 68

test_8					98.67 62	94.06 49	96.37 210	98.01 91	88.58 222	95.98 87	97.55 107	92.73 25	99.58 71

agg_prior					98.67 62	93.79 55		98.00 93		95.68 97			99.57 79

test_prior4						93.66 59	96.42 202

test_prior				97.23 62	98.67 62	92.99 76		98.00 93					99.41 107			99.29 62

新几何2							95.79 246

旧先验1					98.38 81	93.38 67		97.75 116			98.09 62	92.30 38			99.01 85	99.16 70

原ACMM2							95.67 249

test222					98.24 93	92.21 100	95.33 263	97.60 137	79.22 341	95.25 108	97.84 81	88.80 94			99.15 73	98.72 113

segment_acmp												92.89 22

testdata1							95.26 270		93.10 84

test12				97.65 44	98.46 74	94.26 37		97.66 131		95.52 106		90.89 69	99.46 101		99.25 65	99.22 67

plane_prior7					96.21 195	89.98 174

plane_prior6					96.10 205	90.00 170						81.32 212

plane_prior4											96.64 150

plane_prior3						90.00 170			94.46 41	91.34 184

plane_prior2							97.74 73		94.85 26

plane_prior1					96.14 203

plane_prior						89.99 172	97.24 127		94.06 49						92.16 215

n2								0.00 374

nn								0.00 374

door-mid								91.06 353

test11								97.88 105

door								91.13 352

HQP5-MVS						89.33 198

HQP-NCC					95.86 210		96.65 185		93.55 65	90.14 208

ACMP_Plane					95.86 210		96.65 185		93.55 65	90.14 208

HQP4-MVS										90.14 208			98.50 189			95.78 233

HQP3-MVS								97.39 170							92.10 216

HQP2-MVS												80.95 215

NP-MVS					95.99 209	89.81 179					95.87 192

ACMMP++_ref															90.30 244

ACMMP++															91.02 233

Test By Simon												88.73 95

High-resolution multi-view benchmark