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.

SteuartSystems-ACMMP		97.53 1	97.44 1	97.77 5	98.54 14	94.08 6	98.20 13	98.21 11	95.12 5	97.97 2	98.48 3	94.19 1	99.68 15	96.64 2	99.64 2	99.57 1
Skip Steuart: Steuart Systems R&D Blog.
NCCC		97.30 2	97.03 3	98.11 2	98.77 9	95.06 2	97.34 32	98.04 22	95.96 2	97.09 5	97.88 16	93.18 3	99.71 10	95.84 7	99.17 14	99.56 2

CP-MVS		97.02 3	96.81 4	97.64 7	99.33 1	93.54 8	98.80 1	98.28 9	92.99 19	96.45 7	98.30 8	91.90 8	99.85 1	95.61 8	99.68 1	99.54 3

DNCC-MVS		96.95 4	96.60 6	98.01 3	99.03 7	94.93 3	97.72 18	98.10 18	91.50 26	98.01 1	98.32 7	92.33 6	99.58 20	94.85 11	99.51 5	99.53 4

DeepC-MVS_fast	93.89 2	96.93 5	96.64 5	97.78 4	98.64 11	94.30 4	97.41 28	98.04 22	94.81 6	96.59 6	98.37 5	91.24 10	99.64 17	95.16 10	99.52 4	99.42 7
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
PHI-MVS		96.77 6	96.46 8	97.71 6	98.40 18	94.07 7	98.21 12	98.45 5	89.86 42	97.11 4	98.01 14	92.52 5	99.69 13	96.03 4	99.53 3	99.36 8

MVS_111021_HR		96.68 7	96.58 7	96.99 14	98.46 15	92.31 19	96.20 63	98.90 1	94.30 10	95.86 9	97.74 22	92.33 6	99.38 32	96.04 3	99.42 8	99.28 10

DELS-MVS		96.61 8	96.38 9	97.30 10	97.79 32	93.19 10	95.96 68	98.18 12	95.23 4	95.87 8	97.65 24	91.45 9	99.70 12	95.87 5	99.44 7	99.00 19

DeepPCF-MVS	93.97 1	96.61 8	97.09 2	95.15 51	98.09 27	86.63 81	96.00 67	98.15 13	95.43 3	97.95 3	98.56 2	93.40 2	99.36 33	96.77 1	99.48 6	99.45 5

ACMMP		96.27 10	95.93 11	97.28 12	99.24 3	92.62 14	98.25 11	98.81 2	92.99 19	94.56 18	98.39 4	88.96 17	99.85 1	94.57 13	97.63 36	99.36 8
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
MVS_111021_LR		96.24 11	96.19 10	96.39 30	98.23 24	91.35 34	96.24 62	98.79 3	93.99 12	95.80 10	97.65 24	89.92 12	99.24 38	95.87 5	99.20 13	98.58 28

DeepC-MVS	93.07 3	96.06 12	95.66 13	97.29 11	97.96 29	93.17 11	97.30 33	98.06 20	93.92 13	93.38 31	98.66 1	86.83 29	99.73 7	95.60 9	99.22 12	98.96 21
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
CSCG		96.05 13	95.91 12	96.46 28	99.24 3	90.47 49	98.30 8	98.57 4	89.01 53	93.97 26	97.57 28	92.62 4	99.76 5	94.66 12	99.27 11	99.15 12

DP-MVS Recon		95.68 14	95.12 16	97.37 9	99.19 5	94.19 5	97.03 39	98.08 19	88.35 65	95.09 13	97.65 24	89.97 11	99.48 28	92.08 28	98.59 23	98.44 37

MG-MVS		95.61 15	95.38 14	96.31 33	98.42 17	90.53 48	96.04 66	97.48 34	93.47 16	95.67 11	98.10 10	89.17 15	99.25 37	91.27 35	98.77 18	99.13 14

CPTT-MVS		95.57 16	95.19 15	96.70 17	99.27 2	91.48 31	98.33 7	98.11 17	87.79 72	95.17 12	98.03 12	87.09 27	99.61 18	93.51 16	99.42 8	99.02 16

3Dnovator+	91.43 4	95.40 17	94.48 26	98.16 1	96.90 50	95.34 1	98.48 4	97.87 25	94.65 7	88.53 88	98.02 13	83.69 47	99.71 10	93.18 22	98.96 16	99.44 6

PVSNet_Blended_VisFu		95.27 18	94.91 18	96.38 31	98.20 25	90.86 44	97.27 34	98.25 10	90.21 36	94.18 24	97.27 31	87.48 25	99.73 7	93.53 15	97.77 35	98.55 29

Vis-MVSNet		95.23 19	94.81 19	96.51 24	97.18 42	91.58 30	98.26 10	98.12 14	94.38 9	94.90 14	98.15 9	82.28 57	98.92 50	91.45 33	98.58 24	99.01 18
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
EPP-MVSNet		95.22 20	95.04 17	95.76 45	97.49 39	89.56 55	98.67 2	97.00 56	90.69 32	94.24 23	97.62 27	89.79 13	98.81 52	93.39 21	96.49 48	98.92 24

EPNet		95.20 21	94.56 22	97.14 13	92.80 110	92.68 13	97.85 16	94.87 104	96.64 1	92.46 42	97.80 19	86.23 33	99.65 16	93.72 14	98.62 22	99.10 15

3Dnovator	91.36 5	95.19 22	94.44 27	97.44 8	96.56 55	93.36 9	98.65 3	98.36 6	94.12 11	89.25 81	98.06 11	82.20 58	99.77 4	93.41 20	99.32 10	99.18 11

OMC-MVS		95.09 23	94.70 20	96.25 37	98.46 15	91.28 35	96.43 54	97.57 33	92.04 23	94.77 16	97.96 15	87.01 28	99.09 45	91.31 34	96.77 45	98.36 41

PAPM_NR		95.01 24	94.59 21	96.26 36	98.89 8	90.68 46	97.24 35	97.73 27	91.80 24	92.93 39	96.62 49	89.13 16	99.14 41	89.21 47	97.78 34	98.97 20

IS-MVSNet		94.90 25	94.52 24	96.05 40	97.67 35	90.56 47	98.44 5	96.22 78	93.21 17	93.99 25	97.74 22	85.55 37	98.45 63	89.98 41	97.86 32	99.14 13

PVSNet_Blended		94.87 26	94.56 22	95.81 44	98.27 22	89.46 57	95.47 77	98.36 6	88.84 57	94.36 20	96.09 60	88.02 21	99.58 20	93.44 18	98.18 28	98.40 39

API-MVS		94.84 27	94.49 25	95.90 42	97.90 31	92.00 24	97.80 17	97.48 34	89.19 50	94.81 15	96.71 43	88.84 18	99.17 40	88.91 50	98.76 19	96.53 76

WTY-MVS		94.71 28	94.02 28	96.79 16	97.71 34	92.05 23	96.59 50	97.35 46	90.61 35	94.64 17	96.93 40	86.41 32	99.39 31	91.20 37	94.71 60	98.94 22

sss		94.51 29	93.80 31	96.64 18	97.07 46	91.97 25	96.32 58	98.06 20	88.94 56	94.50 19	96.78 42	84.60 42	99.27 36	91.90 29	96.02 51	98.68 27

AdaColmap		94.34 30	93.68 32	96.31 33	98.59 12	91.68 27	96.59 50	97.81 26	89.87 41	92.15 48	97.06 39	83.62 48	99.54 24	89.34 45	98.07 30	97.70 59

CNLPA		94.28 31	93.53 34	96.52 22	98.38 20	92.55 16	96.59 50	96.88 64	90.13 38	91.91 51	97.24 32	85.21 38	99.09 45	87.64 56	97.83 33	97.92 51

MAR-MVS		94.22 32	93.46 35	96.51 24	98.00 28	92.19 22	97.67 19	97.47 36	88.13 68	93.00 34	95.84 65	84.86 41	99.51 26	87.99 54	98.17 29	97.83 55
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
PAPR		94.18 33	93.42 38	96.48 26	97.64 36	91.42 33	95.55 74	97.71 29	88.99 54	92.34 47	95.82 66	89.19 14	99.11 43	86.14 66	97.38 40	98.90 25

Vis-MVSNet (Re-imp)		94.15 34	93.88 30	94.95 56	97.61 37	87.92 69	98.10 14	95.80 84	92.22 22	93.02 33	97.45 30	84.53 44	97.91 88	88.24 52	97.97 31	99.02 16

CDS-MVSNet		94.14 35	93.54 33	95.93 41	96.18 67	91.46 32	96.33 57	97.04 55	88.97 55	93.56 27	96.51 51	87.55 24	97.89 89	89.80 43	95.95 52	98.44 37
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
PLC	91.00 6	94.11 36	93.43 37	96.13 39	98.58 13	91.15 40	96.69 46	97.39 42	87.29 77	91.37 58	96.71 43	88.39 20	99.52 25	87.33 58	97.13 43	97.73 57
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
PVSNet_BlendedMVS		94.06 37	93.92 29	94.47 63	98.27 22	89.46 57	96.73 44	98.36 6	90.17 37	94.36 20	95.24 73	88.02 21	99.58 20	93.44 18	90.72 88	94.36 103

UGNet		94.04 38	93.28 39	96.31 33	96.85 51	91.19 36	97.88 15	97.68 30	94.40 8	93.00 34	96.18 58	73.39 99	99.61 18	91.72 31	98.46 25	98.13 45
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
TAMVS		94.01 39	93.46 35	95.64 46	96.16 69	90.45 50	96.71 45	96.89 63	89.27 49	93.46 30	96.92 41	87.29 26	97.94 85	88.70 51	95.74 54	98.53 30

114514_t		93.95 40	93.06 40	96.63 20	99.07 6	91.61 28	97.46 27	97.96 24	77.99 114	93.00 34	97.57 28	86.14 34	99.33 34	89.22 46	99.15 15	98.94 22

HY-MVS	89.66 9	93.87 41	92.95 42	96.63 20	97.10 45	92.49 18	95.64 72	96.64 70	89.05 52	93.00 34	95.79 68	85.77 36	99.45 30	89.16 49	94.35 61	97.96 50

F-COLMAP		93.58 42	92.98 41	95.37 49	98.40 18	88.98 62	97.18 37	97.29 47	87.75 73	90.49 62	97.10 38	85.21 38	99.50 27	86.70 63	96.72 46	97.63 60

ab-mvs		93.57 43	92.55 47	96.64 18	97.28 40	91.96 26	95.40 78	97.45 40	89.81 44	93.22 32	96.28 55	79.62 72	99.46 29	90.74 38	93.11 70	98.50 33

LS3D		93.57 43	92.61 45	96.47 27	97.59 38	91.61 28	97.67 19	97.72 28	85.17 85	90.29 63	98.34 6	84.60 42	99.73 7	83.85 83	98.27 27	98.06 49

QAPM		93.45 45	92.27 52	96.98 15	96.77 53	92.62 14	98.39 6	98.12 14	84.50 91	88.27 90	97.77 20	82.39 56	99.81 3	85.40 72	98.81 17	98.51 32

1112_ss		93.37 46	92.42 50	96.21 38	97.05 47	90.99 41	96.31 59	96.72 67	86.87 79	89.83 71	96.69 45	86.51 31	99.14 41	88.12 53	93.67 65	98.50 33

MVSTER		93.20 47	92.81 43	94.37 65	96.56 55	89.59 54	97.06 38	97.12 51	91.24 31	91.30 59	95.96 61	82.02 60	98.05 82	93.48 17	90.55 89	95.47 84

CLD-MVS		92.98 48	92.53 48	94.32 67	96.12 72	89.20 60	95.28 80	97.47 36	92.66 21	89.90 70	95.62 70	80.58 65	98.40 64	92.73 23	92.40 73	95.38 86
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
ACMM	89.79 8	92.96 49	92.50 49	94.35 66	96.30 64	88.71 64	97.58 22	97.36 45	91.40 28	90.53 61	96.65 47	79.77 70	98.75 54	91.24 36	91.64 82	95.59 83
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
LPG-MVS_test		92.94 50	92.56 46	94.10 70	96.16 69	88.26 65	97.65 21	97.46 38	91.29 29	90.12 67	97.16 35	79.05 75	98.73 55	92.25 24	91.89 79	95.31 89

BH-untuned		92.94 50	92.62 44	93.92 75	97.22 41	86.16 83	96.40 55	96.25 77	90.06 39	89.79 72	96.17 59	83.19 50	98.35 67	87.19 60	97.27 42	97.24 68

PatchMatch-RL		92.90 52	92.02 54	95.56 47	98.19 26	90.80 45	95.27 81	97.18 50	87.96 69	91.86 53	95.68 69	80.44 66	98.99 48	84.01 81	97.54 38	96.89 71

PMMVS		92.86 53	92.34 51	94.42 64	94.92 91	86.73 79	94.53 84	96.38 74	84.78 89	94.27 22	95.12 75	83.13 51	98.40 64	91.47 32	96.49 48	98.12 46

OpenMVS	89.19 12	92.86 53	91.68 60	96.40 29	95.34 80	92.73 12	98.27 9	98.12 14	84.86 88	85.78 101	97.75 21	78.89 77	99.74 6	87.50 57	98.65 21	96.73 73

Test_1112_low_res		92.84 55	91.84 57	95.85 43	97.04 48	89.97 51	95.53 75	96.64 70	85.38 84	89.65 73	95.18 74	85.86 35	99.10 44	87.70 55	93.58 69	98.49 35

DP-MVS		92.76 56	91.51 61	96.52 22	98.77 9	90.99 41	97.38 30	96.08 80	82.38 101	89.29 79	97.87 17	83.77 46	99.69 13	81.37 95	96.69 47	98.89 26

BH-RMVSNet		92.72 57	91.97 55	94.97 54	97.16 43	87.99 68	96.15 64	95.60 89	90.62 34	91.87 52	97.15 37	78.41 81	98.57 59	83.16 84	97.60 37	98.36 41

ACMP	89.59 10	92.62 58	92.14 53	94.05 72	96.40 62	88.20 67	97.36 31	97.25 49	91.52 25	88.30 89	96.64 48	78.46 80	98.72 57	91.86 30	91.48 84	95.23 95
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
TAPA-MVS	90.10 7	92.30 59	91.22 66	95.56 47	98.33 21	89.60 53	96.79 43	97.65 31	81.83 103	91.52 55	97.23 33	87.94 23	98.91 51	71.31 115	98.37 26	98.17 44
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
IterMVS-LS		92.29 60	91.94 56	93.34 82	96.25 65	86.97 78	96.57 53	97.05 54	90.67 33	89.50 74	94.80 80	86.59 30	97.64 93	89.91 42	86.11 111	95.40 85
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
PVSNet	86.66 18	92.24 61	91.74 59	93.73 78	97.77 33	83.69 97	92.88 104	96.72 67	87.91 70	93.00 34	94.86 78	78.51 79	99.05 47	86.53 64	97.45 39	98.47 36

XXY-MVS		92.16 62	91.23 65	94.95 56	94.75 97	90.94 43	97.47 26	97.43 41	89.14 51	88.90 83	96.43 53	79.71 71	98.24 72	89.56 44	87.68 105	95.67 82

BH-w/o		92.14 63	91.75 58	93.31 83	96.99 49	85.73 84	95.67 70	95.69 86	88.73 62	89.26 80	94.82 79	82.97 52	98.07 81	85.26 73	96.32 50	96.13 80

PatchmatchNet		91.91 64	91.35 62	93.59 80	95.38 78	84.11 93	93.15 101	95.39 91	89.54 46	92.10 49	93.68 92	82.82 53	98.13 77	84.81 77	95.32 56	98.52 31
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
FMVSNet3		91.78 65	90.69 70	95.03 52	96.53 57	92.27 20	97.02 40	96.93 59	89.79 45	89.35 76	94.65 84	77.01 85	97.47 96	86.12 67	88.82 97	95.35 87

EPNet_dtu		91.71 66	91.28 64	92.99 86	93.76 106	83.71 96	96.69 46	95.28 96	93.15 18	87.02 98	95.95 62	83.37 49	97.38 100	79.46 100	96.84 44	97.88 54

PCF-MVS	89.48 11	91.56 67	89.95 79	96.36 32	96.60 54	92.52 17	92.51 107	97.26 48	79.41 110	88.90 83	96.56 50	84.04 45	99.55 23	77.01 107	97.30 41	97.01 69
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
PAPM		91.52 68	90.30 73	95.20 50	95.30 82	89.83 52	93.38 100	96.85 65	86.26 82	88.59 87	95.80 67	84.88 40	98.15 76	75.67 109	95.93 53	97.63 60

TR-MVS		91.48 69	90.59 71	94.16 69	96.40 62	87.33 73	95.67 70	95.34 95	87.68 74	91.46 56	95.52 72	76.77 86	98.35 67	82.85 87	93.61 68	96.79 72

tpmrst		91.44 70	91.32 63	91.79 97	95.15 87	79.20 113	93.42 99	95.37 92	88.55 63	93.49 29	93.67 93	82.49 54	98.27 71	90.41 40	89.34 95	97.90 52

MSDG		91.42 71	90.24 76	94.96 55	97.15 44	88.91 63	93.69 95	96.32 76	85.72 83	86.93 99	96.47 52	80.24 67	98.98 49	80.57 97	95.05 58	96.98 70

GBi-Net		91.35 72	90.27 74	94.59 59	96.51 58	91.18 37	97.50 23	96.93 59	88.82 59	89.35 76	94.51 85	73.87 95	97.29 101	86.12 67	88.82 97	95.31 89

test1		91.35 72	90.27 74	94.59 59	96.51 58	91.18 37	97.50 23	96.93 59	88.82 59	89.35 76	94.51 85	73.87 95	97.29 101	86.12 67	88.82 97	95.31 89

FMVSNet2		91.31 74	90.08 77	94.99 53	96.51 58	92.21 21	97.41 28	96.95 57	88.82 59	88.62 85	94.75 81	73.87 95	97.42 98	85.20 74	88.55 102	95.35 87

cascas		91.20 75	90.08 77	94.58 62	94.97 90	89.16 61	93.65 97	97.59 32	79.90 109	89.40 75	92.92 100	75.36 91	98.36 66	92.14 26	94.75 59	96.23 78

CostFormer		91.18 76	90.70 69	92.62 89	94.84 94	81.76 103	94.09 90	94.43 107	84.15 92	92.72 41	93.77 91	79.43 73	98.20 73	90.70 39	92.18 75	97.90 52

HyFIR		91.07 77	90.50 72	92.77 88	96.08 73	83.37 98	91.52 109	97.09 52	84.77 90	89.13 82	90.06 112	82.20 58	96.76 109	84.95 75	98.67 20	98.18 43

LTVRE_ROB	88.41 13	90.99 78	89.92 80	94.19 68	96.18 67	89.55 56	96.31 59	97.09 52	87.88 71	85.67 102	95.91 63	78.79 78	98.57 59	81.50 94	89.98 91	94.44 102
Andreas Kuhn, Heiko Hirschmüller, Daniel Scharstein, Helmut Mayer: A TV Prior for High-Quality Scalable Multi-View Stereo Reconstruction. International Journal of Computer Vision 2016
CR-MVSNet		90.82 79	89.77 82	93.95 73	94.45 101	87.19 75	90.23 114	95.68 87	86.89 78	92.40 43	92.36 105	80.91 63	97.05 105	81.09 96	93.95 63	97.60 62

RPSCF		90.75 80	90.86 67	90.42 110	96.84 52	76.29 115	95.61 73	96.34 75	83.89 95	91.38 57	97.87 17	76.45 87	98.78 53	87.16 61	92.23 74	96.20 79

EPMVS		90.70 81	89.81 81	93.37 81	94.73 98	84.21 92	93.67 96	88.02 123	89.50 47	92.38 45	93.49 94	77.82 83	97.78 90	86.03 70	92.68 72	98.11 48

ACMH	87.59 16	90.53 82	89.42 85	93.87 76	96.21 66	87.92 69	97.24 35	96.94 58	88.45 64	83.91 107	96.27 56	71.92 101	98.62 58	84.43 79	89.43 94	95.05 96
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
COLMAP_ROB	87.81 15	90.40 83	89.28 86	93.79 77	97.95 30	87.13 77	96.92 42	95.89 83	82.83 100	86.88 100	97.18 34	73.77 98	99.29 35	78.44 104	93.62 67	94.95 97
Johannes L. Schönberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
tpm		90.25 84	89.74 83	91.76 99	93.92 105	79.73 111	93.98 91	93.54 115	88.28 66	91.99 50	93.25 98	77.51 84	97.44 97	87.30 59	87.94 104	98.12 46

ACMH+	87.92 14	90.20 85	89.18 88	93.25 84	96.48 61	86.45 82	96.99 41	96.68 69	88.83 58	84.79 104	96.22 57	70.16 105	98.53 61	84.42 80	88.04 103	94.77 100

IterMVS		90.15 86	89.67 84	91.61 101	95.48 76	83.72 95	94.33 86	96.12 79	89.99 40	87.31 96	94.15 88	75.78 90	96.27 112	86.97 62	86.89 110	94.83 99
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
tpm2		89.96 87	89.21 87	92.23 93	94.91 92	81.25 106	93.78 94	94.42 108	80.62 108	91.56 54	93.44 96	76.44 88	97.94 85	85.60 71	92.08 77	97.49 66

IB-MVS	87.33 17	89.91 88	88.28 91	94.79 58	95.26 84	87.70 72	95.12 82	93.95 114	89.35 48	87.03 97	92.49 103	70.74 104	99.19 39	89.18 48	81.37 119	97.49 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
FMVSNet1		89.88 89	88.31 90	94.59 59	95.41 77	91.18 37	97.50 23	96.93 59	86.62 80	87.41 95	94.51 85	65.94 110	97.29 101	83.04 85	87.43 107	95.31 89

tpmp4_e23		89.58 90	88.59 89	92.54 90	95.16 86	81.53 104	94.11 89	95.09 100	81.66 104	88.60 86	93.44 96	75.11 92	98.33 70	82.45 91	91.72 81	97.75 56

USDC		88.94 91	87.83 93	92.27 92	94.66 99	84.96 89	93.86 93	95.90 82	87.34 76	83.40 109	95.56 71	67.43 108	98.19 74	82.64 90	89.67 93	93.66 107

dp		88.90 92	88.26 92	90.81 107	94.58 100	76.62 114	92.85 105	94.93 102	85.12 86	90.07 69	93.07 99	75.81 89	98.12 79	80.53 98	87.42 108	97.71 58

PatchT		88.87 93	87.42 94	93.22 85	94.08 104	85.10 87	89.51 117	94.64 105	81.92 102	92.36 46	88.15 117	80.05 69	97.01 108	72.43 112	93.65 66	97.54 65

Patchmtry		88.64 94	87.25 95	92.78 87	94.09 103	86.64 80	89.82 116	95.68 87	80.81 107	87.63 93	92.36 105	80.91 63	97.03 107	78.86 102	85.12 113	94.67 101

RPMNet		88.52 95	86.72 100	93.95 73	94.45 101	87.19 75	90.23 114	94.99 101	77.87 116	92.40 43	87.55 119	80.17 68	97.05 105	68.84 117	93.95 63	97.60 62

MIMVSNet		88.50 96	86.76 99	93.72 79	94.84 94	87.77 71	91.39 110	94.05 111	86.41 81	87.99 91	92.59 102	63.27 114	95.82 115	77.44 105	92.84 71	97.57 64

tpm cat1		88.36 97	87.21 97	91.81 96	95.13 88	80.55 107	92.58 106	95.70 85	74.97 119	87.45 94	91.96 108	78.01 82	98.17 75	80.39 99	88.74 100	96.72 74

JIA-IIPM		88.26 98	87.04 98	91.91 94	93.52 107	81.42 105	89.38 118	94.38 109	80.84 106	90.93 60	80.74 122	79.22 74	97.92 87	82.76 88	91.62 83	96.38 77

LF4IMVS		87.94 99	87.25 95	89.98 112	92.38 114	80.05 110	94.38 85	95.25 99	87.59 75	84.34 105	94.74 82	64.31 113	97.66 92	84.83 76	87.45 106	92.23 114

FMVSNet5		87.29 100	85.79 102	91.78 98	94.80 96	87.28 74	95.49 76	95.28 96	84.09 93	83.85 108	91.82 109	62.95 115	94.17 118	78.48 103	85.34 112	93.91 106

EG-PatchMatch MVS		87.02 101	85.44 103	91.76 99	92.67 112	85.00 88	96.08 65	96.45 73	83.41 99	79.52 118	93.49 94	57.10 120	97.72 91	79.34 101	90.87 87	92.56 111

TinyColmap		86.82 102	85.35 104	91.21 103	94.91 92	82.99 99	93.94 92	94.02 113	83.58 98	81.56 113	94.68 83	62.34 116	98.13 77	75.78 108	87.35 109	92.52 112

TDRefinement		86.53 103	84.76 107	91.85 95	82.23 123	84.25 91	96.38 56	95.35 93	84.97 87	84.09 106	94.94 76	65.76 111	98.34 69	84.60 78	74.52 121	92.97 110

test_0402		86.46 104	84.79 106	91.45 102	95.02 89	85.55 85	96.29 61	94.89 103	80.90 105	82.21 111	93.97 89	68.21 107	97.29 101	62.98 120	88.68 101	91.51 117

DSMNet-mixed		86.34 105	86.12 101	87.00 115	89.88 118	70.43 119	94.93 83	90.08 121	77.97 115	85.42 103	92.78 101	74.44 94	93.96 119	74.43 110	95.14 57	96.62 75

UnsupCasMVSNet_eth		85.99 106	84.45 108	90.62 108	89.97 117	82.40 101	93.62 98	97.37 44	89.86 42	78.59 119	92.37 104	65.25 112	95.35 116	82.27 93	70.75 124	94.10 104

CMPMVS	62.92 21	85.62 107	84.92 105	87.74 114	89.14 121	73.12 118	94.17 87	96.80 66	73.98 120	73.65 123	94.93 77	66.36 109	97.61 95	83.95 82	91.28 85	92.48 113
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
PVSNet_0	82.17 19	85.46 108	83.64 109	90.92 105	95.27 83	79.49 112	90.55 113	95.60 89	83.76 97	83.00 110	89.95 113	71.09 102	97.97 83	82.75 89	60.79 126	95.31 89

MIMVSNet1		84.93 109	83.05 110	90.56 109	89.56 120	84.84 90	95.40 78	95.35 93	83.91 94	80.38 116	92.21 107	57.23 119	93.34 120	70.69 116	82.75 117	93.50 108

OpenMVS_ROB	81.14 20	84.42 110	82.28 111	90.83 106	90.06 116	84.05 94	95.73 69	94.04 112	73.89 121	80.17 117	91.53 111	59.15 118	97.64 93	66.92 119	89.05 96	90.80 118

LP		84.13 111	81.85 112	90.97 104	93.20 108	82.12 102	87.68 120	94.27 110	76.80 117	81.93 112	88.52 114	72.97 100	95.95 114	59.53 123	81.73 118	94.84 98

new_pmnet		82.89 112	81.12 114	88.18 113	89.63 119	80.18 109	91.77 108	92.57 117	76.79 118	75.56 122	88.23 116	61.22 117	94.48 117	71.43 114	82.92 115	89.87 119

MVS-HIRNet		82.47 113	81.21 113	86.26 116	95.38 78	69.21 122	88.96 119	89.49 122	66.28 124	80.79 115	74.08 124	68.48 106	97.39 99	71.93 113	95.47 55	92.18 115

UnsupCasMVSNet_bld		82.13 114	79.46 115	90.14 111	88.00 122	82.47 100	90.89 112	96.62 72	78.94 111	75.61 121	84.40 121	56.63 121	96.31 111	77.30 106	66.77 125	91.63 116

HyFIR lowres test		81.01 115	79.31 116	86.13 117	93.19 109	74.03 116	86.45 122	93.06 116	68.57 123	76.44 120	87.03 120	43.72 124	96.27 112	73.27 111	92.05 78	88.09 120

N_pmnet		78.73 116	78.71 117	78.79 119	92.80 110	46.50 130	94.14 88	43.71 130	78.61 112	80.83 114	91.66 110	74.94 93	96.36 110	67.24 118	84.45 114	93.50 108

FPMVS		71.27 117	69.85 118	75.50 121	74.64 124	59.03 126	91.30 111	91.50 118	58.80 126	57.92 125	88.28 115	29.98 127	85.53 124	53.43 125	82.84 116	81.95 122

no-one		68.12 118	63.78 120	81.13 118	74.01 125	70.22 121	87.61 121	90.71 120	72.63 122	53.13 127	71.89 125	30.29 126	91.45 121	61.53 122	32.21 130	81.72 123

Gipuma		67.86 119	65.41 119	75.18 122	92.66 113	73.45 117	66.50 128	94.52 106	53.33 127	57.80 126	66.07 128	30.81 125	89.20 122	48.15 128	78.88 120	62.90 129
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
ANet_high		63.94 120	59.58 121	77.02 120	61.24 130	66.06 123	85.66 123	87.93 124	78.53 113	42.94 128	71.04 126	25.42 128	80.71 126	52.60 126	30.83 131	84.28 121

PNet_i23d		59.01 121	55.87 122	68.44 125	73.98 126	51.37 128	81.36 124	82.41 127	52.37 128	42.49 129	70.39 127	11.39 129	79.99 128	49.77 127	38.71 128	73.97 126

PMVS	53.92 22	58.58 122	55.40 123	68.12 126	51.00 131	48.64 129	78.86 126	87.10 125	46.77 130	35.84 131	74.28 123	8.76 130	86.34 123	42.07 129	73.91 122	69.38 127
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
wuykxyi23d		56.92 123	51.11 124	74.38 124	62.30 129	61.47 125	80.09 125	84.87 126	49.62 129	30.80 132	57.20 130	7.03 131	82.94 125	55.69 124	32.36 129	78.72 125

MVE	50.73 23	53.25 124	48.81 125	66.58 127	65.34 128	57.50 127	72.49 127	70.94 129	40.15 131	39.28 130	63.51 129	6.89 133	73.48 129	38.29 130	42.38 127	68.76 128
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
wuyk23d		25.11 125	24.57 126	26.74 128	73.98 126	39.89 131	57.88 129	9.80 131	12.27 132	10.39 133	6.97 131	7.03 131	36.44 130	25.43 131	17.39 132	3.89 130

ab-mvs-re		8.06 126	10.74 127	0.00 129	0.00 132	0.00 132	0.00 130	0.00 132	0.00 133	0.00 134	96.69 45	0.00 134	0.00 131	0.00 132	0.00 133	0.00 131

DeepMVS_CX				74.68 123	90.84 115	64.34 124		81.61 128	65.34 125	67.47 124	88.01 118	48.60 123	80.13 127	62.33 121	73.68 123	79.58 124

HQP-MVS												51.25 122

door								91.13 119

HQP5-MVS						89.33 59

HQP-NCC					95.86 74		96.65 48		93.55 14	90.14 64

ITE_SJBPF				92.43 91	95.34 80	85.37 86		95.92 81	91.47 27	87.75 92	96.39 54	71.00 103	97.96 84	82.36 92	89.86 92	93.97 105

ACMP_Plane					95.86 74		96.65 48		93.55 14	90.14 64

BP-MVS														92.13 27

ACMMP++															91.02 86

HQP4-MVS										90.14 64			98.50 62			95.78 81

HQP3-MVS								97.39 42							92.10 76

HQP2-MVS												80.95 62

ACMMP++_ref															90.30 90

Test By Simon												88.73 19

MDTV_nov1_ep13			90.76 68		95.22 85	80.33 108	93.03 103	95.28 96	88.14 67	92.84 40	93.83 90	81.34 61	98.08 80	82.86 86	94.34 62

NP-MVS											95.87 64

MDTV_nov1_ep13_2view						70.35 120	93.10 102		83.88 96	93.55 28		82.47 55		86.25 65		98.38 40

LGP-MVS_train				94.10 70	96.16 69	88.26 65		97.46 38	91.29 29	90.12 67	97.16 35	79.05 75	98.73 55	92.25 24	91.89 79	95.31 89

High-resolution multi-view benchmark