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		95.56 2	96.26 2	94.73 2	96.93 16	98.19 1	96.62 6	92.81 4	96.15 1	91.73 5	95.01 7	95.31 2	93.41 1	95.95 2	94.77 7	96.90 4	98.46 1
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
SED-MVS		95.61 1	96.36 1	94.73 2	96.84 19	98.15 2	97.08 3	92.92 2	95.64 2	91.84 4	95.98 4	95.33 1	92.83 6	96.00 1	94.94 3	96.90 4	98.45 2

DPE-MVS		95.53 3	96.13 3	94.82 1	96.81 22	98.05 3	97.42 1	93.09 1	94.31 8	91.49 6	97.12 1	95.03 3	93.27 3	95.55 5	94.58 11	96.86 6	98.25 3
Kehua Chen, Zhenlong Yuan, Tianlu Mao, Zhaoqi Wang: Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization. AAAI2025
APDe-MVS		95.23 4	95.69 5	94.70 4	97.12 10	97.81 5	97.19 2	92.83 3	95.06 5	90.98 10	96.47 2	92.77 10	93.38 2	95.34 8	94.21 15	96.68 8	98.17 4

SMA-MVS		94.70 6	95.35 6	93.93 12	97.57 2	97.57 7	95.98 11	91.91 13	94.50 6	90.35 14	93.46 17	92.72 11	91.89 18	95.89 3	95.22 1	95.88 27	98.10 5
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
SD-MVS		94.53 9	95.22 7	93.73 15	95.69 36	97.03 14	95.77 22	91.95 12	94.41 7	91.35 7	94.97 8	93.34 7	91.80 20	94.72 18	93.99 19	95.82 34	98.07 6
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
MSP-MVS		95.12 5	95.83 4	94.30 5	96.82 21	97.94 4	96.98 4	92.37 11	95.40 3	90.59 13	96.16 3	93.71 5	92.70 7	94.80 15	94.77 7	96.37 14	97.99 7
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
TSAR-MVS + MP.		94.48 10	94.97 8	93.90 13	95.53 37	97.01 15	96.69 5	90.71 24	94.24 9	90.92 11	94.97 8	92.19 15	93.03 4	94.83 14	93.60 26	96.51 13	97.97 8
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
CSCG		92.76 26	93.16 27	92.29 29	96.30 28	97.74 6	94.67 33	88.98 36	92.46 23	89.73 20	86.67 37	92.15 17	88.69 42	92.26 54	92.92 42	95.40 57	97.89 9

SteuartSystems-ACMMP		94.06 13	94.65 11	93.38 19	96.97 15	97.36 8	96.12 9	91.78 14	92.05 28	87.34 30	94.42 12	90.87 24	91.87 19	95.47 7	94.59 10	96.21 19	97.77 10
Skip Steuart: Steuart Systems R&D Blog.
TSAR-MVS + ACMM		92.97 23	94.51 13	91.16 37	95.88 34	96.59 30	95.09 29	90.45 30	93.42 17	83.01 53	94.68 10	90.74 26	88.74 40	94.75 16	93.78 23	93.82 129	97.63 11

DeepPCF-MVS	88.51 2	92.64 29	94.42 16	90.56 41	94.84 44	96.92 18	91.31 61	89.61 32	95.16 4	84.55 46	89.91 29	91.45 21	90.15 33	95.12 10	94.81 6	92.90 148	97.58 12

HPM-MVS++		94.60 8	94.91 10	94.24 7	97.86 1	96.53 32	96.14 8	92.51 8	93.87 15	90.76 12	93.45 18	93.84 4	92.62 8	95.11 11	94.08 18	95.58 50	97.48 13

ACMMP_NAP		93.94 15	94.49 14	93.30 20	97.03 13	97.31 9	95.96 12	91.30 18	93.41 18	88.55 24	93.00 19	90.33 28	91.43 26	95.53 6	94.41 13	95.53 52	97.47 14

CNVR-MVS		94.37 11	94.65 11	94.04 11	97.29 6	97.11 10	96.00 10	92.43 10	93.45 16	89.85 19	90.92 25	93.04 8	92.59 9	95.77 4	94.82 5	96.11 21	97.42 15

xxxxxxxxxxxxxcwj		92.95 24	91.88 33	94.20 8	96.75 24	97.07 11	95.82 18	92.60 6	93.98 12	91.09 8	95.89 5	71.01 126	91.93 15	94.40 25	93.56 28	97.04 2	97.27 16

SF-MVS		94.61 7	94.96 9	94.20 8	96.75 24	97.07 11	95.82 18	92.60 6	93.98 12	91.09 8	95.89 5	92.54 12	91.93 15	94.40 25	93.56 28	97.04 2	97.27 16

APD-MVS		94.37 11	94.47 15	94.26 6	97.18 8	96.99 16	96.53 7	92.68 5	92.45 24	89.96 17	94.53 11	91.63 20	92.89 5	94.58 20	93.82 22	96.31 17	97.26 18
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
DeepC-MVS	87.86 3	92.26 31	91.86 34	92.73 25	96.18 29	96.87 19	95.19 28	91.76 15	92.17 27	86.58 35	81.79 50	85.85 50	90.88 29	94.57 21	94.61 9	95.80 35	97.18 19
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
MP-MVS		93.35 20	93.59 23	93.08 23	97.39 4	96.82 22	95.38 25	90.71 24	90.82 36	88.07 27	92.83 21	90.29 29	91.32 27	94.03 29	93.19 38	95.61 48	97.16 20
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
TSAR-MVS + GP.		92.71 28	93.91 20	91.30 35	91.96 72	96.00 41	93.43 41	87.94 41	92.53 22	86.27 40	93.57 15	91.94 18	91.44 25	93.29 40	92.89 43	96.78 7	97.15 21

MCST-MVS		93.81 16	94.06 18	93.53 17	96.79 23	96.85 20	95.95 13	91.69 16	92.20 26	87.17 32	90.83 27	93.41 6	91.96 14	94.49 23	93.50 31	97.61 1	97.12 22

HFP-MVS		94.02 14	94.22 17	93.78 14	97.25 7	96.85 20	95.81 20	90.94 23	94.12 10	90.29 16	94.09 14	89.98 31	92.52 10	93.94 32	93.49 33	95.87 29	97.10 23

CP-MVS		93.25 21	93.26 26	93.24 21	96.84 19	96.51 33	95.52 24	90.61 27	92.37 25	88.88 22	90.91 26	89.52 34	91.91 17	93.64 36	92.78 44	95.69 41	97.09 24

canonicalmvs		89.36 49	89.92 44	88.70 58	91.38 76	95.92 43	91.81 57	82.61 94	90.37 40	82.73 57	82.09 48	79.28 84	88.30 47	91.17 70	93.59 27	95.36 60	97.04 25

NCCC		93.69 19	93.66 22	93.72 16	97.37 5	96.66 29	95.93 16	92.50 9	93.40 19	88.35 25	87.36 35	92.33 14	92.18 12	94.89 13	94.09 17	96.00 23	96.91 26

zzz-MVS		93.80 17	93.45 25	94.20 8	97.53 3	96.43 36	95.88 17	91.12 20	94.09 11	92.74 3	87.68 33	90.77 25	92.04 13	94.74 17	93.56 28	95.91 26	96.85 27

3Dnovator+	86.06 4	91.60 36	90.86 42	92.47 27	96.00 33	96.50 35	94.70 32	87.83 43	90.49 39	89.92 18	74.68 90	89.35 36	90.66 30	94.02 30	94.14 16	95.67 43	96.85 27

X-MVS		92.36 30	92.75 30	91.90 33	96.89 17	96.70 25	95.25 27	90.48 29	91.50 33	83.95 48	88.20 31	88.82 40	89.11 36	93.75 35	93.43 34	95.75 39	96.83 29

MSLP-MVS++		92.02 34	91.40 37	92.75 24	96.01 32	95.88 44	93.73 40	89.00 34	89.89 44	90.31 15	81.28 56	88.85 39	91.45 23	92.88 47	94.24 14	96.00 23	96.76 30

train_agg		92.87 25	93.53 24	92.09 30	96.88 18	95.38 49	95.94 14	90.59 28	90.65 38	83.65 51	94.31 13	91.87 19	90.30 31	93.38 39	92.42 46	95.17 69	96.73 31

ACMMPR		93.72 18	93.94 19	93.48 18	97.07 11	96.93 17	95.78 21	90.66 26	93.88 14	89.24 21	93.53 16	89.08 38	92.24 11	93.89 34	93.50 31	95.88 27	96.73 31

PVSNet_Blended_VisFu		87.40 69	87.80 64	86.92 73	92.86 64	95.40 48	88.56 97	83.45 82	79.55 103	82.26 58	74.49 91	84.03 57	79.24 124	92.97 46	91.53 54	95.15 71	96.65 33

DeepC-MVS_fast	88.76 1	93.10 22	93.02 29	93.19 22	97.13 9	96.51 33	95.35 26	91.19 19	93.14 21	88.14 26	85.26 41	89.49 35	91.45 23	95.17 9	95.07 2	95.85 32	96.48 34
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
CPTT-MVS		91.39 37	90.95 40	91.91 32	95.06 39	95.24 51	95.02 30	88.98 36	91.02 35	86.71 34	84.89 43	88.58 43	91.60 22	90.82 81	89.67 92	94.08 116	96.45 35

ACMMP		92.03 33	92.16 31	91.87 34	95.88 34	96.55 31	94.47 35	89.49 33	91.71 31	85.26 42	91.52 24	84.48 55	90.21 32	92.82 48	91.63 52	95.92 25	96.42 36
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		92.76 26	93.03 28	92.45 28	97.03 13	96.67 28	95.73 23	87.92 42	90.15 43	86.53 36	92.97 20	88.33 44	91.69 21	93.62 37	93.03 39	95.83 33	96.41 37

QAPM		89.49 48	89.58 49	89.38 52	94.73 45	95.94 42	92.35 49	85.00 60	85.69 60	80.03 72	76.97 78	87.81 46	87.87 50	92.18 58	92.10 48	96.33 15	96.40 38

HQP-MVS		89.13 51	89.58 49	88.60 60	93.53 56	93.67 73	93.29 43	87.58 45	88.53 49	75.50 89	87.60 34	80.32 74	87.07 61	90.66 87	89.95 84	94.62 95	96.35 39

PHI-MVS		92.05 32	93.74 21	90.08 44	94.96 41	97.06 13	93.11 45	87.71 44	90.71 37	80.78 68	92.40 22	91.03 22	87.68 53	94.32 27	94.48 12	96.21 19	96.16 40

MVS_0304		90.88 40	91.35 38	90.34 42	93.91 52	96.79 23	94.49 34	86.54 50	86.57 55	82.85 55	81.68 53	89.70 33	87.57 55	94.64 19	93.93 20	96.67 10	96.15 41

UGNet		85.90 78	88.23 58	83.18 108	88.96 106	94.10 68	87.52 106	83.60 74	81.66 83	77.90 83	80.76 59	83.19 61	66.70 185	91.13 75	90.71 67	94.39 109	96.06 42
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
CANet		91.33 38	91.46 36	91.18 36	95.01 40	96.71 24	93.77 38	87.39 46	87.72 51	87.26 31	81.77 51	89.73 32	87.32 59	94.43 24	93.86 21	96.31 17	96.02 43

DELS-MVS		89.71 46	89.68 48	89.74 47	93.75 54	96.22 38	93.76 39	85.84 53	82.53 73	85.05 44	78.96 68	84.24 56	84.25 75	94.91 12	94.91 4	95.78 38	96.02 43
Christian Sormann, Emanuele Santellani, Mattia Rossi, Andreas Kuhn, Friedrich Fraundorfer: DELS-MVS: Deep Epipolar Line Search for Multi-View Stereo. Winter Conference on Applications of Computer Vision (WACV), 2023
CDPH-MVS		91.14 39	92.01 32	90.11 43	96.18 29	96.18 39	94.89 31	88.80 38	88.76 48	77.88 84	89.18 30	87.71 47	87.29 60	93.13 42	93.31 36	95.62 46	95.84 45

LGP-MVS_train		88.25 61	88.55 54	87.89 66	92.84 66	93.66 74	93.35 42	85.22 59	85.77 58	74.03 98	86.60 38	76.29 98	86.62 65	91.20 68	90.58 71	95.29 65	95.75 46

3Dnovator	85.17 5	90.48 42	89.90 46	91.16 37	94.88 43	95.74 46	93.82 37	85.36 57	89.28 45	87.81 28	74.34 92	87.40 48	88.56 43	93.07 43	93.74 25	96.53 12	95.71 47

MVS_111021_HR		90.56 41	91.29 39	89.70 49	94.71 46	95.63 47	91.81 57	86.38 51	87.53 52	81.29 62	87.96 32	85.43 52	87.69 52	93.90 33	92.93 41	96.33 15	95.69 48

anonymousdsp		77.94 154	79.00 140	76.71 167	79.03 195	87.83 156	79.58 179	72.87 171	65.80 183	58.86 183	65.82 135	62.48 160	75.99 143	86.77 132	88.66 112	93.92 123	95.68 49

PCF-MVS	84.60 6	88.66 54	87.75 67	89.73 48	93.06 62	96.02 40	93.22 44	90.00 31	82.44 76	80.02 73	77.96 74	85.16 53	87.36 58	88.54 110	88.54 114	94.72 89	95.61 50
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
EPNet		89.60 47	89.91 45	89.24 54	96.45 27	93.61 75	92.95 47	88.03 40	85.74 59	83.36 52	87.29 36	83.05 62	80.98 94	92.22 55	91.85 50	93.69 134	95.58 51
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
EPP-MVSNet		86.55 71	87.76 66	85.15 83	90.52 87	94.41 64	87.24 113	82.32 97	81.79 82	73.60 100	78.57 70	82.41 65	82.07 88	91.23 66	90.39 73	95.14 72	95.48 52

abl_6				90.66 40	94.65 47	96.27 37	92.21 50	86.94 48	90.23 41	86.38 37	85.50 40	92.96 9	88.37 46			95.40 57	95.46 53

ACMP	83.90 8	88.32 60	88.06 60	88.62 59	92.18 70	93.98 71	91.28 62	85.24 58	86.69 54	81.23 63	85.62 39	75.13 102	87.01 63	89.83 94	89.77 89	94.79 83	95.43 54
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
ETV-MVS		89.22 50	89.76 47	88.60 60	91.60 73	94.61 63	89.48 81	83.46 81	85.20 62	81.58 60	82.75 46	82.59 64	88.80 38	94.57 21	93.28 37	96.68 8	95.31 55

DCV-MVSNet		85.88 79	86.17 77	85.54 81	89.10 105	89.85 126	89.34 82	80.70 108	83.04 71	78.08 82	76.19 81	79.00 85	82.42 85	89.67 97	90.30 74	93.63 137	95.12 56

test_part1		83.23 101	80.55 123	86.35 75	88.60 109	90.61 110	90.78 64	81.13 106	70.89 158	83.01 53	55.72 186	74.60 104	82.19 86	87.79 118	89.26 100	92.39 153	95.01 57

CLD-MVS		88.66 54	88.52 55	88.82 56	91.37 77	94.22 66	92.82 48	82.08 98	88.27 50	85.14 43	81.86 49	78.53 88	85.93 69	91.17 70	90.61 69	95.55 51	95.00 58
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
IS_MVSNet		86.18 74	88.18 59	83.85 101	91.02 80	94.72 62	87.48 107	82.46 95	81.05 90	70.28 113	76.98 77	82.20 67	76.65 140	93.97 31	93.38 35	95.18 68	94.97 59

CS-MVS		88.97 52	89.44 51	88.41 64	91.45 75	95.24 51	90.03 70	82.43 96	84.08 68	81.16 64	81.02 58	83.83 58	88.74 40	94.25 28	92.73 45	96.67 10	94.95 60

Anonymous20231211		84.42 92	83.02 98	86.05 77	88.85 107	92.70 90	88.92 92	83.40 83	79.99 98	78.31 79	55.83 185	78.92 86	83.33 79	89.06 105	89.76 90	93.50 139	94.90 61

UniMVSNet (Re)		81.22 120	81.08 115	81.39 126	85.35 143	91.76 101	84.93 141	82.88 85	76.13 122	65.02 142	64.94 144	63.09 154	75.17 148	87.71 120	89.04 107	94.97 76	94.88 62

PVSNet_BlendedMVS		88.19 62	88.00 62	88.42 62	92.71 68	94.82 60	89.08 86	83.81 70	84.91 65	86.38 37	79.14 65	78.11 89	82.66 82	93.05 44	91.10 56	95.86 30	94.86 63

PVSNet_Blended		88.19 62	88.00 62	88.42 62	92.71 68	94.82 60	89.08 86	83.81 70	84.91 65	86.38 37	79.14 65	78.11 89	82.66 82	93.05 44	91.10 56	95.86 30	94.86 63

OpenMVS	82.53 11	87.71 65	86.84 72	88.73 57	94.42 48	95.06 55	91.02 63	83.49 78	82.50 75	82.24 59	67.62 129	85.48 51	85.56 70	91.19 69	91.30 55	95.67 43	94.75 65

IB-MVS	79.09 12	82.60 106	82.19 104	83.07 109	91.08 79	93.55 76	80.90 174	81.35 103	76.56 119	80.87 66	64.81 146	69.97 130	68.87 175	85.64 149	90.06 80	95.36 60	94.74 66
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
MVS_111021_LR		90.14 45	90.89 41	89.26 53	93.23 59	94.05 70	90.43 66	84.65 62	90.16 42	84.52 47	90.14 28	83.80 59	87.99 49	92.50 52	90.92 61	94.74 87	94.70 67

FC-MVSNet-train		85.18 82	85.31 86	85.03 84	90.67 84	91.62 102	87.66 105	83.61 73	79.75 101	74.37 96	78.69 69	71.21 124	78.91 125	91.23 66	89.96 83	94.96 77	94.69 68

Vis-MVSNet		84.38 93	86.68 76	81.70 122	87.65 119	94.89 58	88.14 100	80.90 107	74.48 133	68.23 125	77.53 75	80.72 72	69.98 172	92.68 49	91.90 49	95.33 63	94.58 69
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
GeoE		84.62 87	83.98 94	85.35 82	89.34 101	92.83 88	88.34 98	78.95 132	79.29 105	77.16 88	68.10 126	74.56 105	83.40 78	89.31 103	89.23 102	94.92 78	94.57 70

casdiffmvs		87.45 68	87.15 70	87.79 69	90.15 96	94.22 66	89.96 72	83.93 69	85.08 63	80.91 65	75.81 83	77.88 92	86.08 67	91.86 61	90.86 62	95.74 40	94.37 71

EIA-MVS		87.94 64	88.05 61	87.81 67	91.46 74	95.00 57	88.67 93	82.81 86	82.53 73	80.81 67	80.04 61	80.20 75	87.48 56	92.58 51	91.61 53	95.63 45	94.36 72

diffmvs		86.52 72	86.76 75	86.23 76	88.31 112	92.63 92	89.58 78	81.61 102	86.14 56	80.26 70	79.00 67	77.27 94	83.58 76	88.94 106	89.06 106	94.05 118	94.29 73

UniMVSNet_NR-MVSNet		81.87 112	81.33 112	82.50 114	85.31 144	91.30 103	85.70 132	84.25 64	75.89 123	64.21 145	66.95 131	64.65 148	80.22 106	87.07 125	89.18 104	95.27 67	94.29 73

DU-MVS		81.20 121	80.30 125	82.25 117	84.98 151	90.94 107	85.70 132	83.58 76	75.74 124	64.21 145	65.30 141	59.60 176	80.22 106	86.89 128	89.31 98	94.77 85	94.29 73

Anonymous202405211			82.75 102		89.58 100	92.97 86	89.04 89	84.13 67	78.72 109		57.18 181	76.64 97	83.13 80	89.55 99	89.92 85	93.38 142	94.28 76

Effi-MVS+		85.33 81	85.08 87	85.63 80	89.69 99	93.42 78	89.90 73	80.31 115	79.32 104	72.48 108	73.52 98	74.03 109	86.55 66	90.99 77	89.98 82	94.83 82	94.27 77

MAR-MVS		88.39 59	88.44 56	88.33 65	94.90 42	95.06 55	90.51 65	83.59 75	85.27 61	79.07 76	77.13 76	82.89 63	87.70 51	92.19 57	92.32 47	94.23 113	94.20 78
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
DI_MVS_plusplus_trai		86.41 73	85.54 85	87.42 71	89.24 102	93.13 81	92.16 52	82.65 92	82.30 77	80.75 69	68.30 125	80.41 73	85.01 72	90.56 88	90.07 79	94.70 91	94.01 79

OMC-MVS		90.23 44	90.40 43	90.03 45	93.45 57	95.29 50	91.89 56	86.34 52	93.25 20	84.94 45	81.72 52	86.65 49	88.90 37	91.69 62	90.27 75	94.65 93	93.95 80

NR-MVSNet		80.25 127	79.98 131	80.56 137	85.20 146	90.94 107	85.65 134	83.58 76	75.74 124	61.36 169	65.30 141	56.75 189	72.38 164	88.46 112	88.80 111	95.16 70	93.87 81

DPM-MVS		91.72 35	91.48 35	92.00 31	95.53 37	95.75 45	95.94 14	91.07 21	91.20 34	85.58 41	81.63 54	90.74 26	88.40 45	93.40 38	93.75 24	95.45 56	93.85 82

CNLPA		88.40 57	87.00 71	90.03 45	93.73 55	94.28 65	89.56 79	85.81 54	91.87 29	87.55 29	69.53 118	81.49 68	89.23 35	89.45 101	88.59 113	94.31 112	93.82 83

tttt0517		85.11 84	85.81 82	84.30 92	89.24 102	92.68 91	87.12 118	80.11 118	81.98 80	74.31 97	78.08 73	73.57 114	79.90 112	91.01 76	89.58 93	95.11 75	93.77 84

thisisatest0530		85.15 83	85.86 81	84.33 91	89.19 104	92.57 95	87.22 114	80.11 118	82.15 79	74.41 95	78.15 72	73.80 112	79.90 112	90.99 77	89.58 93	95.13 73	93.75 85

OPM-MVS		87.56 67	85.80 83	89.62 50	93.90 53	94.09 69	94.12 36	88.18 39	75.40 127	77.30 87	76.41 79	77.93 91	88.79 39	92.20 56	90.82 63	95.40 57	93.72 86

UniMVSNet_ETH3D		79.24 141	76.47 167	82.48 115	85.66 139	90.97 106	86.08 129	81.63 101	64.48 188	68.94 123	54.47 188	57.65 184	78.83 126	85.20 159	88.91 110	93.72 133	93.60 87

AdaColmap		90.29 43	88.38 57	92.53 26	96.10 31	95.19 53	92.98 46	91.40 17	89.08 47	88.65 23	78.35 71	81.44 69	91.30 28	90.81 82	90.21 76	94.72 89	93.59 88

MVS_Test		86.93 70	87.24 69	86.56 74	90.10 97	93.47 77	90.31 67	80.12 117	83.55 70	78.12 80	79.58 64	79.80 79	85.45 71	90.17 90	90.59 70	95.29 65	93.53 89

TranMVSNet+NR-MVSNet		80.52 124	79.84 133	81.33 128	84.92 153	90.39 113	85.53 137	84.22 66	74.27 136	60.68 174	64.93 145	59.96 171	77.48 135	86.75 133	89.28 99	95.12 74	93.29 90

CP-MVSNet		76.36 172	76.41 168	76.32 171	82.73 180	88.64 150	79.39 181	79.62 124	67.21 174	53.70 189	60.72 164	55.22 195	67.91 180	83.52 173	86.34 143	94.55 99	93.19 91

Baseline_NR-MVSNet		79.84 131	78.37 148	81.55 125	84.98 151	86.66 165	85.06 139	83.49 78	75.57 126	63.31 152	58.22 180	60.97 166	78.00 131	86.89 128	87.13 127	94.47 104	93.15 92

PS-CasMVS		75.90 177	75.86 176	75.96 173	82.59 181	88.46 153	79.23 184	79.56 126	66.00 181	52.77 191	59.48 172	54.35 199	67.14 183	83.37 174	86.23 144	94.47 104	93.10 93

baseline		84.89 85	86.06 80	83.52 106	87.25 123	89.67 133	87.76 103	75.68 162	84.92 64	78.40 78	80.10 60	80.98 70	80.20 108	86.69 135	87.05 129	91.86 160	92.99 94

ACMM	83.27 10	87.68 66	86.09 79	89.54 51	93.26 58	92.19 98	91.43 60	86.74 49	86.02 57	82.85 55	75.63 84	75.14 101	88.41 44	90.68 86	89.99 81	94.59 96	92.97 95
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
v7n		77.22 160	76.23 170	78.38 157	81.89 186	89.10 146	82.24 165	76.36 156	65.96 182	61.21 171	56.56 183	55.79 193	75.07 150	86.55 137	86.68 135	93.52 138	92.95 96

Fast-Effi-MVS+		83.77 96	82.98 99	84.69 85	87.98 113	91.87 100	88.10 101	77.70 146	78.10 113	73.04 104	69.13 120	68.51 137	86.66 64	90.49 89	89.85 87	94.67 92	92.88 97

WR-MVS_H		75.84 178	76.93 164	74.57 184	82.86 177	89.50 137	78.34 187	79.36 129	66.90 176	52.51 192	60.20 168	59.71 173	59.73 194	83.61 172	85.77 152	94.65 93	92.84 98

v144192		78.81 145	77.22 160	80.67 135	82.95 174	89.79 129	86.40 125	77.42 147	68.26 173	63.13 153	59.50 171	58.13 182	80.08 111	85.93 145	86.08 147	94.06 117	92.83 99

MVSTER		86.03 76	86.12 78	85.93 78	88.62 108	89.93 124	89.33 83	79.91 122	81.87 81	81.35 61	81.07 57	74.91 103	80.66 99	92.13 59	90.10 78	95.68 42	92.80 100

WR-MVS		76.63 165	78.02 153	75.02 179	84.14 161	89.76 130	78.34 187	80.64 109	69.56 165	52.32 193	61.26 158	61.24 165	60.66 193	84.45 167	87.07 128	93.99 121	92.77 101

v1921920		78.57 150	76.99 163	80.41 140	82.93 175	89.63 135	86.38 126	77.14 150	68.31 172	61.80 165	58.89 177	56.79 188	80.19 109	86.50 140	86.05 149	94.02 119	92.76 102

thisisatest0515		79.76 133	80.59 122	78.80 150	84.40 156	88.91 149	79.48 180	76.94 152	72.29 151	67.33 128	67.82 128	65.99 143	70.80 170	88.50 111	87.84 119	93.86 127	92.75 103

v1192		78.94 144	77.33 158	80.82 133	83.25 169	89.90 125	86.91 120	77.72 145	68.63 171	62.61 157	59.17 173	57.53 185	80.62 102	86.89 128	86.47 140	93.79 131	92.75 103

IterMVS-LS		83.28 100	82.95 100	83.65 102	88.39 111	88.63 151	86.80 122	78.64 137	76.56 119	73.43 101	72.52 103	75.35 100	80.81 96	86.43 141	88.51 115	93.84 128	92.66 105
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
LTVRE_ROB	74.41 16	75.78 179	74.72 185	77.02 165	85.88 134	89.22 141	82.44 161	77.17 149	50.57 210	45.45 204	65.44 139	52.29 203	81.25 91	85.50 152	87.42 124	89.94 180	92.62 106
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
V42		79.59 135	78.43 146	80.94 132	82.79 179	89.71 131	86.66 123	76.73 155	71.38 154	67.42 127	61.01 161	62.30 161	78.39 128	85.56 151	86.48 139	93.65 136	92.60 107

thres600view7		82.53 108	81.02 116	84.28 93	90.61 86	93.05 83	88.57 96	82.67 90	74.12 138	68.56 124	65.09 143	62.13 163	80.40 104	91.15 72	89.02 108	94.88 80	92.59 108

Effi-MVS+-dtu		82.05 110	81.76 106	82.38 116	87.72 116	90.56 111	86.90 121	78.05 142	73.85 141	66.85 130	71.29 106	71.90 122	82.00 89	86.64 136	85.48 155	92.76 150	92.58 109

TAPA-MVS	84.37 7	88.91 53	88.93 53	88.89 55	93.00 63	94.85 59	92.00 53	84.84 61	91.68 32	80.05 71	79.77 63	84.56 54	88.17 48	90.11 91	89.00 109	95.30 64	92.57 110
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
v10		79.62 134	78.19 149	81.28 129	83.73 164	89.69 132	87.27 112	76.86 153	70.50 162	65.46 137	60.58 166	60.47 168	80.44 103	86.91 127	86.63 137	93.93 122	92.55 111

ACMH	78.52 14	81.86 113	80.45 124	83.51 107	90.51 89	91.22 104	85.62 135	84.23 65	70.29 164	62.21 159	69.04 122	64.05 150	84.48 74	87.57 121	88.45 116	94.01 120	92.54 112
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
v1144		79.38 140	77.83 154	81.18 130	83.62 165	90.23 116	87.15 117	78.35 139	69.13 167	64.02 148	60.20 168	59.41 177	80.14 110	86.78 131	86.57 138	93.81 130	92.53 113

tfpn200view9		82.86 102	81.46 109	84.48 88	90.30 94	93.09 82	89.05 88	82.71 88	75.14 128	69.56 116	65.72 136	63.13 152	80.38 105	91.15 72	89.51 95	94.91 79	92.50 114

v1240		78.15 152	76.53 166	80.04 141	82.85 178	89.48 138	85.61 136	76.77 154	67.05 175	61.18 172	58.37 179	56.16 192	79.89 114	86.11 144	86.08 147	93.92 123	92.47 115

thres400		82.68 105	81.15 114	84.47 89	90.52 87	92.89 87	88.95 91	82.71 88	74.33 135	69.22 121	65.31 140	62.61 158	80.63 100	90.96 79	89.50 96	94.79 83	92.45 116

CHOSEN 1792x2688		82.16 109	80.91 119	83.61 103	91.14 78	92.01 99	89.55 80	79.15 131	79.87 99	70.29 112	52.51 194	72.56 119	81.39 90	88.87 108	88.17 117	90.15 178	92.37 117

tfpnnormal		77.46 159	74.86 184	80.49 138	86.34 132	88.92 148	84.33 148	81.26 104	61.39 196	61.70 166	51.99 195	53.66 201	74.84 151	88.63 109	87.38 125	94.50 101	92.08 118

CANet_DTU		85.43 80	87.72 68	82.76 112	90.95 83	93.01 85	89.99 71	75.46 163	82.67 72	64.91 143	83.14 45	80.09 76	80.68 98	92.03 60	91.03 58	94.57 98	92.08 118

thres200		82.77 104	81.25 113	84.54 87	90.38 91	93.05 83	89.13 85	82.67 90	74.40 134	69.53 118	65.69 138	63.03 155	80.63 100	91.15 72	89.42 97	94.88 80	92.04 120

PEN-MVS		76.02 175	76.07 171	75.95 174	83.17 171	87.97 155	79.65 178	80.07 121	66.57 178	51.45 195	60.94 162	55.47 194	66.81 184	82.72 177	86.80 133	94.59 96	92.03 121

v8		79.90 130	78.39 147	81.66 123	83.97 162	89.81 127	87.16 116	77.40 148	71.49 153	67.71 126	61.24 159	62.49 159	79.83 115	85.48 153	86.17 145	93.89 125	92.02 122

ET-MVSNet_ETH3D		84.65 86	85.58 84	83.56 105	74.99 206	92.62 94	90.29 68	80.38 110	82.16 78	73.01 105	83.41 44	71.10 125	87.05 62	87.77 119	90.17 77	95.62 46	91.82 123

TSAR-MVS + COLMAP		88.40 57	89.09 52	87.60 70	92.72 67	93.92 72	92.21 50	85.57 56	91.73 30	73.72 99	91.75 23	73.22 118	87.64 54	91.49 64	89.71 91	93.73 132	91.82 123

GBi-Net		84.51 89	84.80 88	84.17 95	84.20 158	89.95 121	89.70 75	80.37 111	81.17 86	75.50 89	69.63 114	79.69 81	79.75 116	90.73 83	90.72 64	95.52 53	91.71 125

test1		84.51 89	84.80 88	84.17 95	84.20 158	89.95 121	89.70 75	80.37 111	81.17 86	75.50 89	69.63 114	79.69 81	79.75 116	90.73 83	90.72 64	95.52 53	91.71 125

FMVSNet2		83.87 94	83.73 96	84.05 99	84.20 158	89.95 121	89.70 75	80.21 116	79.17 107	74.89 93	65.91 134	77.49 93	79.75 116	90.87 80	91.00 60	95.52 53	91.71 125

v2v482		79.84 131	78.07 151	81.90 120	83.75 163	90.21 118	87.17 115	79.85 123	70.65 160	65.93 135	61.93 156	60.07 170	80.82 95	85.25 155	86.71 134	93.88 126	91.70 128

FMVSNet1		81.64 117	80.61 121	82.84 111	82.36 183	89.20 142	88.67 93	79.58 125	70.79 159	72.63 107	58.95 176	72.26 121	79.34 122	90.73 83	90.72 64	94.47 104	91.62 129

thres100view900		82.55 107	81.01 118	84.34 90	90.30 94	92.27 96	89.04 89	82.77 87	75.14 128	69.56 116	65.72 136	63.13 152	79.62 119	89.97 93	89.26 100	94.73 88	91.61 130

FMVSNet3		84.44 91	84.64 90	84.21 94	84.32 157	90.13 119	89.85 74	80.37 111	81.17 86	75.50 89	69.63 114	79.69 81	79.62 119	89.72 96	90.52 72	95.59 49	91.58 131

Vis-MVSNet (Re-imp)		83.65 97	86.81 74	79.96 142	90.46 90	92.71 89	84.84 143	82.00 99	80.93 92	62.44 158	76.29 80	82.32 66	65.54 188	92.29 53	91.66 51	94.49 103	91.47 132

Fast-Effi-MVS+-dtu		79.95 129	80.69 120	79.08 147	86.36 131	89.14 144	85.85 130	72.28 173	72.85 150	59.32 179	70.43 112	68.42 138	77.57 134	86.14 143	86.44 141	93.11 146	91.39 133

pm-mvs1		78.51 151	77.75 156	79.40 145	84.83 154	89.30 139	83.55 154	79.38 128	62.64 192	63.68 150	58.73 178	64.68 147	70.78 171	89.79 95	87.84 119	94.17 115	91.28 134

ACMH+	79.08 13	81.84 114	80.06 129	83.91 100	89.92 98	90.62 109	86.21 127	83.48 80	73.88 140	65.75 136	66.38 133	65.30 146	84.63 73	85.90 146	87.25 126	93.45 140	91.13 135

baseline2		82.80 103	82.86 101	82.73 113	87.68 118	90.50 112	84.92 142	78.93 133	78.07 114	73.06 103	75.08 88	69.77 131	77.31 136	88.90 107	86.94 131	94.50 101	90.74 136

UA-Net		86.07 75	87.78 65	84.06 98	92.85 65	95.11 54	87.73 104	84.38 63	73.22 147	73.18 102	79.99 62	89.22 37	71.47 168	93.22 41	93.03 39	94.76 86	90.69 137

CDS-MVSNet		81.63 118	82.09 105	81.09 131	87.21 124	90.28 115	87.46 109	80.33 114	69.06 168	70.66 110	71.30 105	73.87 110	67.99 178	89.58 98	89.87 86	92.87 149	90.69 137
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
v148		78.59 149	76.84 165	80.62 136	83.61 166	89.16 143	83.65 153	79.24 130	69.38 166	69.34 120	59.88 170	60.41 169	75.19 147	83.81 171	84.63 164	92.70 151	90.63 139

pmmvs6		74.83 183	72.89 190	77.09 163	82.11 184	87.50 159	80.88 175	76.97 151	52.79 208	61.91 164	46.66 200	60.49 167	69.28 174	86.74 134	85.46 156	91.39 165	90.56 140

DTE-MVSNet		75.14 182	75.44 181	74.80 181	83.18 170	87.19 162	78.25 189	80.11 118	66.05 180	48.31 200	60.88 163	54.67 196	64.54 189	82.57 179	86.17 145	94.43 107	90.53 141

PLC	83.76 9	88.61 56	86.83 73	90.70 39	94.22 49	92.63 92	91.50 59	87.19 47	89.16 46	86.87 33	75.51 85	80.87 71	89.98 34	90.01 92	89.20 103	94.41 108	90.45 142
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
IterMVS-SCA-FT		79.41 139	80.20 127	78.49 155	85.88 134	86.26 167	83.95 150	71.94 174	73.55 145	61.94 162	70.48 111	70.50 127	75.23 146	85.81 148	84.61 165	91.99 159	90.18 143

GA-MVS		79.52 136	79.71 136	79.30 146	85.68 138	90.36 114	84.55 145	78.44 138	70.47 163	57.87 184	68.52 124	61.38 164	76.21 142	89.40 102	87.89 118	93.04 147	89.96 144

LS3D		85.96 77	84.37 92	87.81 67	94.13 50	93.27 80	90.26 69	89.00 34	84.91 65	72.84 106	71.74 104	72.47 120	87.45 57	89.53 100	89.09 105	93.20 144	89.60 145

gg-mvs-nofinetune		75.64 180	77.26 159	73.76 185	87.92 114	92.20 97	87.32 110	64.67 203	51.92 209	35.35 213	46.44 201	77.05 96	71.97 165	92.64 50	91.02 59	95.34 62	89.53 146

SixPastTwentyTwo		76.02 175	75.72 177	76.36 170	83.38 167	87.54 158	75.50 194	76.22 157	65.50 185	57.05 185	70.64 108	53.97 200	74.54 153	80.96 185	82.12 180	91.44 164	89.35 147

HyFIR lowres test		81.62 119	79.45 139	84.14 97	91.00 81	93.38 79	88.27 99	78.19 140	76.28 121	70.18 114	48.78 198	73.69 113	83.52 77	87.05 126	87.83 121	93.68 135	89.15 148

CostFormer		80.94 122	80.21 126	81.79 121	87.69 117	88.58 152	87.47 108	70.66 179	80.02 97	77.88 84	73.03 99	71.40 123	78.24 129	79.96 189	79.63 185	88.82 184	88.84 149

IterMVS		78.79 146	79.71 136	77.71 159	85.26 145	85.91 170	84.54 146	69.84 185	73.38 146	61.25 170	70.53 110	70.35 128	74.43 155	85.21 158	83.80 170	90.95 172	88.77 150
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
baseline1		84.54 88	84.43 91	84.67 86	90.62 85	91.16 105	88.63 95	83.75 72	79.78 100	71.16 109	75.14 87	74.10 108	77.84 133	91.56 63	90.67 68	96.04 22	88.58 151

pmmvs5		76.93 162	76.33 169	77.62 160	81.97 185	88.40 154	81.32 170	74.35 167	65.42 186	61.42 168	63.07 152	57.95 183	73.23 162	85.60 150	85.35 157	93.41 141	88.55 152

RPMNet		77.07 161	77.63 157	76.42 169	85.56 141	85.15 178	81.37 168	65.27 200	74.71 131	60.29 175	63.71 151	66.59 142	73.64 158	82.71 178	82.12 180	92.38 154	88.39 153

test-mter		77.79 155	80.02 130	75.18 178	81.18 191	82.85 189	80.52 177	62.03 207	73.62 144	62.16 160	73.55 97	73.83 111	73.81 157	84.67 164	83.34 172	91.37 166	88.31 154

CR-MVSNet		78.71 147	78.86 141	78.55 154	85.85 137	85.15 178	82.30 163	68.23 189	74.71 131	65.37 139	64.39 148	69.59 133	77.18 137	85.10 161	84.87 160	92.34 155	88.21 155

PatchT		76.42 169	77.81 155	74.80 181	78.46 198	84.30 183	71.82 201	65.03 202	73.89 139	65.37 139	61.58 157	66.70 141	77.18 137	85.10 161	84.87 160	90.94 173	88.21 155

pmmvs4		79.99 128	78.08 150	82.22 118	83.04 173	87.16 163	84.95 140	78.80 136	78.64 110	74.53 94	64.61 147	59.41 177	79.45 121	84.13 169	84.54 166	92.53 152	88.08 157

PM-MVS		74.17 187	73.10 188	75.41 176	76.07 203	82.53 192	77.56 190	71.69 175	71.04 155	61.92 163	61.23 160	47.30 209	74.82 152	81.78 183	79.80 184	90.42 175	88.05 158

CVMVSNet		76.70 164	78.46 145	74.64 183	83.34 168	84.48 182	81.83 167	74.58 165	68.88 169	51.23 197	69.77 113	70.05 129	67.49 181	84.27 168	83.81 169	89.38 182	87.96 159

TransMVSNet (Re)		76.57 166	75.16 183	78.22 158	85.60 140	87.24 161	82.46 159	81.23 105	59.80 200	59.05 182	57.07 182	59.14 180	66.60 186	88.09 115	86.82 132	94.37 110	87.95 160

EG-PatchMatch MVS		76.40 171	75.47 180	77.48 161	85.86 136	90.22 117	82.45 160	73.96 169	59.64 201	59.60 178	52.75 193	62.20 162	68.44 177	88.23 114	87.50 122	94.55 99	87.78 161

EPNet_dtu		81.98 111	83.82 95	79.83 144	94.10 51	85.97 169	87.29 111	84.08 68	80.61 95	59.96 176	81.62 55	77.19 95	62.91 192	87.21 123	86.38 142	90.66 174	87.77 162
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
TDRefinement		79.05 143	77.05 162	81.39 126	88.45 110	89.00 147	86.92 119	82.65 92	74.21 137	64.41 144	59.17 173	59.16 179	74.52 154	85.23 156	85.09 158	91.37 166	87.51 163

CHOSEN 280x420		80.28 126	81.66 107	78.67 153	82.92 176	79.24 203	85.36 138	66.79 195	78.11 112	70.32 111	75.03 89	79.87 77	81.09 93	89.07 104	83.16 173	85.54 200	87.17 164

test-LLR		79.47 138	79.84 133	79.03 148	87.47 120	82.40 194	81.24 171	78.05 142	73.72 142	62.69 155	73.76 95	74.42 106	73.49 159	84.61 165	82.99 175	91.25 168	87.01 165

TESTMET0.1,1		77.78 156	79.84 133	75.38 177	80.86 192	82.40 194	81.24 171	62.72 206	73.72 142	62.69 155	73.76 95	74.42 106	73.49 159	84.61 165	82.99 175	91.25 168	87.01 165

PatchMatch-RL		83.34 99	81.36 111	85.65 79	90.33 93	89.52 136	84.36 147	81.82 100	80.87 94	79.29 74	74.04 93	62.85 157	86.05 68	88.40 113	87.04 130	92.04 157	86.77 167

FC-MVSNet-test		76.53 168	81.62 108	70.58 193	84.99 150	85.73 172	74.81 195	78.85 135	77.00 118	39.13 211	75.90 82	73.50 115	54.08 200	86.54 138	85.99 150	91.65 162	86.68 168

tpm		76.30 173	76.05 173	76.59 168	86.97 126	83.01 188	83.83 151	67.06 194	71.83 152	63.87 149	69.56 117	62.88 156	73.41 161	79.79 190	78.59 189	84.41 202	86.68 168

PMMVS		81.65 116	84.05 93	78.86 149	78.56 197	82.63 191	83.10 155	67.22 193	81.39 84	70.11 115	84.91 42	79.74 80	82.12 87	87.31 122	85.70 153	92.03 158	86.67 170

MS-PatchMatch		81.79 115	81.44 110	82.19 119	90.35 92	89.29 140	88.08 102	75.36 164	77.60 115	69.00 122	64.37 149	78.87 87	77.14 139	88.03 116	85.70 153	93.19 145	86.24 171

RPSCF		83.46 98	83.36 97	83.59 104	87.75 115	87.35 160	84.82 144	79.46 127	83.84 69	78.12 80	82.69 47	79.87 77	82.60 84	82.47 180	81.13 183	88.78 185	86.13 172

pmmvs-eth3d		74.32 186	71.96 192	77.08 164	77.33 200	82.71 190	78.41 186	76.02 160	66.65 177	65.98 134	54.23 190	49.02 208	73.14 163	82.37 181	82.69 177	91.61 163	86.05 173

MSDG		83.87 94	81.02 116	87.19 72	92.17 71	89.80 128	89.15 84	85.72 55	80.61 95	79.24 75	66.66 132	68.75 136	82.69 81	87.95 117	87.44 123	94.19 114	85.92 174

EU-MVSNet		69.98 194	72.30 191	67.28 198	75.67 205	79.39 202	73.12 199	69.94 184	63.59 191	42.80 207	62.93 153	56.71 190	55.07 198	79.13 194	78.55 190	87.06 194	85.82 175

tpm cat1		77.78 156	75.28 182	80.70 134	87.14 125	85.84 171	85.81 131	70.40 180	77.44 116	78.80 77	63.72 150	64.01 151	76.55 141	75.60 201	75.21 199	85.51 201	85.12 176

ambc			61.92 203		70.98 208	73.54 208	63.64 209		60.06 198	52.23 194	38.44 208	19.17 219	57.12 195	82.33 182	75.03 201	83.21 206	84.89 177

COLMAP_ROB	76.78 15	80.50 125	78.49 144	82.85 110	90.96 82	89.65 134	86.20 128	83.40 83	77.15 117	66.54 131	62.27 154	65.62 145	77.89 132	85.23 156	84.70 163	92.11 156	84.83 178
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
USDC		80.69 123	79.89 132	81.62 124	86.48 130	89.11 145	86.53 124	78.86 134	81.15 89	63.48 151	72.98 100	59.12 181	81.16 92	87.10 124	85.01 159	93.23 143	84.77 179

gm-plane-assit		70.29 193	70.65 195	69.88 194	85.03 149	78.50 204	58.41 211	65.47 199	50.39 211	40.88 209	49.60 197	50.11 205	75.14 149	91.43 65	89.78 88	94.32 111	84.73 180

GG-mvs-BLEND		57.56 204	82.61 103	28.34 212	0.22 220	90.10 120	79.37 182	0.14 218	79.56 102	0.40 221	71.25 107	83.40 60	0.30 218	86.27 142	83.87 168	89.59 181	83.83 181

Anonymous20231206		70.80 192	70.59 196	71.04 192	81.60 188	82.49 193	74.64 196	75.87 161	64.17 189	49.27 199	44.85 204	53.59 202	54.68 199	83.07 175	82.34 179	90.17 177	83.65 182

dps		78.02 153	75.94 175	80.44 139	86.06 133	86.62 166	82.58 158	69.98 183	75.14 128	77.76 86	69.08 121	59.93 172	78.47 127	79.47 191	77.96 192	87.78 189	83.40 183

tpmrst		76.55 167	75.99 174	77.20 162	87.32 122	83.05 187	82.86 157	65.62 198	78.61 111	67.22 129	69.19 119	65.71 144	75.87 144	76.75 199	75.33 198	84.31 203	83.28 184

TAMVS		76.42 169	77.16 161	75.56 175	83.05 172	85.55 175	80.58 176	71.43 176	65.40 187	61.04 173	67.27 130	69.22 135	67.99 178	84.88 163	84.78 162	89.28 183	83.01 185

SCA		79.51 137	80.15 128	78.75 151	86.58 129	87.70 157	83.07 156	68.53 188	81.31 85	66.40 132	73.83 94	75.38 99	79.30 123	80.49 187	79.39 188	88.63 187	82.96 186

TinyColmap		76.73 163	73.95 187	79.96 142	85.16 148	85.64 174	82.34 162	78.19 140	70.63 161	62.06 161	60.69 165	49.61 206	80.81 96	85.12 160	83.69 171	91.22 170	82.27 187

MDTV_nov1_ep13		79.14 142	79.49 138	78.74 152	85.40 142	86.89 164	84.32 149	70.29 181	78.85 108	69.42 119	75.37 86	73.29 117	75.64 145	80.61 186	79.48 187	87.36 191	81.91 188

PatchmatchNet		78.67 148	78.85 142	78.46 156	86.85 128	86.03 168	83.77 152	68.11 191	80.88 93	66.19 133	72.90 101	73.40 116	78.06 130	79.25 193	77.71 193	87.75 190	81.75 189
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
test0.0.03 1		76.03 174	78.51 143	73.12 189	87.47 120	85.13 180	76.32 192	78.05 142	73.19 149	50.98 198	70.64 108	69.28 134	55.53 196	85.33 154	84.38 167	90.39 176	81.63 190

CMPMVS	56.49 17	73.84 188	71.73 194	76.31 172	85.20 146	85.67 173	75.80 193	73.23 170	62.26 193	65.40 138	53.40 192	59.70 174	71.77 167	80.25 188	79.56 186	86.45 197	81.28 191
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
MIMVSNet		74.69 184	75.60 179	73.62 186	76.02 204	85.31 177	81.21 173	67.43 192	71.02 156	59.07 181	54.48 187	64.07 149	66.14 187	86.52 139	86.64 136	91.83 161	81.17 192

MDTV_nov1_ep13_2view		73.21 189	72.91 189	73.56 187	80.01 193	84.28 184	78.62 185	66.43 197	68.64 170	59.12 180	60.39 167	59.69 175	69.81 173	78.82 195	77.43 194	87.36 191	81.11 193

pmnet_mix02		71.95 190	71.83 193	72.10 190	81.40 190	80.63 200	73.78 197	72.85 172	70.90 157	54.89 187	62.17 155	57.42 186	62.92 191	76.80 198	73.98 202	86.74 196	80.87 194

EPMVS		77.53 158	78.07 151	76.90 166	86.89 127	84.91 181	82.18 166	66.64 196	81.00 91	64.11 147	72.75 102	69.68 132	74.42 156	79.36 192	78.13 191	87.14 193	80.68 195

ADS-MVSNet		74.53 185	75.69 178	73.17 188	81.57 189	80.71 199	79.27 183	63.03 205	79.27 106	59.94 177	67.86 127	68.32 140	71.08 169	77.33 197	76.83 195	84.12 205	79.53 196

MDA-MVSNet-bldmvs		66.22 198	64.49 201	68.24 196	61.67 210	82.11 196	70.07 203	76.16 158	59.14 202	47.94 201	54.35 189	35.82 216	67.33 182	64.94 208	75.68 197	86.30 198	79.36 197

FMVSNet5		75.50 181	76.07 171	74.83 180	76.16 202	81.19 197	81.34 169	70.21 182	73.20 148	61.59 167	58.97 175	68.33 139	68.50 176	85.87 147	85.85 151	91.18 171	79.11 198

testgi		71.92 191	74.20 186	69.27 195	84.58 155	83.06 186	73.40 198	74.39 166	64.04 190	46.17 203	68.90 123	57.15 187	48.89 204	84.07 170	83.08 174	88.18 188	79.09 199

MIMVSNet1		65.00 199	66.24 200	63.55 202	58.41 213	80.01 201	69.00 204	74.03 168	55.81 206	41.88 208	36.81 209	49.48 207	47.89 205	81.32 184	82.40 178	90.08 179	77.88 200

test20.03		68.31 196	70.05 197	66.28 200	82.41 182	80.84 198	67.35 205	76.11 159	58.44 203	40.80 210	53.77 191	54.54 197	42.28 207	83.07 175	81.96 182	88.73 186	77.76 201

pmmvs3		61.89 202	61.74 204	62.06 203	64.30 209	70.83 210	64.22 207	52.14 211	48.78 212	44.47 205	41.67 207	41.70 214	63.03 190	76.06 200	76.02 196	84.18 204	77.14 202

new-patchmatchnet		63.80 200	63.31 202	64.37 201	76.49 201	75.99 205	63.73 208	70.99 178	57.27 204	43.08 206	45.86 202	43.80 210	45.13 206	73.20 203	70.68 206	86.80 195	76.34 203

N_pmnet		66.85 197	66.63 198	67.11 199	78.73 196	74.66 207	70.53 202	71.07 177	66.46 179	46.54 202	51.68 196	51.91 204	55.48 197	74.68 202	72.38 203	80.29 208	74.65 204

MVS-HIRNet		68.83 195	66.39 199	71.68 191	77.58 199	75.52 206	66.45 206	65.05 201	62.16 194	62.84 154	44.76 205	56.60 191	71.96 166	78.04 196	75.06 200	86.18 199	72.56 205

new_pmnet		59.28 203	61.47 205	56.73 205	61.66 211	68.29 211	59.57 210	54.91 208	60.83 197	34.38 214	44.66 206	43.65 211	49.90 203	71.66 204	71.56 205	79.94 209	69.67 206

FPMVS		63.63 201	60.08 206	67.78 197	80.01 193	71.50 209	72.88 200	69.41 187	61.82 195	53.11 190	45.12 203	42.11 213	50.86 202	66.69 206	63.84 207	80.41 207	69.46 207

PMVS	50.48 18	55.81 205	51.93 207	60.33 204	72.90 207	49.34 213	48.78 212	69.51 186	43.49 213	54.25 188	36.26 210	41.04 215	39.71 209	65.07 207	60.70 208	76.85 210	67.58 208
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
DeepMVS_CX						48.31 215	48.03 213	26.08 214	56.42 205	25.77 216	47.51 199	31.31 217	51.30 201	48.49 212		53.61 214	61.52 209

test_method		41.78 207	48.10 208	34.42 210	10.74 219	19.78 220	44.64 214	17.73 215	59.83 199	38.67 212	35.82 211	54.41 198	34.94 210	62.87 209	43.13 212	59.81 213	60.82 210

PMMVS2		41.68 208	44.74 210	38.10 207	46.97 216	52.32 212	40.63 216	48.08 212	35.51 214	7.36 220	26.86 212	24.64 218	16.72 214	55.24 211	59.03 209	68.85 212	59.59 211

Gipuma		49.17 206	47.05 209	51.65 206	59.67 212	48.39 214	41.98 215	63.47 204	55.64 207	33.33 215	14.90 213	13.78 220	41.34 208	69.31 205	72.30 204	70.11 211	55.00 212
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
MVE	30.17 19	30.88 210	33.52 211	27.80 213	23.78 218	39.16 216	18.69 220	46.90 213	21.88 217	15.39 217	14.37 215	7.31 223	24.41 213	41.63 213	56.22 210	37.64 218	54.07 213
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
E-PMN		31.40 209	26.80 212	36.78 208	51.39 215	29.96 217	20.20 218	54.17 209	25.93 216	12.75 218	14.73 214	8.58 222	34.10 212	27.36 214	37.83 213	48.07 216	43.18 214

EMVS		30.49 211	25.44 213	36.39 209	51.47 214	29.89 218	20.17 219	54.00 210	26.49 215	12.02 219	13.94 216	8.84 221	34.37 211	25.04 215	34.37 214	46.29 217	39.53 215

test123		0.87 213	1.40 215	0.25 215	0.03 222	0.25 222	0.35 223	0.08 219	1.21 219	0.05 223	2.84 218	0.03 225	0.89 216	0.43 217	1.16 216	0.13 220	3.87 216

testmvs		1.03 212	1.63 214	0.34 214	0.09 221	0.35 221	0.61 222	0.16 217	1.49 218	0.10 222	3.15 217	0.15 224	0.86 217	1.32 216	1.18 215	0.20 219	3.76 217

uanet_test		0.00 214	0.00 216	0.00 216	0.00 223	0.00 223	0.00 224	0.00 220	0.00 220	0.00 224	0.00 219	0.00 226	0.00 219	0.00 218	0.00 217	0.00 221	0.00 218

sosnet-low-res		0.00 214	0.00 216	0.00 216	0.00 223	0.00 223	0.00 224	0.00 220	0.00 220	0.00 224	0.00 219	0.00 226	0.00 219	0.00 218	0.00 217	0.00 221	0.00 218

sosnet		0.00 214	0.00 216	0.00 216	0.00 223	0.00 223	0.00 224	0.00 220	0.00 220	0.00 224	0.00 219	0.00 226	0.00 219	0.00 218	0.00 217	0.00 221	0.00 218

RE-MVS-def										56.08 186

9.14												92.16 16

SR-MVS					96.58 26			90.99 22				92.40 13

our_test_3					81.81 187	83.96 185	76.61 191

MTAPA										92.97 2		91.03 22

MTMP										93.14 1		90.21 30

Patchmatch-RL test							8.55 221

tmp_tt				32.73 211	43.96 217	21.15 219	26.71 217	8.99 216	65.67 184	51.39 196	56.01 184	42.64 212	11.76 215	56.60 210	50.81 211	53.55 215

XVS					93.11 60	96.70 25	91.91 54			83.95 48		88.82 40				95.79 36

X-MVStestdata					93.11 60	96.70 25	91.91 54			83.95 48		88.82 40				95.79 36

mPP-MVS					97.06 12							88.08 45

NP-MVS									87.47 53

Patchmtry						85.54 176	82.30 163	68.23 189		65.37 139

High-resolution multi-view benchmark