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.

DVP-MVS		97.93 2	98.23 2	97.58 2	99.05 6	99.31 1	98.64 5	96.62 4	97.56 1	95.08 5	96.61 13	99.64 1	97.32 1	97.91 3	97.31 6	98.77 11	99.26 1

SED-MVS		97.98 1	98.36 1	97.54 3	98.94 17	99.29 2	98.81 3	96.64 3	97.14 2	95.16 4	97.96 2	99.61 2	96.92 11	98.00 1	97.24 8	98.75 12	99.25 2

MSP-MVS		97.70 5	98.09 4	97.24 6	99.00 11	99.17 4	98.76 4	96.41 9	96.91 4	93.88 15	97.72 4	99.04 6	96.93 10	97.29 15	97.31 6	98.45 31	99.23 3

SD-MVS		97.35 7	97.73 7	96.90 15	97.35 45	98.66 13	97.85 25	96.25 11	96.86 5	94.54 9	96.75 11	99.13 5	96.99 7	96.94 23	96.58 22	98.39 39	99.20 4
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
APDe-MVS		97.79 4	97.96 5	97.60 1	99.20 2	99.10 5	98.88 2	96.68 2	96.81 6	94.64 6	97.84 3	98.02 10	97.24 3	97.74 7	97.02 13	98.97 3	99.16 5

TSAR-MVS + MP.		97.31 8	97.64 8	96.92 14	97.28 47	98.56 22	98.61 6	95.48 29	96.72 7	94.03 14	96.73 12	98.29 8	97.15 4	97.61 11	96.42 25	98.96 4	99.13 6
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
TSAR-MVS + ACMM		96.19 23	97.39 12	94.78 38	97.70 40	98.41 34	97.72 27	95.49 28	96.47 10	86.66 66	96.35 15	97.85 12	93.99 50	97.19 18	96.37 27	97.12 124	99.13 6

DPE-MVS		97.83 3	98.13 3	97.48 4	98.83 23	99.19 3	98.99 1	96.70 1	96.05 19	94.39 10	98.30 1	99.47 3	97.02 6	97.75 6	97.02 13	98.98 2	99.10 8

CSCG		95.68 30	95.46 35	95.93 28	98.71 25	99.07 6	97.13 36	93.55 38	95.48 25	93.35 20	90.61 45	93.82 46	95.16 35	94.60 76	95.57 48	97.70 99	99.08 9

SteuartSystems-ACMMP		97.10 14	97.49 9	96.65 19	98.97 13	98.95 8	98.43 8	95.96 18	95.12 29	91.46 29	96.85 9	97.60 17	96.37 24	97.76 5	97.16 10	98.68 13	98.97 10
Skip Steuart: Steuart Systems R&D Blog.
SMA-MVS		97.53 6	97.93 6	97.07 11	99.21 1	99.02 7	98.08 19	96.25 11	96.36 11	93.57 16	96.56 14	99.27 4	96.78 16	97.91 3	97.43 3	98.51 21	98.94 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
canonicalmvs		93.08 54	93.09 54	93.07 61	94.24 78	97.86 50	95.45 57	87.86 99	94.00 44	87.47 58	88.32 54	82.37 102	95.13 36	93.96 92	96.41 26	98.27 53	98.73 12

HPM-MVS++		97.22 10	97.40 11	97.01 12	99.08 4	98.55 23	98.19 14	96.48 6	96.02 20	93.28 21	96.26 17	98.71 7	96.76 17	97.30 14	96.25 34	98.30 49	98.68 13

DeepPCF-MVS	92.65 2	95.50 34	96.96 18	93.79 51	96.44 57	98.21 38	93.51 92	94.08 37	96.94 3	89.29 43	93.08 31	96.77 27	93.82 54	97.68 8	97.40 4	95.59 170	98.65 14

TSAR-MVS + GP.		95.86 28	96.95 20	94.60 43	94.07 83	98.11 42	96.30 44	91.76 51	95.67 21	91.07 32	96.82 10	97.69 16	95.71 31	95.96 48	95.75 46	98.68 13	98.63 15

DeepC-MVS	92.10 3	95.22 35	94.77 39	95.75 31	97.77 38	98.54 24	97.63 28	95.96 18	95.07 32	88.85 47	85.35 73	91.85 54	95.82 29	96.88 25	97.10 11	98.44 32	98.63 15
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
3Dnovator+	90.56 5	95.06 37	94.56 42	95.65 32	98.11 32	98.15 41	97.19 34	91.59 53	95.11 31	93.23 23	81.99 99	94.71 43	95.43 34	96.48 35	96.88 17	98.35 41	98.63 15

ACMMP_NAP		96.93 16	97.27 14	96.53 24	99.06 5	98.95 8	98.24 13	96.06 15	95.66 22	90.96 34	95.63 24	97.71 15	96.53 20	97.66 9	96.68 19	98.30 49	98.61 18

HFP-MVS		97.11 13	97.19 15	97.00 13	98.97 13	98.73 11	98.37 11	95.69 22	96.60 8	93.28 21	96.87 8	96.64 28	97.27 2	96.64 31	96.33 32	98.44 32	98.56 19

MP-MVS		96.56 21	96.72 22	96.37 25	98.93 19	98.48 29	98.04 20	95.55 24	94.32 41	90.95 36	95.88 22	97.02 25	96.29 25	96.77 28	96.01 42	98.47 26	98.56 19
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
APD-MVS		97.12 12	97.05 17	97.19 7	99.04 7	98.63 18	98.45 7	96.54 5	94.81 37	93.50 17	96.10 19	97.40 21	96.81 13	97.05 20	96.82 18	98.80 7	98.56 19
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
CP-MVS		96.68 20	96.59 26	96.77 18	98.85 22	98.58 21	98.18 15	95.51 27	95.34 26	92.94 24	95.21 28	96.25 31	96.79 15	96.44 38	95.77 45	98.35 41	98.56 19

PVSNet_Blended_VisFu		91.92 67	92.39 66	91.36 84	95.45 71	97.85 51	92.25 108	89.54 74	88.53 103	87.47 58	79.82 109	90.53 65	85.47 143	96.31 42	95.16 56	97.99 81	98.56 19

xxxxxxxxxxxxxcwj		95.62 31	94.35 45	97.10 9	98.95 15	98.51 27	97.51 29	96.48 6	96.17 15	94.64 6	97.32 5	76.98 135	96.23 26	96.78 26	96.15 37	98.79 9	98.55 24

SF-MVS		97.20 11	97.29 13	97.10 9	98.95 15	98.51 27	97.51 29	96.48 6	96.17 15	94.64 6	97.32 5	97.57 18	96.23 26	96.78 26	96.15 37	98.79 9	98.55 24

MCST-MVS		96.83 18	97.06 16	96.57 20	98.88 21	98.47 31	98.02 21	96.16 14	95.58 24	90.96 34	95.78 23	97.84 13	96.46 22	97.00 22	96.17 36	98.94 5	98.55 24

CNVR-MVS		97.30 9	97.41 10	97.18 8	99.02 10	98.60 20	98.15 16	96.24 13	96.12 17	94.10 12	95.54 25	97.99 11	96.99 7	97.97 2	97.17 9	98.57 19	98.50 27

X-MVS		96.07 26	96.33 28	95.77 30	98.94 17	98.66 13	97.94 23	95.41 31	95.12 29	88.03 51	93.00 32	96.06 32	95.85 28	96.65 30	96.35 28	98.47 26	98.48 28

train_agg		96.15 25	96.64 25	95.58 34	98.44 28	98.03 45	98.14 18	95.40 32	93.90 45	87.72 55	96.26 17	98.10 9	95.75 30	96.25 43	95.45 50	98.01 79	98.47 29

MSLP-MVS++		96.05 27	95.63 31	96.55 22	98.33 30	98.17 40	96.94 37	94.61 35	94.70 39	94.37 11	89.20 51	95.96 35	96.81 13	95.57 54	97.33 5	98.24 57	98.47 29

zzz-MVS		96.98 15	96.68 23	97.33 5	99.09 3	98.71 12	98.43 8	96.01 16	96.11 18	95.19 3	92.89 33	97.32 22	96.84 12	97.20 16	96.09 40	98.44 32	98.46 31

ACMMPR		96.92 17	96.96 18	96.87 16	98.99 12	98.78 10	98.38 10	95.52 25	96.57 9	92.81 25	96.06 20	95.90 36	97.07 5	96.60 33	96.34 31	98.46 28	98.42 32

QAPM		94.13 47	94.33 47	93.90 48	97.82 37	98.37 36	96.47 42	90.89 60	92.73 55	85.63 77	85.35 73	93.87 45	94.17 48	95.71 53	95.90 43	98.40 37	98.42 32

UGNet		91.52 74	93.41 52	89.32 102	94.13 80	97.15 70	91.83 118	89.01 78	90.62 73	85.86 74	86.83 59	91.73 56	77.40 183	94.68 73	94.43 64	97.71 97	98.40 34
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
DELS-MVS		93.71 50	93.47 51	94.00 45	96.82 54	98.39 35	96.80 39	91.07 58	89.51 93	89.94 41	83.80 83	89.29 70	90.95 84	97.32 12	97.65 2	98.42 35	98.32 35
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
NCCC		96.75 19	96.67 24	96.85 17	99.03 9	98.44 33	98.15 16	96.28 10	96.32 12	92.39 26	92.16 35	97.55 19	96.68 19	97.32 12	96.65 21	98.55 20	98.26 36

MVS_0304		94.30 46	94.68 40	93.86 50	96.33 59	98.48 29	97.41 31	91.20 55	92.75 53	86.96 63	86.03 68	93.81 47	92.64 68	96.89 24	96.54 24	98.61 17	98.24 37

ACMMP		95.54 32	95.49 34	95.61 33	98.27 31	98.53 25	97.16 35	94.86 33	94.88 35	89.34 42	95.36 27	91.74 55	95.50 33	95.51 55	94.16 69	98.50 23	98.22 38
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
CANet		94.85 39	94.92 37	94.78 38	97.25 48	98.52 26	97.20 33	91.81 49	93.25 49	91.06 33	86.29 65	94.46 44	92.99 64	97.02 21	96.68 19	98.34 43	98.20 39

PGM-MVS		96.16 24	96.33 28	95.95 27	99.04 7	98.63 18	98.32 12	92.76 43	93.42 48	90.49 39	96.30 16	95.31 41	96.71 18	96.46 36	96.02 41	98.38 40	98.19 40

3Dnovator	90.28 7	94.70 43	94.34 46	95.11 36	98.06 33	98.21 38	96.89 38	91.03 59	94.72 38	91.45 30	82.87 90	93.10 49	94.61 39	96.24 44	97.08 12	98.63 16	98.16 41

EPNet		93.92 48	94.40 43	93.36 54	97.89 35	96.55 84	96.08 47	92.14 46	91.65 63	89.16 44	94.07 30	90.17 69	87.78 118	95.24 58	94.97 58	97.09 126	98.15 42
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
anonymousdsp		84.51 150	85.85 139	82.95 175	86.30 189	93.51 132	85.77 184	80.38 171	78.25 175	63.42 190	73.51 141	72.20 146	84.64 149	93.21 106	92.16 116	97.19 119	98.14 43

CPTT-MVS		95.54 32	95.07 36	96.10 26	97.88 36	97.98 48	97.92 24	94.86 33	94.56 40	92.16 27	91.01 42	95.71 37	96.97 9	94.56 77	93.50 85	96.81 147	98.14 43

HQP-MVS		92.39 62	92.49 63	92.29 70	95.65 65	95.94 97	95.64 55	92.12 47	92.46 57	79.65 110	91.97 37	82.68 98	92.92 66	93.47 100	92.77 104	97.74 95	98.12 45

DeepC-MVS_fast	93.32 1	96.48 22	96.42 27	96.56 21	98.70 26	98.31 37	97.97 22	95.76 21	96.31 13	92.01 28	91.43 40	95.42 40	96.46 22	97.65 10	97.69 1	98.49 25	98.12 45
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
PHI-MVS		95.86 28	96.93 21	94.61 42	97.60 42	98.65 17	96.49 41	93.13 41	94.07 43	87.91 54	97.12 7	97.17 24	93.90 53	96.46 36	96.93 16	98.64 15	98.10 47

DCV-MVSNet		91.24 76	91.26 80	91.22 85	92.84 108	93.44 133	93.82 83	86.75 109	91.33 68	85.61 78	84.00 82	85.46 83	91.27 80	92.91 107	93.62 80	97.02 130	98.05 48

ETV-MVS		93.80 49	94.57 41	92.91 64	93.98 85	97.50 61	93.62 89	88.70 82	91.95 59	87.57 56	90.21 47	90.79 61	94.56 40	97.20 16	96.35 28	99.02 1	97.98 49

EPP-MVSNet		92.13 64	93.06 55	91.05 86	93.66 97	97.30 65	92.18 109	87.90 95	90.24 80	83.63 90	86.14 67	90.52 67	90.76 86	94.82 69	94.38 65	98.18 63	97.98 49

Anonymous20231211		89.82 95	88.18 108	91.74 74	92.52 113	96.09 96	93.38 94	89.30 77	88.95 98	85.90 73	64.55 184	84.39 87	92.41 71	92.24 120	93.06 99	96.93 139	97.95 51

CDPH-MVS		94.80 42	95.50 33	93.98 47	98.34 29	98.06 43	97.41 31	93.23 40	92.81 52	82.98 93	92.51 34	94.82 42	93.53 58	96.08 46	96.30 33	98.42 35	97.94 52

UniMVSNet (Re)		86.22 127	85.46 143	87.11 125	88.34 157	94.42 111	89.65 147	87.10 108	84.39 135	74.61 126	70.41 155	68.10 163	85.10 146	91.17 137	91.79 125	97.84 89	97.94 52

FC-MVSNet-train		90.55 86	90.19 90	90.97 87	93.78 94	95.16 102	92.11 113	88.85 79	87.64 108	83.38 92	84.36 80	78.41 124	89.53 95	94.69 72	93.15 96	98.15 64	97.92 54

MVS_111021_HR		94.84 40	95.91 30	93.60 52	97.35 45	98.46 32	95.08 59	91.19 56	94.18 42	85.97 70	95.38 26	92.56 51	93.61 57	96.61 32	96.25 34	98.40 37	97.92 54

CLD-MVS		92.50 61	91.96 72	93.13 58	93.93 89	96.24 92	95.69 53	88.77 81	92.92 50	89.01 45	88.19 56	81.74 107	93.13 63	93.63 94	93.08 97	98.23 58	97.91 56
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
LGP-MVS_train		91.83 69	92.04 71	91.58 76	95.46 69	96.18 94	95.97 50	89.85 67	90.45 76	77.76 115	91.92 38	80.07 115	92.34 72	94.27 82	93.47 86	98.11 70	97.90 57

IS_MVSNet		91.87 68	93.35 53	90.14 97	94.09 82	97.73 56	93.09 99	88.12 90	88.71 100	79.98 109	84.49 78	90.63 64	87.49 122	97.07 19	96.96 15	98.07 73	97.88 58

PVSNet_BlendedMVS		92.80 56	92.44 64	93.23 55	96.02 61	97.83 52	93.74 86	90.58 61	91.86 60	90.69 37	85.87 71	82.04 104	90.01 91	96.39 39	95.26 53	98.34 43	97.81 59

PVSNet_Blended		92.80 56	92.44 64	93.23 55	96.02 61	97.83 52	93.74 86	90.58 61	91.86 60	90.69 37	85.87 71	82.04 104	90.01 91	96.39 39	95.26 53	98.34 43	97.81 59

IB-MVS	85.10 14	87.98 111	87.97 112	87.99 116	94.55 76	96.86 80	84.52 187	88.21 89	86.48 120	88.54 50	74.41 137	77.74 130	74.10 193	89.65 164	92.85 103	98.06 75	97.80 61
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
ACMP	89.13 9	92.03 65	91.70 76	92.41 68	94.92 74	96.44 90	93.95 78	89.96 66	91.81 62	85.48 82	90.97 43	79.12 118	92.42 70	93.28 105	92.55 108	97.76 93	97.74 62
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
CS-MVS		93.68 53	94.33 47	92.93 63	94.15 79	98.04 44	94.43 64	87.99 91	91.64 64	87.54 57	88.22 55	92.09 52	94.56 40	96.77 28	95.85 44	98.88 6	97.71 63

Anonymous202405211			88.00 110		93.16 101	96.38 91	93.58 90	89.34 76	87.92 107		65.04 180	83.03 94	92.07 73	92.67 110	93.33 89	96.96 134	97.63 64

OpenMVS	88.18 11	92.51 60	91.61 77	93.55 53	97.74 39	98.02 46	95.66 54	90.46 63	89.14 96	86.50 67	75.80 131	90.38 68	92.69 67	94.99 61	95.30 52	98.27 53	97.63 64

test_part1		87.53 115	84.97 144	90.52 88	92.11 116	93.31 138	93.32 96	85.79 116	79.56 168	87.38 60	62.89 188	78.60 121	89.25 102	90.65 147	92.17 115	95.24 177	97.62 66

UniMVSNet_NR-MVSNet		86.80 121	85.86 138	87.89 119	88.17 159	94.07 118	90.15 133	88.51 85	84.20 139	73.45 132	72.38 148	70.30 155	88.95 109	90.25 152	92.21 113	98.12 68	97.62 66

DU-MVS		86.12 129	84.81 147	87.66 120	87.77 166	93.78 123	90.15 133	87.87 97	84.40 133	73.45 132	70.59 152	64.82 182	88.95 109	90.14 153	92.33 110	97.76 93	97.62 66

EIA-MVS		92.72 58	92.96 57	92.44 67	93.86 92	97.76 54	93.13 98	88.65 84	89.78 90	86.68 65	86.69 62	87.57 71	93.74 55	96.07 47	95.32 51	98.58 18	97.53 69

PCF-MVS	90.19 8	92.98 55	92.07 70	94.04 44	96.39 58	97.87 49	96.03 48	95.47 30	87.16 111	85.09 87	84.81 77	93.21 48	93.46 60	91.98 125	91.98 122	97.78 92	97.51 70
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
MVS_111021_LR		94.84 40	95.57 32	94.00 45	97.11 50	97.72 58	94.88 62	91.16 57	95.24 28	88.74 48	96.03 21	91.52 58	94.33 46	95.96 48	95.01 57	97.79 91	97.49 71

NR-MVSNet		85.46 139	84.54 149	86.52 133	88.33 158	93.78 123	90.45 126	87.87 97	84.40 133	71.61 140	70.59 152	62.09 191	82.79 162	91.75 127	91.75 126	98.10 71	97.44 72

MAR-MVS		92.71 59	92.63 60	92.79 65	97.70 40	97.15 70	93.75 85	87.98 93	90.71 70	85.76 75	86.28 66	86.38 76	94.35 45	94.95 62	95.49 49	97.22 117	97.44 72
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
abl_6				94.78 38	97.46 43	97.99 47	95.76 52	91.80 50	93.72 46	91.25 31	91.33 41	96.47 29	94.28 47			98.14 66	97.39 74

tttt0517		91.01 81	91.71 75	90.19 95	92.98 103	97.07 74	91.96 117	87.63 104	90.61 74	81.42 98	86.76 61	82.26 103	89.23 103	94.86 68	93.03 101	97.90 86	97.36 75

Vis-MVSNet		89.36 102	91.49 79	86.88 128	92.10 117	97.60 60	92.16 112	85.89 114	84.21 138	75.20 125	82.58 94	87.13 72	77.40 183	95.90 50	95.63 47	98.51 21	97.36 75
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
thisisatest0530		91.04 80	91.74 74	90.21 93	92.93 107	97.00 75	92.06 114	87.63 104	90.74 69	81.51 97	86.81 60	82.48 99	89.23 103	94.81 70	93.03 101	97.90 86	97.33 77

DI_MVS_plusplus_trai		91.05 79	90.15 91	92.11 71	92.67 112	96.61 82	96.03 48	88.44 86	90.25 79	85.92 72	73.73 138	84.89 86	91.92 74	94.17 85	94.07 73	97.68 101	97.31 78

UniMVSNet_ETH3D		84.57 148	81.40 182	88.28 112	89.34 147	94.38 114	90.33 127	86.50 111	74.74 192	77.52 117	59.90 195	62.04 192	88.78 114	88.82 174	92.65 106	97.22 117	97.24 79

casdiffmvs		91.72 72	91.16 82	92.38 69	93.16 101	97.15 70	93.95 78	89.49 75	91.58 66	86.03 69	80.75 106	80.95 110	93.16 62	95.25 57	95.22 55	98.50 23	97.23 80

diffmvs		91.37 75	91.09 83	91.70 75	92.71 111	96.47 87	94.03 76	88.78 80	92.74 54	85.43 84	83.63 85	80.37 112	91.76 77	93.39 102	93.78 77	97.50 109	97.23 80

Effi-MVS+		89.79 96	89.83 95	89.74 98	92.98 103	96.45 89	93.48 93	84.24 132	87.62 109	76.45 121	81.76 100	77.56 132	93.48 59	94.61 75	93.59 81	97.82 90	97.22 82

CP-MVSNet		83.11 172	82.15 172	84.23 157	87.20 176	92.70 157	86.42 179	83.53 143	77.83 177	67.67 170	66.89 169	60.53 198	82.47 163	89.23 169	90.65 145	98.08 72	97.20 83

TranMVSNet+NR-MVSNet		85.57 137	84.41 150	86.92 127	87.67 169	93.34 136	90.31 129	88.43 87	83.07 148	70.11 153	69.99 158	65.28 177	86.96 127	89.73 161	92.27 111	98.06 75	97.17 84

Baseline_NR-MVSNet		85.28 141	83.42 159	87.46 124	87.77 166	90.80 191	89.90 143	87.69 101	83.93 143	74.16 128	64.72 182	66.43 172	87.48 123	90.14 153	90.83 137	97.73 96	97.11 85

PS-CasMVS		82.53 177	81.54 180	83.68 164	87.08 181	92.54 163	86.20 181	83.46 144	76.46 185	65.73 182	65.71 176	59.41 203	81.61 171	89.06 171	90.55 147	98.03 77	97.07 86

baseline		91.19 77	91.89 73	90.38 89	92.76 109	95.04 104	93.55 91	84.54 130	92.92 50	85.71 76	86.68 63	86.96 73	89.28 101	92.00 124	92.62 107	96.46 152	96.99 87

WR-MVS		83.14 170	83.38 161	82.87 176	87.55 170	93.29 139	86.36 180	84.21 133	80.05 164	66.41 177	66.91 167	66.92 170	75.66 189	88.96 172	90.56 146	97.05 128	96.96 88

CHOSEN 1792x2688		88.57 107	87.82 114	89.44 101	95.46 69	96.89 79	93.74 86	85.87 115	89.63 91	77.42 118	61.38 191	83.31 92	88.80 113	93.44 101	93.16 95	95.37 175	96.95 89

tfpnnormal		83.80 162	81.26 184	86.77 130	89.60 144	93.26 142	89.72 146	87.60 106	72.78 194	70.44 150	60.53 194	61.15 195	85.55 141	92.72 109	91.44 131	97.71 97	96.92 90

WR-MVS_H		82.86 175	82.66 169	83.10 172	87.44 172	93.33 137	85.71 185	83.20 147	77.36 179	68.20 167	66.37 170	65.23 178	76.05 188	89.35 165	90.13 155	97.99 81	96.89 91

v7n		82.25 180	81.54 180	83.07 173	85.55 193	92.58 161	86.68 178	81.10 169	76.54 183	65.97 180	62.91 187	60.56 197	82.36 164	91.07 139	90.35 150	96.77 148	96.80 92

MVS_Test		91.81 70	92.19 68	91.37 83	93.24 99	96.95 77	94.43 64	86.25 112	91.45 67	83.45 91	86.31 64	85.15 84	92.93 65	93.99 88	94.71 62	97.92 85	96.77 93

thres600view7		89.28 104	87.47 123	91.39 81	94.12 81	97.25 67	93.94 80	89.74 69	85.62 127	80.63 106	75.24 135	69.33 158	91.66 79	94.92 64	93.23 92	98.27 53	96.72 94

AdaColmap		95.02 38	93.71 49	96.54 23	98.51 27	97.76 54	96.69 40	95.94 20	93.72 46	93.50 17	89.01 52	90.53 65	96.49 21	94.51 79	93.76 78	98.07 73	96.69 95

LTVRE_ROB	81.71 16	82.44 179	81.84 177	83.13 170	89.01 148	92.99 149	88.90 158	82.32 156	66.26 205	54.02 206	74.68 136	59.62 202	88.87 112	90.71 145	92.02 120	95.68 167	96.62 96
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
tfpn200view9		89.55 99	87.86 113	91.53 78	93.90 90	97.26 66	94.31 71	89.74 69	85.87 122	81.15 100	76.46 126	70.38 153	91.76 77	94.92 64	93.51 82	98.28 52	96.61 97

thres400		89.40 101	87.58 120	91.53 78	94.06 84	97.21 69	94.19 75	89.83 68	85.69 124	81.08 102	75.50 133	69.76 157	91.80 75	94.79 71	93.51 82	98.20 61	96.60 98

ACMH	85.51 13	87.31 118	86.59 127	88.14 114	93.96 86	94.51 108	89.00 157	87.99 91	81.58 154	70.15 152	78.41 115	71.78 149	90.60 88	91.30 134	91.99 121	97.17 120	96.58 99
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
CNLPA		93.69 51	92.50 62	95.06 37	97.11 50	97.36 64	93.88 82	93.30 39	95.64 23	93.44 19	80.32 107	90.73 63	94.99 37	93.58 95	93.33 89	97.67 102	96.57 100

thisisatest0515		85.70 134	87.00 124	84.19 158	88.16 160	93.67 128	84.20 189	84.14 135	83.39 147	72.91 134	76.79 123	74.75 142	78.82 181	92.57 114	91.26 134	96.94 136	96.56 101

Vis-MVSNet (Re-imp)		90.54 87	92.76 59	87.94 117	93.73 95	96.94 78	92.17 111	87.91 94	88.77 99	76.12 123	83.68 84	90.80 60	79.49 179	96.34 41	96.35 28	98.21 60	96.46 102

Fast-Effi-MVS+		88.56 108	87.99 111	89.22 103	91.56 124	95.21 101	92.29 107	82.69 149	86.82 113	77.73 116	76.24 129	73.39 143	93.36 61	94.22 84	93.64 79	97.65 103	96.43 103

thres200		89.49 100	87.72 115	91.55 77	93.95 87	97.25 67	94.34 69	89.74 69	85.66 125	81.18 99	76.12 130	70.19 156	91.80 75	94.92 64	93.51 82	98.27 53	96.40 104

MVSTER		91.73 71	91.61 77	91.86 73	93.18 100	94.56 106	94.37 67	87.90 95	90.16 84	88.69 49	89.23 50	81.28 109	88.92 111	95.75 52	93.95 75	98.12 68	96.37 105

v144192		83.48 166	82.23 171	84.94 147	86.65 185	92.84 153	89.63 148	82.48 153	77.87 176	67.36 172	65.33 178	63.50 185	86.51 131	89.72 162	89.99 163	97.03 129	96.35 106

IterMVS-LS		88.60 106	88.45 103	88.78 107	92.02 118	92.44 166	92.00 116	83.57 142	86.52 118	78.90 114	78.61 114	81.34 108	89.12 106	90.68 146	93.18 94	97.10 125	96.35 106
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
v1192		83.56 165	82.35 170	84.98 146	86.84 184	92.84 153	90.01 138	82.70 148	78.54 172	66.48 176	64.88 181	62.91 186	86.91 128	90.72 144	90.25 153	96.94 136	96.32 108

v1921920		83.30 168	82.09 174	84.70 150	86.59 187	92.67 159	89.82 144	82.23 157	78.32 173	65.76 181	64.64 183	62.35 189	86.78 130	90.34 151	90.02 161	97.02 130	96.31 109

v10		84.18 155	83.17 165	85.37 141	87.34 173	92.68 158	90.32 128	81.33 165	79.93 167	69.23 160	66.33 171	65.74 175	87.03 126	90.84 141	90.38 149	96.97 132	96.29 110

V42		84.48 152	83.36 162	85.79 138	87.14 178	93.28 140	90.03 136	83.98 137	80.30 161	71.20 145	66.90 168	67.17 167	85.55 141	89.35 165	90.27 152	96.82 146	96.27 111

PEN-MVS		82.49 178	81.58 179	83.56 166	86.93 182	92.05 174	86.71 177	83.84 138	76.94 182	64.68 186	67.24 164	60.11 199	81.17 173	87.78 178	90.70 144	98.02 78	96.21 112

thres100view900		89.36 102	87.61 118	91.39 81	93.90 90	96.86 80	94.35 68	89.66 73	85.87 122	81.15 100	76.46 126	70.38 153	91.17 81	94.09 86	93.43 88	98.13 67	96.16 113

v1144		84.03 159	82.88 167	85.37 141	87.17 177	93.15 147	90.18 132	83.31 145	78.83 171	67.85 168	65.99 173	64.99 180	86.79 129	90.75 143	90.33 151	96.90 141	96.15 114

OMC-MVS		94.49 44	94.36 44	94.64 41	97.17 49	97.73 56	95.49 56	92.25 45	96.18 14	90.34 40	88.51 53	92.88 50	94.90 38	94.92 64	94.17 68	97.69 100	96.15 114

Effi-MVS+-dtu		87.51 116	88.13 109	86.77 130	91.10 130	94.90 105	90.91 122	82.67 150	83.47 145	71.55 141	81.11 105	77.04 134	89.41 97	92.65 112	91.68 129	95.00 182	96.09 116

v8		84.45 154	83.30 163	85.80 137	87.53 171	92.95 150	90.31 129	82.46 154	80.46 159	71.43 142	66.99 166	67.16 168	86.14 137	89.26 168	90.22 154	96.94 136	96.06 117

v1240		82.88 174	81.66 178	84.29 156	86.46 188	92.52 165	89.06 155	81.82 161	77.16 180	65.09 185	64.17 185	61.50 193	86.36 132	90.12 155	90.13 155	96.95 135	96.04 118

DPM-MVS		95.07 36	94.84 38	95.34 35	97.44 44	97.49 62	97.76 26	95.52 25	94.88 35	88.92 46	87.25 58	96.44 30	94.41 42	95.78 51	96.11 39	97.99 81	95.95 119

CDS-MVSNet		88.34 109	88.71 102	87.90 118	90.70 137	94.54 107	92.38 104	86.02 113	80.37 160	79.42 111	79.30 110	83.43 91	82.04 166	93.39 102	94.01 74	96.86 145	95.93 120
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
Fast-Effi-MVS+-dtu		86.25 125	87.70 116	84.56 153	90.37 140	93.70 126	90.54 125	78.14 179	83.50 144	65.37 184	81.59 103	75.83 141	86.09 139	91.70 128	91.70 127	96.88 143	95.84 121

ET-MVSNet_ETH3D		89.93 93	90.84 85	88.87 106	79.60 204	96.19 93	94.43 64	86.56 110	90.63 72	80.75 105	90.71 44	77.78 129	93.73 56	91.36 133	93.45 87	98.15 64	95.77 122

OPM-MVS		91.08 78	89.34 97	93.11 60	96.18 60	96.13 95	96.39 43	92.39 44	82.97 149	81.74 96	82.55 96	80.20 114	93.97 52	94.62 74	93.23 92	98.00 80	95.73 123

baseline2		88.97 105	89.50 96	88.36 110	91.14 129	95.30 100	90.13 135	85.17 124	87.24 110	80.80 104	84.46 79	78.44 123	85.60 140	93.54 98	91.87 123	97.31 114	95.66 124

CANet_DTU		90.74 85	92.93 58	88.19 113	94.36 77	96.61 82	94.34 69	84.66 127	90.66 71	68.75 162	90.41 46	86.89 74	89.78 93	95.46 56	94.87 59	97.25 116	95.62 125

v2v482		84.51 150	83.05 166	86.20 135	87.25 175	93.28 140	90.22 131	85.40 122	79.94 166	69.78 155	67.74 163	65.15 179	87.57 120	89.12 170	90.55 147	96.97 132	95.60 126

TAPA-MVS	90.35 6	93.69 51	93.52 50	93.90 48	96.89 53	97.62 59	96.15 45	91.67 52	94.94 33	85.97 70	87.72 57	91.96 53	94.40 43	93.76 93	93.06 99	98.30 49	95.58 127
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
ACMM	88.76 10	91.70 73	90.43 87	93.19 57	95.56 66	95.14 103	93.35 95	91.48 54	92.26 58	87.12 61	84.02 81	79.34 117	93.99 50	94.07 87	92.68 105	97.62 106	95.50 128
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
GBi-Net		90.21 90	90.11 92	90.32 91	88.66 153	93.65 129	94.25 72	85.78 117	90.03 85	85.56 79	77.38 117	86.13 77	89.38 98	93.97 89	94.16 69	98.31 46	95.47 129

test1		90.21 90	90.11 92	90.32 91	88.66 153	93.65 129	94.25 72	85.78 117	90.03 85	85.56 79	77.38 117	86.13 77	89.38 98	93.97 89	94.16 69	98.31 46	95.47 129

FMVSNet2		89.61 98	89.14 99	90.16 96	88.66 153	93.65 129	94.25 72	85.44 121	88.57 102	84.96 88	73.53 140	83.82 89	89.38 98	94.23 83	94.68 63	98.31 46	95.47 129

pm-mvs1		84.55 149	83.46 156	85.82 136	88.16 160	93.39 135	89.05 156	85.36 123	74.03 193	72.43 138	65.08 179	71.11 150	82.30 165	93.48 99	91.70 127	97.64 104	95.43 132

FMVSNet3		90.19 92	90.06 94	90.34 90	88.69 152	93.85 121	94.58 63	85.78 117	90.03 85	85.56 79	77.38 117	86.13 77	89.22 105	93.29 104	94.36 66	98.20 61	95.40 133

FMVSNet1		87.33 117	86.00 135	88.89 105	87.13 179	92.83 155	93.08 100	84.46 131	81.35 156	82.20 95	66.33 171	77.96 127	88.96 108	93.97 89	94.16 69	97.54 108	95.38 134

v148		83.61 164	82.10 173	85.37 141	87.34 173	92.94 151	87.48 169	85.72 120	78.92 170	73.87 130	65.71 176	64.69 183	81.78 170	87.82 177	89.35 172	96.01 159	95.26 135

HyFIR lowres test		87.87 112	86.42 129	89.57 99	95.56 66	96.99 76	92.37 105	84.15 134	86.64 115	77.17 119	57.65 197	83.97 88	91.08 83	92.09 123	92.44 109	97.09 126	95.16 136

DTE-MVSNet		81.76 183	81.04 185	82.60 180	86.63 186	91.48 185	85.97 183	83.70 139	76.45 186	62.44 191	67.16 165	59.98 200	78.98 180	87.15 182	89.93 164	97.88 88	95.12 137

ACMH+	85.75 12	87.19 119	86.02 134	88.56 109	93.42 98	94.41 112	89.91 141	87.66 103	83.45 146	72.25 139	76.42 128	71.99 148	90.78 85	89.86 159	90.94 136	97.32 113	95.11 138

PLC	90.69 4	94.32 45	92.99 56	95.87 29	97.91 34	96.49 86	95.95 51	94.12 36	94.94 33	94.09 13	85.90 69	90.77 62	95.58 32	94.52 78	93.32 91	97.55 107	95.00 139
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
IterMVS-SCA-FT		85.44 140	86.71 125	83.97 162	90.59 138	90.84 189	89.73 145	78.34 178	84.07 142	66.40 178	77.27 122	78.66 120	83.06 159	91.20 135	90.10 160	95.72 165	94.78 140

pmmvs6		80.90 185	78.77 190	83.38 169	85.84 190	91.61 181	86.01 182	82.54 152	64.17 206	70.43 151	54.14 204	67.06 169	80.73 175	90.50 150	89.17 174	94.74 183	94.75 141

GA-MVS		85.08 143	85.65 140	84.42 155	89.77 142	94.25 115	89.26 151	84.62 128	81.19 157	62.25 192	75.72 132	68.44 162	84.14 154	93.57 96	91.68 129	96.49 150	94.71 142

TSAR-MVS + COLMAP		92.39 62	92.31 67	92.47 66	95.35 73	96.46 88	96.13 46	92.04 48	95.33 27	80.11 108	94.95 29	77.35 133	94.05 49	94.49 80	93.08 97	97.15 121	94.53 143

gg-mvs-nofinetune		81.83 182	83.58 155	79.80 189	91.57 123	96.54 85	93.79 84	68.80 205	62.71 208	43.01 213	55.28 200	85.06 85	83.65 157	96.13 45	94.86 60	97.98 84	94.46 144

LS3D		91.97 66	90.98 84	93.12 59	97.03 52	97.09 73	95.33 58	95.59 23	92.47 56	79.26 112	81.60 102	82.77 97	94.39 44	94.28 81	94.23 67	97.14 123	94.45 145

UA-Net		90.81 82	92.58 61	88.74 108	94.87 75	97.44 63	92.61 102	88.22 88	82.35 152	78.93 113	85.20 75	95.61 38	79.56 178	96.52 34	96.57 23	98.23 58	94.37 146

pmmvs5		83.37 167	82.68 168	84.18 159	87.13 179	93.18 144	86.74 176	82.08 158	76.48 184	67.28 173	71.26 149	62.70 188	84.71 148	90.77 142	90.12 158	97.15 121	94.24 147

IterMVS		85.25 142	86.49 128	83.80 163	90.42 139	90.77 192	90.02 137	78.04 180	84.10 140	66.27 179	77.28 121	78.41 124	83.01 160	90.88 140	89.72 169	95.04 180	94.24 147
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
test-mter		86.09 131	88.38 104	83.43 168	87.89 163	92.61 160	86.89 175	77.11 184	84.30 136	68.62 164	82.57 95	82.45 100	84.34 150	92.40 116	90.11 159	95.74 163	94.21 149

SixPastTwentyTwo		83.12 171	83.44 158	82.74 177	87.71 168	93.11 148	82.30 194	82.33 155	79.24 169	64.33 187	78.77 113	62.75 187	84.11 155	88.11 176	87.89 178	95.70 166	94.21 149

CR-MVSNet		85.48 138	86.29 130	84.53 154	91.08 132	92.10 170	89.18 152	73.30 196	84.75 129	71.08 146	73.12 146	77.91 128	86.27 135	91.48 130	90.75 141	96.27 156	93.94 151

PatchT		83.86 160	85.51 142	81.94 184	88.41 156	91.56 182	78.79 200	71.57 200	84.08 141	71.08 146	70.62 151	76.13 140	86.27 135	91.48 130	90.75 141	95.52 173	93.94 151

FC-MVSNet-test		86.15 128	89.10 100	82.71 178	89.83 141	93.18 144	87.88 167	84.69 126	86.54 117	62.18 193	82.39 97	83.31 92	74.18 192	92.52 115	91.86 124	97.50 109	93.88 153

baseline1		90.81 82	90.29 88	91.42 80	93.67 96	95.86 98	93.94 80	89.69 72	89.29 95	82.85 94	82.91 89	80.30 113	89.60 94	95.05 60	94.79 61	98.80 7	93.82 154

RPMNet		84.82 147	85.90 137	83.56 166	91.10 130	92.10 170	88.73 161	71.11 201	84.75 129	68.79 161	73.56 139	77.62 131	85.33 144	90.08 157	89.43 171	96.32 155	93.77 155

CHOSEN 280x420		90.77 84	92.14 69	89.17 104	93.86 92	92.81 156	93.16 97	80.22 172	90.21 81	84.67 89	89.89 48	91.38 59	90.57 89	94.94 63	92.11 117	92.52 191	93.65 156

test-LLR		86.88 120	88.28 105	85.24 144	91.22 127	92.07 172	87.41 170	83.62 140	84.58 131	69.33 158	83.00 87	82.79 95	84.24 151	92.26 118	89.81 165	95.64 168	93.44 157

TESTMET0.1,1		86.11 130	88.28 105	83.59 165	87.80 164	92.07 172	87.41 170	77.12 183	84.58 131	69.33 158	83.00 87	82.79 95	84.24 151	92.26 118	89.81 165	95.64 168	93.44 157

CostFormer		86.78 122	86.05 132	87.62 123	92.15 115	93.20 143	91.55 120	75.83 186	88.11 106	85.29 85	81.76 100	76.22 139	87.80 117	84.45 193	85.21 189	93.12 187	93.42 159

PM-MVS		80.29 187	79.30 189	81.45 187	81.91 201	88.23 197	82.61 192	79.01 176	79.99 165	67.15 174	69.07 159	51.39 208	82.92 161	87.55 180	85.59 185	95.08 179	93.28 160

tpm		83.16 169	83.64 154	82.60 180	90.75 134	91.05 186	88.49 162	73.99 191	82.36 151	67.08 175	78.10 116	68.79 159	84.17 153	85.95 189	85.96 184	91.09 198	93.23 161

pmmvs4		86.00 132	84.28 151	88.00 115	87.80 164	92.01 175	89.94 140	84.91 125	86.79 114	80.98 103	73.41 143	66.34 173	88.12 116	89.31 167	88.90 176	96.24 157	93.20 162

PMMVS		89.88 94	91.19 81	88.35 111	89.73 143	91.97 176	90.62 124	81.92 159	90.57 75	80.58 107	92.16 35	86.85 75	91.17 81	92.31 117	91.35 133	96.11 158	93.11 163

PatchMatch-RL		90.30 89	88.93 101	91.89 72	95.41 72	95.68 99	90.94 121	88.67 83	89.80 89	86.95 64	85.90 69	72.51 144	92.46 69	93.56 97	92.18 114	96.93 139	92.89 164

EPNet_dtu		88.32 110	90.61 86	85.64 140	96.79 55	92.27 168	92.03 115	90.31 64	89.05 97	65.44 183	89.43 49	85.90 81	74.22 191	92.76 108	92.09 118	95.02 181	92.76 165
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
CVMVSNet		83.83 161	85.53 141	81.85 185	89.60 144	90.92 187	87.81 168	83.21 146	80.11 163	60.16 196	76.47 125	78.57 122	76.79 185	89.76 160	90.13 155	93.51 184	92.75 166

TransMVSNet (Re)		82.67 176	80.93 187	84.69 151	88.71 151	91.50 183	87.90 166	87.15 107	71.54 199	68.24 166	63.69 186	64.67 184	78.51 182	91.65 129	90.73 143	97.64 104	92.73 167

EG-PatchMatch MVS		81.70 184	81.31 183	82.15 183	88.75 150	93.81 122	87.14 173	78.89 177	71.57 197	64.12 189	61.20 193	68.46 161	76.73 186	91.48 130	90.77 140	97.28 115	91.90 168

TDRefinement		84.97 145	83.39 160	86.81 129	92.97 105	94.12 116	92.18 109	87.77 100	82.78 150	71.31 144	68.43 161	68.07 164	81.10 174	89.70 163	89.03 175	95.55 172	91.62 169

pmmvs-eth3d		79.78 189	77.58 194	82.34 182	81.57 202	87.46 200	82.92 191	81.28 166	75.33 191	71.34 143	61.88 189	52.41 207	81.59 172	87.56 179	86.90 181	95.36 176	91.48 170

gm-plane-assit		77.65 193	78.50 191	76.66 194	87.96 162	85.43 204	64.70 210	74.50 189	64.15 207	51.26 208	61.32 192	58.17 204	84.11 155	95.16 59	93.83 76	97.45 111	91.41 171

EU-MVSNet		78.43 190	80.25 188	76.30 195	83.81 198	87.27 202	80.99 196	79.52 174	76.01 187	54.12 205	70.44 154	64.87 181	67.40 199	86.23 187	85.54 187	91.95 196	91.41 171

MSDG		90.42 88	88.25 107	92.94 62	96.67 56	94.41 112	93.96 77	92.91 42	89.59 92	86.26 68	76.74 124	80.92 111	90.43 90	92.60 113	92.08 119	97.44 112	91.41 171

test0.0.03 1		85.58 136	87.69 117	83.11 171	91.22 127	92.54 163	85.60 186	83.62 140	85.66 125	67.84 169	82.79 92	79.70 116	73.51 195	91.15 138	90.79 138	96.88 143	91.23 174

GG-mvs-BLEND		62.84 202	90.21 89	30.91 210	0.57 218	94.45 110	86.99 174	0.34 216	88.71 100	0.98 218	81.55 104	91.58 57	0.86 215	92.66 111	91.43 132	95.73 164	91.11 175

TAMVS		84.94 146	84.95 145	84.93 148	88.82 149	93.18 144	88.44 163	81.28 166	77.16 180	73.76 131	75.43 134	76.57 138	82.04 166	90.59 148	90.79 138	95.22 178	90.94 176

MS-PatchMatch		87.63 113	87.61 118	87.65 121	93.95 87	94.09 117	92.60 103	81.52 164	86.64 115	76.41 122	73.46 142	85.94 80	85.01 147	92.23 121	90.00 162	96.43 154	90.93 177

tpm cat1		84.13 156	81.99 176	86.63 132	91.74 120	91.50 183	90.68 123	75.69 187	86.12 121	85.44 83	72.39 147	70.72 151	85.16 145	80.89 201	81.56 198	91.07 199	90.71 178

ambc			67.96 203		73.69 207	79.79 208	73.82 205		71.61 196	59.80 197	46.00 206	20.79 216	66.15 201	86.92 184	80.11 202	89.13 206	90.50 179

COLMAP_ROB	84.39 15	87.61 114	86.03 133	89.46 100	95.54 68	94.48 109	91.77 119	90.14 65	87.16 111	75.50 124	73.41 143	76.86 137	87.33 124	90.05 158	89.76 168	96.48 151	90.46 180
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
RPSCF		89.68 97	89.24 98	90.20 94	92.97 105	92.93 152	92.30 106	87.69 101	90.44 77	85.12 86	91.68 39	85.84 82	90.69 87	87.34 181	86.07 183	92.46 192	90.37 181

Anonymous20231206		78.09 192	78.11 193	78.07 193	85.19 195	89.17 194	80.99 196	81.24 168	75.46 190	58.25 200	54.78 203	59.90 201	66.73 200	88.94 173	88.26 177	96.01 159	90.25 182

dps		85.00 144	83.21 164	87.08 126	90.73 135	92.55 162	89.34 149	75.29 188	84.94 128	87.01 62	79.27 111	67.69 166	87.27 125	84.22 194	83.56 194	92.83 189	90.25 182

USDC		86.73 123	85.96 136	87.63 122	91.64 121	93.97 119	92.76 101	84.58 129	88.19 104	70.67 149	80.10 108	67.86 165	89.43 96	91.81 126	89.77 167	96.69 149	90.05 184

tpmrst		83.72 163	83.45 157	84.03 161	92.21 114	91.66 180	88.74 160	73.58 195	88.14 105	72.67 136	77.37 120	72.11 147	86.34 133	82.94 198	82.05 197	90.63 201	89.86 185

SCA		86.25 125	87.52 121	84.77 149	91.59 122	93.90 120	89.11 154	73.25 198	90.38 78	72.84 135	83.26 86	83.79 90	88.49 115	86.07 188	85.56 186	93.33 185	89.67 186

testgi		81.94 181	84.09 152	79.43 190	89.53 146	90.83 190	82.49 193	81.75 162	80.59 158	59.46 198	82.82 91	65.75 174	67.97 197	90.10 156	89.52 170	95.39 174	89.03 187

PatchmatchNet		85.70 134	86.65 126	84.60 152	91.79 119	93.40 134	89.27 150	73.62 193	90.19 82	72.63 137	82.74 93	81.93 106	87.64 119	84.99 191	84.29 193	92.64 190	89.00 188
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
MDTV_nov1_ep13		86.64 124	87.50 122	85.65 139	90.73 135	93.69 127	89.96 139	78.03 181	89.48 94	76.85 120	84.92 76	82.42 101	86.14 137	86.85 185	86.15 182	92.17 193	88.97 189

MIMVSNet		82.97 173	84.00 153	81.77 186	82.23 200	92.25 169	87.40 172	72.73 199	81.48 155	69.55 156	68.79 160	72.42 145	81.82 169	92.23 121	92.25 112	96.89 142	88.61 190

TinyColmap		84.04 158	82.01 175	86.42 134	90.87 133	91.84 177	88.89 159	84.07 136	82.11 153	69.89 154	71.08 150	60.81 196	89.04 107	90.52 149	89.19 173	95.76 162	88.50 191

ADS-MVSNet		84.08 157	84.95 145	83.05 174	91.53 126	91.75 179	88.16 164	70.70 202	89.96 88	69.51 157	78.83 112	76.97 136	86.29 134	84.08 195	84.60 191	92.13 195	88.48 192

EPMVS		85.77 133	86.24 131	85.23 145	92.76 109	93.78 123	89.91 141	73.60 194	90.19 82	74.22 127	82.18 98	78.06 126	87.55 121	85.61 190	85.38 188	93.32 186	88.48 192

MDTV_nov1_ep13_2view		80.43 186	80.94 186	79.84 188	84.82 196	90.87 188	84.23 188	73.80 192	80.28 162	64.33 187	70.05 157	68.77 160	79.67 176	84.83 192	83.50 195	92.17 193	88.25 194

MDA-MVSNet-bldmvs		73.81 196	72.56 200	75.28 196	72.52 209	88.87 195	74.95 204	82.67 150	71.57 197	55.02 203	65.96 174	42.84 214	76.11 187	70.61 206	81.47 199	90.38 203	86.59 195

test20.03		76.41 195	78.49 192	73.98 197	85.64 192	87.50 199	75.89 202	80.71 170	70.84 200	51.07 209	68.06 162	61.40 194	54.99 206	88.28 175	87.20 180	95.58 171	86.15 196

FMVSNet5		84.47 153	84.72 148	84.18 159	83.30 199	88.43 196	88.09 165	79.42 175	84.25 137	74.14 129	73.15 145	78.74 119	83.65 157	91.19 136	91.19 135	96.46 152	86.07 197

pmmvs3		71.13 200	71.06 202	71.21 201	73.54 208	80.19 207	71.69 208	64.86 207	62.04 209	52.10 207	54.92 202	48.00 212	75.03 190	83.75 197	83.24 196	90.04 204	85.27 198

MIMVSNet1		73.19 197	73.70 198	72.60 200	65.42 212	86.69 203	75.56 203	79.65 173	67.87 204	55.30 202	45.24 207	56.41 205	63.79 202	86.98 183	87.66 179	95.85 161	85.04 199

CMPMVS	61.19 17	79.86 188	77.46 196	82.66 179	91.54 125	91.82 178	83.25 190	81.57 163	70.51 201	68.64 163	59.89 196	66.77 171	79.63 177	84.00 196	84.30 192	91.34 197	84.89 200
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
N_pmnet		77.55 194	76.68 197	78.56 192	85.43 194	87.30 201	78.84 199	81.88 160	78.30 174	60.61 194	61.46 190	62.15 190	74.03 194	82.04 199	80.69 201	90.59 202	84.81 201

new-patchmatchnet		72.32 198	71.09 201	73.74 198	81.17 203	84.86 205	72.21 207	77.48 182	68.32 203	54.89 204	55.10 201	49.31 211	63.68 203	79.30 202	76.46 204	93.03 188	84.32 202

MVS-HIRNet		78.16 191	77.57 195	78.83 191	85.83 191	87.76 198	76.67 201	70.22 203	75.82 189	67.39 171	55.61 199	70.52 152	81.96 168	86.67 186	85.06 190	90.93 200	81.58 203

new_pmnet		72.29 199	73.25 199	71.16 202	75.35 206	81.38 206	73.72 206	69.27 204	75.97 188	49.84 210	56.27 198	56.12 206	69.08 196	81.73 200	80.86 200	89.72 205	80.44 204

DeepMVS_CX						71.82 210	68.37 209	48.05 212	77.38 178	46.88 211	65.77 175	47.03 213	67.48 198	64.27 209		76.89 211	76.72 205

FPMVS		69.87 201	67.10 204	73.10 199	84.09 197	78.35 209	79.40 198	76.41 185	71.92 195	57.71 201	54.06 205	50.04 209	56.72 204	71.19 205	68.70 206	84.25 207	75.43 206

PMVS	56.77 18	61.27 203	58.64 205	64.35 203	75.66 205	54.60 213	53.62 212	74.23 190	53.69 210	58.37 199	44.27 208	49.38 210	44.16 209	69.51 207	65.35 208	80.07 209	73.66 207
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
PMMVS2		53.68 205	55.72 207	51.30 205	58.84 213	67.02 211	54.23 211	60.97 210	47.50 211	19.42 215	34.81 209	31.97 215	30.88 211	65.84 208	69.99 205	83.47 208	72.92 208

Gipuma		58.52 204	56.17 206	61.27 204	67.14 211	58.06 212	52.16 213	68.40 206	69.00 202	45.02 212	22.79 210	20.57 217	55.11 205	76.27 203	79.33 203	79.80 210	67.16 209
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
MVE	39.81 19	39.52 207	41.58 208	37.11 209	33.93 216	49.06 214	26.45 217	54.22 211	29.46 214	24.15 214	20.77 212	10.60 220	34.42 210	51.12 210	65.27 209	49.49 215	64.81 210
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
E-PMN		40.00 206	35.74 209	44.98 207	57.69 215	39.15 217	28.05 215	62.70 208	35.52 213	17.78 216	20.90 211	14.36 219	44.47 208	35.89 211	47.86 210	59.15 213	56.47 211

EMVS		39.04 208	34.32 210	44.54 208	58.25 214	39.35 216	27.61 216	62.55 209	35.99 212	16.40 217	20.04 213	14.77 218	44.80 207	33.12 212	44.10 211	57.61 214	52.89 212

test123		3.48 210	5.31 212	1.34 212	0.20 219	1.52 219	2.17 220	0.58 215	6.13 216	0.31 220	9.85 215	0.31 222	3.90 213	2.65 214	5.28 213	0.87 217	11.46 213

testmvs		4.35 209	6.54 211	1.79 211	0.60 217	1.82 218	3.06 219	0.95 214	7.22 215	0.88 219	12.38 214	1.25 221	3.87 214	6.09 213	5.58 212	1.40 216	11.42 214

uanet_test		0.00 211	0.00 213	0.00 213	0.00 220	0.00 220	0.00 221	0.00 217	0.00 217	0.00 221	0.00 216	0.00 223	0.00 216	0.00 215	0.00 214	0.00 218	0.00 215

sosnet-low-res		0.00 211	0.00 213	0.00 213	0.00 220	0.00 220	0.00 221	0.00 217	0.00 217	0.00 221	0.00 216	0.00 223	0.00 216	0.00 215	0.00 214	0.00 218	0.00 215

sosnet		0.00 211	0.00 213	0.00 213	0.00 220	0.00 220	0.00 221	0.00 217	0.00 217	0.00 221	0.00 216	0.00 223	0.00 216	0.00 215	0.00 214	0.00 218	0.00 215

RE-MVS-def										60.19 195

9.14												97.28 23

SR-MVS					98.93 19			96.00 17				97.75 14

our_test_3					86.93 182	89.77 193	81.61 195

MTAPA										95.36 2		97.46 20

MTMP										95.70 1		96.90 26

Patchmatch-RL test							18.47 218

tmp_tt				50.24 206	68.55 210	46.86 215	48.90 214	18.28 213	86.51 119	68.32 165	70.19 156	65.33 176	26.69 212	74.37 204	66.80 207	70.72 212

XVS					95.68 63	98.66 13	94.96 60			88.03 51		96.06 32				98.46 28

X-MVStestdata					95.68 63	98.66 13	94.96 60			88.03 51		96.06 32				98.46 28

mPP-MVS					98.76 24							95.49 39

NP-MVS									91.63 65

Patchmtry						92.39 167	89.18 152	73.30 196		71.08 146

High-resolution multi-view benchmark