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.

DeepPCF-MVS	69.58 1	79.03 1	79.00 1	79.13 3	84.92 7	60.32 5	83.03 7	85.33 5	62.86 5	80.17 1	90.03 2	61.76 1	88.95 4	74.21 2	88.67 3	88.12 3

DeepC-MVS	69.38 2	78.56 2	78.14 2	79.83 1	83.60 9	61.62 2	84.17 5	86.85 1	63.23 3	73.84 3	90.25 1	57.68 3	89.96 1	74.62 1	89.03 2	87.89 5
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	68.24 3	77.25 3	76.63 4	79.12 4	86.15 2	60.86 3	84.71 3	84.85 6	61.98 7	73.06 5	88.88 5	53.72 5	89.06 3	68.27 5	88.04 4	87.42 8
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
CSCG		76.92 4	76.75 3	77.41 6	83.96 8	59.60 7	82.95 8	86.50 2	60.78 11	75.27 2	84.83 22	60.76 2	86.56 13	67.86 6	87.87 5	86.06 9

ACMMP		76.02 5	75.33 6	78.07 5	85.20 4	61.91 1	85.49 2	84.44 8	63.04 4	69.80 9	89.74 3	45.43 27	87.16 9	72.01 3	82.87 14	85.14 15

PHI-MVS		75.87 6	75.36 5	77.41 6	80.62 19	55.91 20	84.28 4	85.78 4	56.08 29	73.41 4	86.58 10	50.94 13	88.54 5	70.79 4	89.71 1	87.79 6

3Dnovator+	66.72 4	75.84 7	74.57 7	79.66 2	82.40 11	59.92 6	85.83 1	86.32 3	66.92 1	67.80 18	89.24 4	42.03 37	89.38 2	64.07 11	86.50 6	89.69 1

MVS_111021_HR		74.02 8	73.46 9	75.69 10	83.01 10	60.63 4	77.29 37	78.40 31	61.18 10	70.58 7	85.97 14	54.18 4	84.00 29	67.52 7	82.98 13	82.45 34

MG-MVS		73.96 9	73.89 8	74.16 21	85.65 3	49.69 48	81.59 14	81.29 22	61.45 8	71.05 6	88.11 6	51.77 9	87.73 8	61.05 13	83.09 11	85.05 17

CLD-MVS		73.33 10	72.68 10	75.29 14	78.82 31	53.33 31	78.23 31	84.79 7	61.30 9	70.41 8	81.04 48	52.41 7	87.12 10	64.61 10	82.49 15	85.41 14
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
PAPM_NR		72.63 11	71.80 11	75.13 15	81.72 15	53.42 29	79.91 24	83.28 11	59.14 18	66.31 23	85.90 15	51.86 8	86.06 16	57.45 19	80.62 18	85.91 11

3Dnovator	64.47 5	72.49 12	71.39 13	75.79 9	77.70 38	58.99 10	80.66 20	83.15 12	62.24 6	65.46 29	86.59 9	42.38 36	85.52 22	59.59 17	84.72 7	82.85 30

ACMP	63.53 6	72.30 13	71.20 16	75.59 12	80.28 21	57.54 16	82.74 9	82.84 15	60.58 13	65.24 31	86.18 12	39.25 45	86.03 18	66.95 8	76.79 34	83.22 23
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
Vis-MVSNet		72.18 14	71.37 14	74.61 18	81.29 16	55.41 23	80.90 18	78.28 32	60.73 12	69.23 11	88.09 7	44.36 32	82.65 35	57.68 18	81.75 16	85.77 12
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
API-MVS		72.17 15	71.41 12	74.45 19	81.95 13	57.22 17	84.03 6	80.38 24	59.89 16	68.40 13	82.33 38	49.64 14	87.83 7	51.87 25	84.16 9	78.30 53

EPP-MVSNet		72.16 16	71.31 15	74.71 16	78.68 33	49.70 47	82.10 13	81.65 18	60.40 14	65.94 25	85.84 16	51.74 10	86.37 14	55.93 20	79.55 23	88.07 4

DP-MVS Recon		72.15 17	70.73 19	76.40 8	86.57 1	57.99 14	81.15 17	82.96 13	57.03 23	66.78 21	85.56 17	44.50 30	88.11 6	51.77 27	80.23 21	83.10 25

PAPR		71.72 18	70.82 18	74.41 20	81.20 17	51.17 42	79.55 27	83.33 10	55.81 31	66.93 20	84.61 25	50.95 12	86.06 16	55.79 23	79.20 25	86.00 10

IS-MVSNet		71.57 19	71.00 17	73.27 28	78.86 30	45.63 57	80.22 22	78.69 30	64.14 2	66.46 22	87.36 8	49.30 16	85.60 21	50.26 33	83.71 10	88.59 2

MAR-MVS		71.51 20	70.15 22	75.60 11	81.84 14	59.39 8	81.38 15	82.90 14	54.90 36	68.08 15	78.70 59	47.73 22	85.51 23	51.68 28	84.17 8	81.88 38
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
PVSNet_Blended_VisFu		71.45 21	70.39 21	74.65 17	82.01 12	58.82 11	79.93 23	80.35 25	55.09 35	65.82 27	82.16 41	49.17 18	82.64 36	60.34 15	78.62 29	82.50 33

OMC-MVS		71.40 22	70.60 20	73.78 24	76.60 50	53.15 32	79.74 26	79.78 27	58.37 20	68.75 12	86.45 11	45.43 27	80.60 43	62.58 12	77.73 31	87.58 7

PCF-MVS	61.88 8	70.95 23	69.49 24	75.35 13	77.63 40	55.71 21	76.04 43	81.81 17	50.30 53	69.66 10	85.40 19	52.51 6	84.89 25	51.82 26	80.24 20	85.45 13
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
ACMM	61.98 7	70.80 24	69.73 23	74.02 22	80.59 20	58.59 12	82.68 10	82.02 16	55.46 34	67.18 19	84.39 26	38.51 49	83.17 32	60.65 14	76.10 36	80.30 42
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
AdaColmap		69.99 25	68.66 26	73.97 23	84.94 5	57.83 15	82.63 11	78.71 29	56.28 28	64.34 38	84.14 28	41.57 41	87.06 11	46.45 41	78.88 26	77.02 60

MVS_111021_LR		69.50 26	68.78 25	71.65 34	78.38 34	59.33 9	74.82 46	70.11 60	58.08 21	67.83 17	84.68 23	41.96 38	76.34 56	65.62 9	77.54 32	79.30 50

QAPM		69.46 27	68.16 28	73.36 27	76.38 53	53.37 30	82.46 12	83.46 9	50.36 52	65.60 28	86.12 13	40.23 42	86.58 12	50.52 30	78.27 30	80.12 46

IterMVS-LS		69.22 28	68.48 27	71.43 36	74.44 60	49.40 50	76.23 41	77.55 36	59.60 17	65.85 26	81.59 45	51.28 11	81.58 38	59.87 16	69.90 59	83.30 22

OpenMVS	61.03 9	68.85 29	67.56 31	72.70 30	74.26 61	53.99 26	81.21 16	81.34 21	52.70 43	62.75 42	85.55 18	38.86 47	84.14 28	48.41 36	83.01 12	79.97 47

BH-RMVSNet		68.81 30	67.42 32	72.97 29	80.11 22	52.53 38	74.26 48	76.29 42	58.48 19	68.38 14	84.20 27	42.59 35	83.83 30	46.53 40	75.91 37	82.56 31

PVSNet_Blended		68.59 31	67.72 29	71.19 39	77.03 46	50.57 44	72.51 53	81.52 19	51.91 47	64.22 39	77.77 64	49.13 19	82.87 33	55.82 21	79.58 22	80.14 45

PVSNet_BlendedMVS		68.56 32	67.72 29	71.07 42	77.03 46	50.57 44	74.50 47	81.52 19	53.66 40	64.22 39	79.72 57	49.13 19	82.87 33	55.82 21	73.92 40	79.77 49

UGNet		67.97 33	66.65 34	71.93 32	77.70 38	49.99 46	78.04 33	75.26 46	60.01 15	61.27 46	83.93 30	31.63 64	80.96 41	50.40 31	81.67 17	84.85 18

PAPM		67.92 34	66.69 33	71.63 35	78.09 35	49.02 51	77.09 38	81.24 23	51.04 50	60.91 48	83.98 29	47.71 23	84.99 24	40.81 61	79.32 24	80.90 41

GBi-Net		67.21 35	66.55 35	69.19 50	77.63 40	43.33 62	77.31 35	77.83 34	56.62 24	65.04 34	82.70 33	41.85 39	80.33 44	47.18 37	72.76 44	83.92 20

FMVSNet2		66.93 36	66.31 37	68.79 53	77.63 40	42.98 64	76.11 42	77.47 37	56.62 24	65.22 33	82.17 40	41.85 39	80.18 46	47.05 39	72.72 46	83.20 24

anonymous		66.80 37	65.37 40	71.10 41	78.98 29	53.13 33	73.27 51	71.07 56	52.15 46	64.72 36	80.23 55	43.56 34	77.10 54	45.48 45	78.88 26	83.05 26

TAMVS		66.78 38	65.27 41	71.33 38	79.16 27	53.67 27	73.84 49	69.59 62	52.32 45	65.28 30	81.72 43	44.49 31	77.40 53	42.32 55	78.66 28	82.92 27

FMVSNet1		66.70 39	65.87 38	69.19 50	77.49 43	43.33 62	77.31 35	77.83 34	56.45 26	64.60 37	82.70 33	38.08 51	80.33 44	46.08 43	72.31 49	83.92 20

ab-mvs		66.65 40	66.42 36	67.37 56	76.17 54	41.73 69	70.41 60	76.14 44	53.99 39	65.98 24	83.51 32	49.48 15	76.24 57	48.60 35	73.46 43	84.14 19

TAPA-MVS	59.36 10	66.60 41	65.20 42	70.81 43	76.63 49	48.75 52	76.52 40	80.04 26	50.64 51	65.24 31	84.93 21	39.15 46	78.54 51	36.77 70	76.88 33	85.14 15
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
TR-MVS		66.59 42	65.07 43	71.17 40	79.18 26	49.63 49	73.48 50	75.20 47	52.95 42	67.90 16	80.33 54	39.81 44	83.68 31	43.20 51	73.56 42	80.20 43

FMVSNet3		66.32 43	65.61 39	68.46 55	76.48 52	42.34 67	74.98 45	77.15 40	55.83 30	65.04 34	81.16 47	39.91 43	80.14 47	47.18 37	72.76 44	82.90 29

ACMH+	57.40 11	66.12 44	64.06 44	72.30 31	77.79 37	52.83 34	80.39 21	78.03 33	57.30 22	57.47 60	82.55 35	27.68 74	84.17 27	45.54 44	69.78 60	79.90 48

cascas		65.98 45	63.42 50	73.64 25	77.26 44	52.58 37	72.26 54	77.21 39	48.56 58	61.21 47	74.60 74	32.57 62	85.82 20	50.38 32	76.75 35	82.52 32

DP-MVS		65.68 46	63.66 48	71.75 33	84.93 6	56.87 18	80.74 19	73.16 53	53.06 41	59.09 52	82.35 37	36.79 53	85.94 19	32.82 77	69.96 58	72.45 75

CNLPA		65.43 47	64.02 45	69.68 47	78.73 32	58.07 13	77.82 34	70.71 58	51.49 48	61.57 45	83.58 31	38.23 50	70.82 65	43.90 49	70.10 56	80.16 44

IB-MVS	56.42 12	65.40 48	62.73 53	73.40 26	74.89 56	52.78 35	73.09 52	75.13 48	55.69 33	58.48 58	73.73 75	32.86 60	86.32 15	50.63 29	70.11 55	81.10 40
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
ACMH	55.70 15	65.20 49	63.57 49	70.07 45	78.07 36	52.01 41	79.48 28	79.69 28	55.75 32	56.59 63	80.98 49	27.12 76	80.94 42	42.90 53	71.58 52	77.25 59
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
PLC	56.13 14	65.09 50	63.21 52	70.72 44	81.04 18	54.87 24	78.57 29	77.47 37	48.51 59	55.71 65	81.89 42	33.71 57	79.71 48	41.66 59	70.37 54	77.58 56
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
LS3D		64.71 51	62.50 54	71.34 37	79.72 24	55.71 21	79.82 25	74.72 49	48.50 60	56.62 62	84.62 24	33.59 58	82.34 37	29.65 83	75.23 38	75.97 62

HY-MVS	56.14 13	64.55 52	63.89 46	66.55 58	74.73 58	41.02 70	69.96 61	74.43 50	49.29 55	61.66 44	80.92 50	47.43 24	76.68 55	44.91 47	71.69 51	81.94 36

1112_ss		64.00 53	63.36 51	65.93 59	79.28 25	42.58 66	71.35 55	72.36 54	46.41 68	60.55 49	77.89 61	46.27 26	73.28 60	46.18 42	69.97 57	81.92 37

Vis-MVSNet (Re-imp)		63.69 54	63.88 47	63.14 64	74.75 57	31.04 91	71.16 57	63.64 75	56.32 27	59.80 51	84.99 20	44.51 29	75.46 58	39.12 63	80.62 18	82.92 27

test_0402		63.25 55	61.01 59	69.96 46	80.00 23	54.37 25	76.86 39	72.02 55	54.58 37	58.71 55	80.79 52	35.00 54	84.36 26	26.41 87	64.71 72	71.15 79

LTVRE_ROB	55.42 16	63.15 56	61.23 58	68.92 52	76.57 51	47.80 53	59.92 80	76.39 41	54.35 38	58.67 56	82.46 36	29.44 69	81.49 39	42.12 56	71.14 53	77.46 57
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
F-COLMAP		63.05 57	60.87 60	69.58 48	76.99 48	53.63 28	78.12 32	76.16 43	47.97 62	52.41 73	81.61 44	27.87 73	78.11 52	40.07 62	66.66 67	77.00 61

IterMVS		62.79 58	61.27 57	67.35 57	69.37 71	52.04 40	71.17 56	68.24 66	52.63 44	59.82 50	76.91 66	37.32 52	72.36 62	52.80 24	63.19 75	77.66 55

Test_1112_low_res		62.32 59	61.77 55	64.00 62	79.08 28	39.53 73	68.17 63	70.17 59	43.25 78	59.03 53	79.90 56	44.08 33	71.24 64	43.79 50	68.42 62	81.25 39

MSDG		61.81 60	59.23 62	69.55 49	72.64 65	52.63 36	70.45 59	75.81 45	51.38 49	53.70 70	76.11 67	29.52 67	81.08 40	37.70 67	65.79 70	74.93 67

COLMAP_ROB	52.97 17	61.27 61	58.81 63	68.64 54	74.63 59	52.51 39	78.42 30	73.30 52	49.92 54	50.96 75	81.51 46	23.06 84	79.40 50	31.63 81	65.85 69	74.01 71
Johannes L. Schönberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
XXY-MVS		60.68 62	61.67 56	57.70 76	70.43 69	38.45 74	64.19 71	66.47 68	48.05 61	63.22 41	80.86 51	49.28 17	60.47 79	45.25 46	67.28 66	74.19 69

OpenMVS_ROB	52.78 18	60.03 63	58.14 65	65.69 60	70.47 68	44.82 58	75.33 44	70.86 57	45.04 71	56.06 64	76.00 68	26.89 78	79.65 49	35.36 73	67.29 65	72.60 73

PatchmatchNet		59.84 64	58.24 64	64.65 61	73.05 63	46.70 56	69.42 62	62.18 78	47.55 64	58.88 54	71.96 79	34.49 55	69.16 67	42.99 52	63.60 74	78.07 54
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
WTY-MVS		59.75 65	60.39 61	57.85 74	72.32 66	37.83 77	61.05 79	64.18 74	45.95 70	61.91 43	79.11 58	47.01 25	60.88 78	42.50 54	69.49 61	74.83 68

PVSNet	50.76 19	58.40 66	57.39 66	61.42 68	75.53 55	44.04 61	61.43 76	63.45 76	47.04 66	56.91 61	73.61 76	27.00 77	64.76 70	39.12 63	72.40 47	75.47 65

CMPMVS	42.80 21	57.81 67	55.97 70	63.32 63	60.98 85	47.38 54	64.66 69	69.50 63	32.06 87	46.83 81	77.80 63	29.50 68	71.36 63	48.68 34	73.75 41	71.21 78
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
MIMVSNet		57.35 68	57.07 67	58.22 73	74.21 62	37.18 78	62.46 73	60.88 80	48.88 57	55.29 69	75.99 70	31.68 63	62.04 77	31.87 80	72.35 48	75.43 66

USDC		56.35 69	54.24 74	62.69 65	64.74 80	40.31 72	65.05 67	73.83 51	43.93 77	47.58 79	77.71 65	15.36 86	75.05 59	38.19 66	61.81 79	72.70 72

PatchMatch-RL		56.25 70	54.55 72	61.32 70	77.06 45	56.07 19	65.57 66	54.10 85	44.13 76	53.49 72	71.27 81	25.20 83	66.78 68	36.52 71	63.66 73	61.12 84

sss		56.17 71	56.57 69	54.96 79	66.93 75	36.32 80	57.94 83	61.69 79	41.67 80	58.64 57	75.32 72	38.72 48	56.25 88	42.04 57	66.19 68	72.31 76

FMVSNet5		55.86 72	54.93 71	58.66 72	71.05 67	36.35 79	64.18 72	62.48 77	46.76 67	50.66 76	74.73 73	25.80 81	64.04 72	33.11 76	65.57 71	75.59 64

RPSCF		55.80 73	54.22 75	60.53 71	65.13 79	42.91 65	64.30 70	57.62 81	36.84 84	58.05 59	82.28 39	28.01 72	56.24 89	37.14 69	58.61 81	82.44 35

MIMVSNet1		55.17 74	54.31 73	57.77 75	70.03 70	32.01 89	65.68 65	64.81 70	49.19 56	46.75 82	76.00 68	25.53 82	64.04 72	28.65 84	62.13 78	77.26 58

TinyColmap		54.14 75	51.72 79	61.40 69	66.84 76	41.97 68	66.52 64	68.51 65	44.81 72	42.69 86	75.77 71	11.66 89	72.94 61	31.96 79	56.77 82	69.27 80

PMMVS		53.96 76	53.26 77	56.04 77	62.60 83	50.92 43	61.17 78	56.09 82	32.81 86	53.51 71	66.84 84	34.04 56	59.93 81	44.14 48	68.18 63	57.27 86

EPMVS		53.96 76	53.69 76	54.79 80	66.12 78	31.96 90	62.34 75	49.05 87	44.42 74	55.54 66	71.33 80	30.22 66	56.70 85	41.65 60	62.54 76	75.71 63

TDRefinement		53.44 78	50.72 81	61.60 67	64.31 81	46.96 55	70.89 58	65.27 69	41.78 79	44.61 84	77.98 60	11.52 90	66.36 69	28.57 85	51.59 86	71.49 77

UnsupCasMVSNet_eth		53.16 79	52.47 78	55.23 78	59.45 87	33.39 85	59.43 81	69.13 64	45.98 69	50.35 77	72.32 78	29.30 70	58.26 84	42.02 58	44.30 91	74.05 70

dp		51.89 80	51.60 80	52.77 84	68.44 73	32.45 87	62.36 74	54.57 84	44.16 75	49.31 78	67.91 83	28.87 71	56.61 86	33.89 75	54.89 84	69.24 81

UnsupCasMVSNet_bld		50.07 81	48.87 82	53.66 81	60.97 86	33.67 84	57.62 84	64.56 72	39.47 83	47.38 80	64.02 87	27.47 75	59.32 82	34.69 74	43.68 92	67.98 82

PVSNet_0	43.31 20	47.46 82	45.64 83	52.92 83	67.60 74	44.65 59	54.06 88	54.64 83	41.59 81	46.15 83	58.75 88	30.99 65	58.66 83	32.18 78	24.81 94	55.46 87

MVS-HIRNet		45.52 83	44.48 84	48.65 85	68.49 72	34.05 83	59.41 82	44.50 91	27.03 90	37.96 88	50.47 93	26.16 80	64.10 71	26.74 86	59.52 80	47.82 88

LF4IMVS		42.95 84	42.26 85	45.04 87	48.30 91	32.50 86	54.80 86	48.49 89	28.03 89	40.51 87	70.16 82	9.24 93	43.89 92	31.63 81	49.18 88	58.72 85

FPMVS		42.18 85	41.11 86	45.39 86	58.03 88	41.01 71	49.50 89	53.81 86	30.07 88	33.71 89	64.03 86	11.69 88	52.08 90	14.01 95	55.11 83	43.09 91

ANet_high		41.38 86	37.47 88	53.11 82	39.73 94	24.45 96	56.94 85	69.69 61	47.65 63	26.04 92	52.32 90	12.44 87	62.38 76	21.80 91	10.61 99	72.49 74

N_pmnet		39.35 87	40.28 87	36.54 92	63.76 82	1.62 99	49.37 90	0.76 98	34.62 85	43.61 85	66.38 85	26.25 79	42.57 94	26.02 88	51.77 85	65.44 83

HyFIR lowres test		36.24 88	34.78 89	40.64 91	50.09 90	35.94 82	43.63 93	33.13 93	20.35 95	30.75 91	51.75 91	10.74 91	51.24 91	25.80 89	46.76 90	34.75 94

PMVS	28.69 22	36.22 89	33.29 91	45.02 88	36.82 95	35.98 81	54.68 87	48.74 88	26.31 91	21.02 93	51.61 92	2.88 96	60.10 80	9.99 97	47.58 89	38.99 93
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
Gipuma		34.77 90	31.91 92	43.33 90	62.05 84	37.87 76	20.39 96	67.03 67	23.23 92	18.41 94	25.84 97	4.24 94	62.73 74	14.71 94	51.32 87	29.38 96
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
new_pmnet		34.13 91	34.29 90	33.64 93	52.63 89	18.23 97	44.43 92	33.90 92	22.81 93	30.89 90	53.18 89	10.48 92	35.72 95	20.77 92	39.51 93	46.98 89

wuykxyi23d		28.12 92	22.54 94	44.87 89	34.97 96	32.11 88	37.96 94	47.31 90	13.32 96	9.29 98	23.72 98	0.45 99	56.58 87	21.85 90	13.98 96	45.93 90

PNet_i23d		27.88 93	25.99 93	33.55 94	47.54 92	25.89 93	47.24 91	32.91 94	21.44 94	15.90 95	38.09 94	0.85 98	42.76 93	16.90 93	13.03 97	32.00 95

MVE	17.77 23	21.41 94	17.77 95	32.34 95	34.34 97	25.44 95	16.11 97	24.11 95	11.19 97	13.22 96	31.92 95	1.58 97	30.95 96	10.47 96	17.03 95	40.62 92
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
wuyk23d		13.32 95	12.52 96	15.71 96	47.54 92	26.27 92	31.06 95	1.98 97	4.93 99	5.18 99	1.94 99	0.45 99	18.54 97	6.81 99	12.83 98	2.33 98

ab-mvs-re		6.49 96	8.65 97	0.00 98	0.00 99	0.00 100	0.00 98	0.00 99	0.00 100	0.00 100	77.89 61	0.00 101	0.00 99	0.00 100	0.00 101	0.00 99

Test By Simon												48.33 21

DeepMVS_CX				12.03 97	17.97 98	10.91 98		10.60 96	7.46 98	11.07 97	28.36 96	3.28 95	11.29 98	8.01 98	9.74 100	13.89 97

ACMMP++															72.16 50

ACMMP++_ref															74.07 39

ITE_SJBPF				62.09 66	66.16 77	44.55 60		64.32 73	47.36 65	55.31 68	80.34 53	19.27 85	62.68 75	36.29 72	62.39 77	79.04 51

MDTV_nov1_ep13			57.00 68		72.73 64	38.26 75	65.02 68	64.73 71	44.74 73	55.46 67	72.48 77	32.61 61	70.47 66	37.47 68	67.75 64

MDTV_nov1_ep13_2view						25.89 93	61.22 77		40.10 82	51.10 74		32.97 59		38.49 65		78.61 52

High-resolution multi-view benchmark