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		88.16 1	88.22 2	87.98 3	92.00 15	72.76 5	92.99 3	93.57 4	79.38 4	87.27 3	94.47 2	71.72 7	96.63 5	87.17 1	95.44 3	94.56 4
Skip Steuart: Steuart Systems R&D Blog.
DeepPCF-MVS	80.84 1	88.10 2	88.56 1	86.73 11	92.24 14	69.03 23	89.57 25	93.39 7	77.53 7	89.79 1	94.12 5	78.98 1	96.58 6	85.66 2	95.72 2	94.58 3

NCCC		88.06 3	88.01 3	88.24 2	94.41 2	73.62 2	91.22 12	92.83 10	81.50 1	85.79 4	93.47 7	73.02 5	97.00 2	84.90 3	94.94 7	94.10 8

DNCC-MVS		87.44 4	87.52 4	87.19 9	94.24 3	72.39 9	91.86 8	92.83 10	73.01 31	88.58 2	94.52 1	73.36 4	96.49 7	84.26 5	95.01 6	92.70 19

CP-MVS		87.11 5	86.92 5	87.68 6	94.20 4	73.86 1	93.98 1	92.82 12	76.62 12	83.68 7	94.46 3	67.93 12	95.95 11	84.20 6	94.39 8	93.23 14

DeepC-MVS	79.81 2	87.08 6	86.88 6	87.69 5	91.16 19	72.32 11	90.31 18	93.94 2	77.12 9	82.82 10	94.23 4	72.13 6	97.09 1	84.83 4	95.37 4	93.65 10
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
DeepC-MVS_fast	79.65 3	86.91 7	86.62 7	87.76 4	93.52 7	72.37 10	91.26 11	93.04 9	76.62 12	84.22 6	93.36 9	71.44 8	96.76 3	80.82 11	95.33 5	94.16 6
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
PHI-MVS		86.43 8	86.17 9	87.24 8	90.88 22	70.96 15	92.27 6	94.07 1	72.45 32	85.22 5	91.90 17	69.47 10	96.42 8	83.28 7	95.94 1	94.35 5

CSCG		86.41 9	86.19 8	87.07 10	92.91 11	72.48 8	90.81 15	93.56 5	73.95 28	83.16 9	91.07 25	75.94 2	95.19 16	79.94 14	94.38 9	93.55 11

ACMMP		85.89 10	85.39 10	87.38 7	93.59 6	72.63 7	92.74 4	93.18 8	76.78 11	80.73 14	93.82 6	64.33 24	96.29 9	82.67 8	90.69 23	93.23 14
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
3Dnovator+	77.84 4	85.48 11	84.47 13	88.51 1	91.08 20	73.49 3	93.18 2	93.78 3	80.79 3	76.66 41	93.37 8	60.40 46	96.75 4	77.20 19	93.73 12	95.29 1

DELS-MVS		85.41 12	85.30 11	85.77 16	88.49 52	67.93 34	85.52 59	93.44 6	78.70 6	83.63 8	89.03 42	74.57 3	95.71 13	80.26 13	94.04 10	93.66 9

MVS_111021_HR		85.14 13	84.75 12	86.32 13	91.65 17	72.70 6	85.98 55	90.33 35	76.11 17	82.08 11	91.61 20	71.36 9	94.17 39	81.02 9	92.58 14	92.08 25

CPTT-MVS		83.73 14	83.33 15	84.92 22	93.28 9	70.86 17	92.09 7	90.38 34	68.75 56	79.57 17	92.83 11	60.60 45	93.04 54	80.92 10	91.56 19	90.86 31

EPNet		83.72 15	82.92 17	86.14 15	84.22 80	69.48 21	91.05 14	85.27 70	81.30 2	76.83 39	91.65 18	66.09 18	95.56 14	76.00 21	93.85 11	93.38 12

Vis-MVSNet		83.46 16	82.80 19	85.43 17	90.25 27	68.74 29	90.30 19	90.13 36	76.33 16	80.87 13	92.89 10	61.00 43	94.20 38	72.45 30	90.97 21	93.35 13
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
MG-MVS		83.41 17	83.45 14	83.28 42	92.74 12	62.28 74	88.17 42	89.50 38	75.22 21	81.49 12	92.74 13	66.75 15	95.11 18	72.85 27	91.58 18	92.45 21

EPP-MVSNet		83.40 18	83.02 16	84.57 25	90.13 28	64.47 61	92.32 5	90.73 33	74.45 25	79.35 19	91.10 23	69.05 11	95.12 17	72.78 28	87.22 37	94.13 7

3Dnovator	76.31 5	83.38 19	82.31 21	86.59 12	87.94 59	72.94 4	90.64 16	92.14 16	77.21 8	75.47 48	92.83 11	58.56 48	94.72 29	73.24 26	92.71 13	92.13 24

IS-MVSNet		83.15 20	82.81 18	84.18 33	89.94 29	63.30 65	91.59 9	88.46 55	79.04 5	79.49 18	92.16 14	65.10 23	94.28 35	67.71 37	91.86 16	94.95 2

DP-MVS Recon		83.11 21	82.09 23	86.15 14	94.44 1	70.92 16	88.79 32	92.20 14	70.53 41	79.17 20	91.03 27	64.12 26	96.03 10	68.39 36	90.14 25	91.50 27

PAPM_NR		83.02 22	82.41 20	84.82 24	92.47 13	66.37 46	87.93 45	91.80 22	73.82 29	77.32 32	90.66 30	67.90 13	94.90 25	70.37 33	89.48 28	93.19 16

OMC-MVS		82.69 23	81.97 24	84.85 23	88.75 46	67.42 38	87.98 44	90.87 32	74.92 22	79.72 16	91.65 18	62.19 37	93.96 40	75.26 22	86.42 46	93.16 17

PVSNet_Blended_VisFu		82.62 24	81.83 25	84.96 21	90.80 23	69.76 20	88.74 33	91.70 24	69.39 50	78.96 22	88.46 47	65.47 21	94.87 27	74.42 24	88.57 31	90.24 41

MVS_111021_LR		82.61 25	82.11 22	84.11 34	88.82 44	71.58 12	85.15 60	86.16 68	74.69 23	80.47 15	91.04 26	62.29 35	90.55 74	80.33 12	90.08 26	90.20 42

CLD-MVS		82.31 26	81.65 26	84.29 30	88.47 53	67.73 37	85.81 58	92.35 13	75.78 18	78.33 26	86.58 70	64.01 27	94.35 34	76.05 20	87.48 35	90.79 32
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
LPG-MVS_test		82.08 27	81.27 27	84.50 27	89.23 38	68.76 27	90.22 20	91.94 20	75.37 19	76.64 42	91.51 21	54.29 63	94.91 23	78.44 15	83.78 51	89.83 49

API-MVS		81.99 28	81.23 28	84.26 31	90.94 21	70.18 18	91.10 13	89.32 41	71.51 38	78.66 24	88.28 50	65.26 22	95.10 20	64.74 46	91.23 20	87.51 73

MAR-MVS		81.84 29	80.70 30	85.27 18	91.32 18	71.53 13	89.82 23	90.92 31	69.77 48	78.50 25	86.21 73	62.36 34	94.52 33	65.36 42	92.05 15	89.77 51
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		81.66 30	80.89 29	83.99 37	90.27 26	64.00 63	86.76 50	91.77 23	68.84 55	77.13 38	89.50 39	67.63 14	94.88 26	67.55 38	88.52 32	93.09 18

ACMP	74.13 6	81.51 31	80.57 31	84.36 29	89.42 31	68.69 30	89.97 22	91.50 27	74.46 24	75.04 59	90.41 32	53.82 68	94.54 31	77.56 17	82.91 54	89.86 48
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
PVSNet_Blended		80.98 32	80.34 32	82.90 48	88.85 42	65.40 54	84.43 67	92.00 17	67.62 59	78.11 28	85.05 80	66.02 19	94.27 36	71.52 31	89.50 27	89.01 58

QAPM		80.88 33	79.50 37	85.03 19	88.01 58	68.97 25	91.59 9	92.00 17	66.63 63	75.15 57	92.16 14	57.70 50	95.45 15	63.52 47	88.76 30	90.66 34

UGNet		80.83 34	79.59 35	84.54 26	88.04 57	68.09 33	89.42 26	88.16 56	76.95 10	76.22 45	89.46 40	49.30 85	93.94 42	68.48 35	90.31 24	91.60 26
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
ACMM	73.20 8	80.78 35	79.84 34	83.58 38	89.31 35	68.37 31	89.99 21	91.60 25	70.28 44	77.25 33	89.66 37	53.37 71	93.53 47	74.24 25	82.85 55	88.85 61
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
114514_t		80.68 36	79.51 36	84.20 32	94.09 5	67.27 42	89.64 24	91.11 29	58.75 104	74.08 64	90.72 29	58.10 49	95.04 22	69.70 34	89.42 29	90.30 40

PVSNet_BlendedMVS		80.60 37	80.02 33	82.36 52	88.85 42	65.40 54	86.16 53	92.00 17	69.34 51	78.11 28	86.09 74	66.02 19	94.27 36	71.52 31	82.06 61	87.39 75

AdaColmap		80.58 38	79.42 38	84.06 36	93.09 10	68.91 26	89.36 27	88.97 51	69.27 52	75.70 47	89.69 36	57.20 53	95.77 12	63.06 48	88.41 33	87.50 74

PCF-MVS	73.52 7	80.38 39	78.84 41	85.01 20	87.71 61	68.99 24	83.65 72	91.46 28	63.00 82	77.77 31	90.28 33	66.10 17	95.09 21	61.40 58	88.22 34	90.94 30
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
IterMVS-LS		80.06 40	79.38 39	82.11 53	85.89 73	63.20 67	86.79 49	89.34 40	74.19 26	75.45 49	86.72 64	66.62 16	92.39 60	72.58 29	76.86 89	90.75 33
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
OpenMVS	72.83 10	79.77 41	78.33 45	84.09 35	85.17 75	69.91 19	90.57 17	90.97 30	66.70 62	72.17 73	91.91 16	54.70 61	93.96 40	61.81 57	90.95 22	88.41 65

BH-RMVSNet		79.61 42	78.44 44	83.14 44	89.38 32	65.93 49	84.95 62	87.15 62	73.56 30	78.19 27	89.79 35	56.67 54	93.36 48	59.53 69	86.74 43	90.13 43

ab-mvs		79.51 43	78.97 40	81.14 65	88.46 54	60.91 79	83.84 70	89.24 45	70.36 43	79.03 21	88.87 44	63.23 30	90.21 76	65.12 43	82.57 59	92.28 22

BH-untuned		79.47 44	78.60 42	82.05 54	89.19 40	65.91 50	86.07 54	88.52 54	72.18 33	75.42 50	87.69 54	61.15 42	93.54 46	60.38 65	86.83 42	86.70 87

TAPA-MVS	73.13 9	79.15 45	77.94 46	82.79 49	89.59 30	62.99 71	88.16 43	91.51 26	65.77 68	77.14 37	91.09 24	60.91 44	93.21 49	50.26 101	87.05 40	92.17 23
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
CDS-MVSNet		79.07 46	77.70 48	83.17 43	87.60 62	68.23 32	84.40 68	86.20 67	67.49 61	76.36 44	86.54 71	61.54 38	90.79 73	61.86 56	87.33 36	90.49 36
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
MVSTER		79.01 47	77.88 47	82.38 51	83.07 86	64.80 60	84.08 69	88.95 52	69.01 54	78.69 23	87.17 62	54.70 61	92.43 59	74.69 23	80.57 72	89.89 47

TAMVS		78.89 48	77.51 49	83.03 46	87.80 60	67.79 36	84.72 65	85.05 73	67.63 58	76.75 40	87.70 53	62.25 36	90.82 72	58.53 72	87.13 39	90.49 36

test1		78.40 49	77.40 50	81.40 60	87.60 62	63.01 68	88.39 37	89.28 42	71.63 35	75.34 52	87.28 57	54.80 58	91.11 65	62.72 49	79.57 75	90.09 44

GBi-Net		78.40 49	77.40 50	81.40 60	87.60 62	63.01 68	88.39 37	89.28 42	71.63 35	75.34 52	87.28 57	54.80 58	91.11 65	62.72 49	79.57 75	90.09 44

Vis-MVSNet (Re-imp)		78.36 51	78.45 43	78.07 84	88.64 48	51.78 109	86.70 51	79.63 103	74.14 27	75.11 58	90.83 28	61.29 41	89.75 78	58.10 76	91.60 17	92.69 20

BH-w/o		78.21 52	77.33 52	80.84 67	88.81 45	65.13 59	84.87 63	87.85 58	69.75 49	74.52 62	84.74 82	61.34 39	93.11 53	58.24 75	85.84 48	84.27 99

FMVSNet2		78.20 53	77.21 53	81.20 63	87.60 62	62.89 72	87.47 47	89.02 49	71.63 35	75.29 56	87.28 57	54.80 58	91.10 68	62.38 53	79.38 78	89.61 53

CNLPA		78.08 54	76.79 56	81.97 56	90.40 25	71.07 14	87.59 46	84.55 74	66.03 67	72.38 72	89.64 38	57.56 51	86.04 92	59.61 68	83.35 53	88.79 62

PLC	70.83 11	78.05 55	76.37 59	83.08 45	91.88 16	67.80 35	88.19 41	89.46 39	64.33 77	69.87 85	88.38 48	53.66 70	93.58 45	58.86 70	82.73 57	87.86 69
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
HY-MVS	69.67 12	77.95 56	77.15 54	80.36 71	87.57 66	60.21 83	83.37 75	87.78 59	66.11 65	75.37 51	87.06 63	63.27 28	90.48 75	61.38 59	82.43 60	90.40 39

FMVSNet3		77.88 57	76.85 55	80.97 66	86.84 69	62.36 73	86.52 52	88.77 53	71.13 39	75.34 52	86.66 69	54.07 66	91.10 68	62.72 49	79.57 75	89.45 54

PAPM		77.68 58	76.40 58	81.51 58	87.29 67	61.85 76	83.78 71	89.59 37	64.74 73	71.23 76	88.70 45	62.59 32	93.66 44	52.66 95	87.03 41	89.01 58

FMVSNet1		77.44 59	76.12 61	81.40 60	86.81 70	63.01 68	88.39 37	89.28 42	70.49 42	74.39 63	87.28 57	49.06 87	91.11 65	60.91 62	78.52 80	90.09 44

TR-MVS		77.44 59	76.18 60	81.20 63	88.24 55	63.24 66	84.61 66	86.40 66	67.55 60	77.81 30	86.48 72	54.10 65	93.15 52	57.75 77	82.72 58	87.20 78

1112_ss		77.40 61	76.43 57	80.32 72	89.11 41	60.41 82	83.65 72	87.72 60	62.13 88	73.05 66	86.72 64	62.58 33	89.97 77	62.11 55	80.80 70	90.59 35

LS3D		76.95 62	74.82 66	83.37 40	90.45 24	67.36 41	89.15 30	86.94 64	61.87 90	69.52 87	90.61 31	51.71 78	94.53 32	46.38 111	86.71 44	88.21 66

DP-MVS		76.78 63	74.57 68	83.42 39	93.29 8	69.46 22	88.55 35	83.70 80	63.98 80	70.20 80	88.89 43	54.01 67	94.80 28	46.66 108	81.88 64	86.01 91

cascas		76.72 64	74.64 67	82.99 47	85.78 74	65.88 51	82.33 76	89.21 46	60.85 94	72.74 68	81.02 100	47.28 89	93.75 43	67.48 39	85.02 49	89.34 55

Test_1112_low_res		76.40 65	75.44 64	79.27 77	89.28 36	58.09 90	81.69 79	87.07 63	59.53 101	72.48 71	86.67 68	61.30 40	89.33 80	60.81 64	80.15 74	90.41 38

F-COLMAP		76.38 66	74.33 70	82.50 50	89.28 36	66.95 44	88.41 36	89.03 48	64.05 79	66.83 97	88.61 46	46.78 91	92.89 55	57.48 78	78.55 79	87.67 71

LTVRE_ROB	69.57 13	76.25 67	74.54 69	81.41 59	88.60 49	64.38 62	79.24 92	89.12 47	70.76 40	69.79 86	87.86 51	49.09 86	93.20 50	56.21 84	80.16 73	86.65 88
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
ACMH+	68.96 14	76.01 68	74.01 71	82.03 55	88.60 49	65.31 56	88.86 31	87.55 61	70.25 45	67.75 94	87.47 56	41.27 104	93.19 51	58.37 73	75.94 94	87.60 72

ACMH	67.68 16	75.89 69	73.93 72	81.77 57	88.71 47	66.61 45	88.62 34	89.01 50	69.81 47	66.78 98	86.70 67	41.95 103	91.51 62	55.64 85	78.14 84	87.17 79
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
IB-MVS	68.01 15	75.85 70	73.36 75	83.31 41	84.76 77	66.03 47	83.38 74	85.06 72	70.21 46	69.40 88	81.05 99	45.76 93	94.66 30	65.10 44	75.49 96	89.25 56
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
WTY-MVS		75.65 71	75.68 62	75.57 94	86.40 71	56.82 95	77.92 96	82.40 89	65.10 72	76.18 46	87.72 52	63.13 31	80.90 106	60.31 66	81.96 62	89.00 60

EPNet_dtu		75.46 72	74.86 65	77.23 87	82.57 90	54.60 103	86.89 48	83.09 82	71.64 34	66.25 100	85.86 75	55.99 55	88.04 81	54.92 86	86.55 45	89.05 57

XXY-MVS		75.41 73	75.56 63	74.96 96	83.59 81	57.82 92	80.59 85	83.87 79	66.54 64	74.93 60	88.31 49	63.24 29	80.09 109	62.16 54	76.85 90	86.97 83

CostFormer		75.24 74	73.90 73	79.27 77	82.65 89	58.27 89	80.80 81	82.73 86	61.57 91	75.33 55	83.13 88	55.52 56	91.07 70	64.98 45	78.34 83	88.45 64

IterMVS		74.29 75	72.94 76	78.35 82	81.53 97	63.49 64	81.58 80	82.49 87	68.06 57	69.99 84	83.69 85	51.66 79	85.54 94	65.85 40	71.64 108	86.01 91
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
EG-PatchMatch MVS		74.04 76	71.82 78	80.71 69	84.92 76	67.42 38	85.86 57	88.08 57	66.04 66	64.22 105	83.85 83	35.10 116	92.56 57	57.44 79	80.83 69	82.16 105

sss		73.60 77	73.64 74	73.51 103	82.80 88	55.01 102	76.12 100	81.69 93	62.47 87	74.68 61	85.85 76	57.32 52	78.11 113	60.86 63	80.93 68	87.39 75

tpmp4_e23		73.45 78	71.17 84	80.31 73	83.55 82	59.56 86	81.88 78	82.33 90	57.94 107	70.51 79	81.62 95	51.19 82	91.63 61	53.96 90	77.51 87	89.75 52

CR-MVSNet		73.37 79	71.27 83	79.67 75	81.32 98	65.19 57	75.92 101	80.30 99	59.92 98	72.73 69	81.19 97	52.50 72	86.69 87	59.84 67	77.71 85	87.11 81

MSDG		73.36 80	70.99 85	80.49 70	84.51 79	65.80 52	80.71 83	86.13 69	65.70 69	65.46 102	83.74 84	44.60 96	90.91 71	51.13 99	76.89 88	84.74 98

tpm2		73.26 81	71.46 80	78.63 79	83.34 83	56.71 96	80.65 84	80.40 98	56.63 109	73.55 65	82.02 93	51.80 77	91.24 63	56.35 83	78.42 82	87.95 67

RPSCF		73.23 82	71.46 80	78.54 80	82.50 91	59.85 85	82.18 77	82.84 85	58.96 102	71.15 77	89.41 41	45.48 95	84.77 96	58.82 71	71.83 107	91.02 29

PatchmatchNet		73.12 83	71.33 82	78.49 81	83.18 84	60.85 80	79.63 89	78.57 106	64.13 78	71.73 74	79.81 106	51.20 81	85.97 93	57.40 80	76.36 91	88.66 63
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
COLMAP_ROB	66.92 17	73.01 84	70.41 90	80.81 68	87.13 68	65.63 53	88.30 40	84.19 77	62.96 83	63.80 107	87.69 54	38.04 110	92.56 57	46.66 108	74.91 98	84.24 100
Johannes L. Schönberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
test_0402		72.79 85	70.44 89	79.84 74	88.13 56	65.99 48	85.93 56	84.29 76	65.57 70	67.40 96	85.49 77	46.92 90	92.61 56	35.88 120	74.38 101	80.94 107

tpmrst		72.39 86	72.13 77	73.18 104	80.54 103	49.91 115	79.91 88	79.08 105	63.11 81	71.69 75	79.95 104	55.32 57	82.77 101	65.66 41	73.89 104	86.87 84

PatchMatch-RL		72.38 87	70.90 86	76.80 90	88.60 49	67.38 40	79.53 90	76.17 110	62.75 85	69.36 89	82.00 94	45.51 94	84.89 95	53.62 91	80.58 71	78.12 114

tpm		72.37 88	71.71 79	74.35 99	82.19 93	52.00 107	79.22 93	77.29 108	64.56 75	72.95 67	83.68 86	51.35 80	83.26 99	58.33 74	75.80 95	87.81 70

PVSNet	64.34 18	72.08 89	70.87 87	75.69 93	86.21 72	56.44 98	74.37 109	80.73 96	62.06 89	70.17 81	82.23 91	42.86 100	83.31 98	54.77 87	84.45 50	87.32 77

RPMNet		71.62 90	68.94 94	79.67 75	81.32 98	65.19 57	75.92 101	78.30 107	57.60 108	72.73 69	76.45 113	52.30 74	86.69 87	48.14 105	77.71 85	87.11 81

HyFIR		71.37 91	70.50 88	73.97 102	80.52 104	59.87 84	70.92 114	82.42 88	56.28 110	65.84 101	76.50 112	53.72 69	83.21 100	61.07 61	87.21 38	78.84 112

Patchmtry		70.74 92	69.16 93	75.49 95	80.72 100	54.07 104	74.94 108	80.30 99	58.34 105	70.01 82	81.19 97	52.50 72	86.54 89	53.37 92	71.09 109	85.87 93

MIMVSNet		70.69 93	69.30 92	74.88 97	84.52 78	56.35 99	75.87 103	79.42 104	64.59 74	67.76 93	82.41 89	41.10 105	81.54 105	46.64 110	81.34 66	86.75 86

tpm cat1		70.57 94	68.31 97	77.35 86	82.41 92	57.95 91	78.08 95	80.22 101	52.04 115	68.54 92	77.66 109	52.00 76	87.84 83	51.77 96	72.07 106	86.25 90

OpenMVS_ROB	64.09 19	70.56 95	68.19 98	77.65 85	80.26 105	59.41 87	85.01 61	82.96 84	58.76 103	65.43 103	82.33 90	37.63 112	91.23 64	45.34 113	76.03 93	82.32 103

USDC		70.33 96	68.37 96	76.21 92	80.60 102	56.23 100	79.19 94	86.49 65	60.89 93	61.29 109	85.47 78	31.78 117	89.47 79	53.37 92	76.21 92	82.94 102

CMPMVS	51.72 21	70.19 97	68.16 99	76.28 91	73.15 118	57.55 93	79.47 91	83.92 78	48.02 120	56.48 118	84.81 81	43.13 98	86.42 91	62.67 52	81.81 65	84.89 97
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
FMVSNet5		69.50 98	67.96 101	74.15 100	82.97 87	55.35 101	80.01 87	82.12 91	62.56 86	63.02 108	81.53 96	36.92 113	81.92 103	48.42 104	74.06 102	85.17 96

PMMVS		69.34 99	68.67 95	71.35 108	75.67 113	62.03 75	75.17 104	73.46 116	50.00 118	68.68 91	79.05 107	52.07 75	78.13 112	61.16 60	82.77 56	73.90 117

EPMVS		69.02 100	68.16 99	71.59 106	79.61 108	49.80 116	77.40 97	66.93 121	62.82 84	70.01 82	79.05 107	45.79 92	77.86 114	56.58 82	75.26 97	87.13 80

MIMVSNet1		68.58 101	66.78 103	73.98 101	80.07 106	51.82 108	80.77 82	84.37 75	64.40 76	59.75 111	82.16 92	36.47 114	83.63 97	42.73 114	70.33 110	86.48 89

PatchT		68.46 102	67.85 102	70.29 110	80.70 101	43.93 120	72.47 111	74.88 114	60.15 97	70.55 78	76.57 111	49.94 83	81.59 104	50.58 100	74.83 99	85.34 95

TDRefinement		67.49 103	64.34 107	76.92 88	73.47 117	61.07 78	84.86 64	82.98 83	59.77 99	58.30 114	85.13 79	26.06 118	87.89 82	47.92 106	60.59 118	81.81 106

UnsupCasMVSNet_eth		67.33 104	65.99 104	71.37 107	73.48 116	51.47 111	75.16 105	85.19 71	65.20 71	60.78 110	80.93 102	42.35 101	77.20 116	57.12 81	53.69 123	85.44 94

TinyColmap		67.30 105	64.81 106	74.76 98	81.92 94	56.68 97	80.29 86	81.49 94	60.33 96	56.27 119	83.22 87	24.77 119	87.66 84	45.52 112	69.47 111	79.95 110

dp		66.80 106	65.43 105	70.90 109	79.74 107	48.82 118	75.12 107	74.77 115	59.61 100	64.08 106	77.23 110	42.89 99	80.72 107	48.86 103	66.58 114	83.16 101

JIA-IIPM		66.32 107	62.82 108	76.82 89	77.09 111	61.72 77	65.34 120	75.38 112	58.04 106	64.51 104	62.32 121	42.05 102	86.51 90	51.45 98	69.22 112	82.21 104

LF4IMVS		64.02 108	62.19 109	69.50 112	70.90 119	53.29 106	76.13 99	77.18 109	52.65 114	58.59 112	80.98 101	23.55 120	76.52 117	53.06 94	66.66 113	78.68 113

UnsupCasMVSNet_bld		63.70 109	61.53 110	70.21 111	73.69 115	51.39 112	72.82 110	81.89 92	55.63 112	57.81 115	71.80 116	38.67 108	78.61 111	49.26 102	52.21 124	80.63 108

PVSNet_0	57.27 20	61.67 110	59.27 111	68.85 113	79.61 108	57.44 94	68.01 118	73.44 117	55.93 111	58.54 113	70.41 117	44.58 97	77.55 115	47.01 107	35.91 125	71.55 118

LP		61.36 111	57.78 112	72.09 105	75.54 114	58.53 88	67.16 119	75.22 113	51.90 116	54.13 120	69.97 118	37.73 111	80.45 108	32.74 123	55.63 121	77.29 115

MVS-HIRNet		59.14 112	57.67 113	63.57 116	81.65 96	43.50 121	71.73 112	65.06 122	39.59 123	51.43 121	57.73 122	38.34 109	82.58 102	39.53 116	73.95 103	64.62 121

DSMNet-mixed		57.77 113	56.90 114	60.38 117	67.70 122	35.61 125	69.18 117	53.97 125	32.30 128	57.49 116	79.88 105	40.39 107	68.57 124	38.78 117	72.37 105	76.97 116

FPMVS		53.68 114	51.64 116	59.81 118	65.08 123	51.03 113	69.48 116	69.58 118	41.46 121	40.67 124	72.32 115	16.46 126	70.00 123	24.24 126	65.42 115	58.40 123

N_pmnet		52.79 115	53.26 115	51.40 124	78.99 110	7.68 131	69.52 115	3.89 130	51.63 117	57.01 117	74.98 114	40.83 106	65.96 125	37.78 118	64.67 116	80.56 109

HyFIR lowres test		51.79 116	50.01 118	57.11 120	68.82 120	49.21 117	60.50 123	53.26 126	34.52 124	43.77 123	64.94 120	20.34 123	71.75 121	39.87 115	64.06 117	50.39 124

no-one		51.08 117	45.79 120	66.95 114	57.92 126	50.49 114	59.63 125	76.04 111	48.04 119	31.85 125	56.10 125	19.12 124	80.08 110	36.89 119	26.52 126	70.29 119

new_pmnet		50.91 118	50.29 117	52.78 123	68.58 121	34.94 128	63.71 122	56.63 124	39.73 122	44.95 122	65.47 119	21.93 121	58.48 127	34.98 121	56.62 120	64.92 120

ANet_high		50.57 119	46.10 119	63.99 115	48.67 127	39.13 123	70.99 113	80.85 95	61.39 92	31.18 126	57.70 123	17.02 125	73.65 120	31.22 124	15.89 131	79.18 111

Gipuma		45.18 120	41.86 121	55.16 122	77.03 112	51.52 110	32.50 128	80.52 97	32.46 126	27.12 127	35.02 129	9.52 128	75.50 119	22.31 127	60.21 119	38.45 128
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
PMVS	37.38 22	44.16 121	40.28 122	55.82 121	40.82 130	42.54 122	65.12 121	63.99 123	34.43 125	24.48 128	57.12 124	3.92 129	76.17 118	17.10 129	55.52 122	48.75 125
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
wuykxyi23d		39.76 122	33.18 124	59.51 119	46.98 128	44.01 119	57.70 126	67.74 120	24.13 129	13.98 132	34.33 130	1.27 132	71.33 122	34.23 122	18.23 129	63.18 122

PNet_i23d		38.26 123	35.42 123	46.79 125	58.74 124	35.48 126	59.65 124	51.25 127	32.45 127	23.44 130	47.53 127	2.04 131	58.96 126	25.60 125	18.09 130	45.92 127

MVE	26.22 23	30.37 124	25.89 125	43.81 126	44.55 129	35.46 127	28.87 129	39.07 128	18.20 130	18.58 131	40.18 128	2.68 130	47.37 128	17.07 130	23.78 128	48.60 126
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
wuyk23d		16.82 125	15.94 126	19.46 128	58.74 124	31.45 129	39.22 127	3.74 131	6.84 132	6.04 133	2.70 131	1.27 132	24.29 130	10.54 131	14.40 132	2.63 130

ab-mvs-re		7.23 126	9.64 127	0.00 129	0.00 132	0.00 132	0.00 130	0.00 132	0.00 133	0.00 134	86.72 64	0.00 134	0.00 131	0.00 132	0.00 133	0.00 131

NP-MVS											90.24 34

HQP5-MVS						66.98 43

HQP-MVS												21.25 122

ACMMP++_ref															81.95 63

MDTV_nov1_ep13			69.97 91		83.18 84	53.48 105	77.10 98	80.18 102	60.45 95	69.33 90	80.44 103	48.89 88	86.90 86	51.60 97	78.51 81

door								69.44 119

MDTV_nov1_ep13_2view						37.79 124	75.16 105		55.10 113	66.53 99		49.34 84		53.98 89		87.94 68

HQP-NCC					89.33 33		89.17 28		76.41 14	77.23 35

Test By Simon												64.33 24

DeepMVS_CX				27.40 127	40.17 131	26.90 130		24.59 129	17.44 131	23.95 129	48.61 126	9.77 127	26.48 129	18.06 128	24.47 127	28.83 129

ACMP_Plane					89.33 33		89.17 28		76.41 14	77.23 35

ACMMP++															81.25 67

BP-MVS														77.47 18

HQP4-MVS										77.24 34			95.11 18			91.03 28

HQP3-MVS								92.19 15							85.99 47

HQP2-MVS												60.17 47

ITE_SJBPF				78.22 83	81.77 95	60.57 81		83.30 81	69.25 53	67.54 95	87.20 61	36.33 115	87.28 85	54.34 88	74.62 100	86.80 85

LGP-MVS_train				84.50 27	89.23 38	68.76 27		91.94 20	75.37 19	76.64 42	91.51 21	54.29 63	94.91 23	78.44 15	83.78 51	89.83 49

High-resolution multi-view benchmark