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.

UA-Net		78.65 36	83.96 42	72.46 41	84.87 1	76.15 42	89.06 30	55.70 45	77.25 63	53.14 108	79.73 77	82.09 120	59.69 67	92.21 5	90.93 3	92.32 3	89.36 6

mPP-MVS					82.97 2							92.12 32

DTE-MVSNet		77.28 46	84.87 34	68.42 60	82.94 3	72.70 62	81.60 62	61.78 13	85.03 12	51.40 115	92.11 4	96.00 6	49.42 119	89.73 23	82.52 47	83.39 72	75.98 82

zzz-MVS		82.61 10	85.04 31	79.79 6	82.59 4	73.90 55	92.42 9	62.39 12	84.54 15	80.21 9	79.86 75	90.74 52	70.63 16	90.01 21	89.71 8	90.48 21	86.49 34

MP-MVS		83.50 5	86.11 18	80.45 3	82.58 5	80.60 24	92.68 5	63.48 3	81.43 42	80.21 9	81.95 59	90.76 51	72.86 6	90.14 19	89.30 11	90.92 19	88.59 19
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
PEN-MVS		77.06 49	85.05 30	67.74 63	82.29 6	72.59 63	80.86 66	61.03 20	84.66 13	50.08 122	92.19 2	96.59 3	49.12 120	89.83 22	82.35 49	83.06 74	77.14 78

HPM-MVS++		80.44 28	82.57 47	77.96 14	81.99 7	72.76 60	90.48 19	61.31 15	80.85 48	77.90 18	81.93 60	87.01 92	68.20 28	84.15 55	85.27 30	87.85 40	86.00 37

CP-MVS		84.06 3	86.79 9	80.86 2	81.81 8	79.66 29	92.67 6	64.48 1	83.13 28	82.32 3	80.89 68	92.97 24	72.51 9	91.74 6	90.02 6	91.40 17	89.14 7

SR-MVS					81.31 9			62.63 9				91.11 43

PS-CasMVS		77.46 45	85.80 22	67.73 64	81.24 10	72.88 59	80.63 68	61.28 16	84.14 19	50.53 119	92.13 3	96.76 1	50.12 116	91.02 9	84.46 32	82.60 81	79.19 65

APD-MVS		80.60 27	84.63 36	75.91 24	81.22 11	81.48 10	90.49 18	58.81 30	77.54 62	67.49 65	85.90 26	89.82 65	69.43 21	86.08 43	83.80 38	88.01 38	87.77 24
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
CP-MVSNet		77.01 50	85.04 31	67.65 65	81.16 12	72.72 61	80.54 69	61.18 17	82.09 37	50.41 120	90.81 7	95.89 7	50.03 117	90.86 10	84.30 35	82.56 83	78.65 71

ACMMPR		83.94 4	87.20 3	80.14 4	81.04 13	81.92 8	92.57 8	63.14 5	84.35 18	79.45 12	83.37 43	92.04 33	72.82 8	90.66 12	88.96 12	91.80 6	89.13 8

train_agg		77.83 40	80.47 57	74.77 28	80.92 14	69.60 73	88.87 32	56.32 40	74.03 79	71.03 46	83.67 42	87.68 82	64.75 40	83.70 57	81.85 53	86.71 46	82.73 48

TSAR-MVS + MP.		81.23 23	86.13 16	75.52 26	80.74 15	83.22 3	90.55 17	55.12 48	80.87 47	67.62 63	88.01 14	92.38 29	70.61 17	86.64 40	83.10 43	88.51 33	88.67 16
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
CPTT-MVS		82.32 13	85.00 33	79.19 7	80.73 16	80.86 21	91.68 13	62.59 10	82.55 33	75.53 29	73.88 108	92.28 30	73.74 5	90.07 20	87.65 19	90.87 20	87.74 25

PGM-MVS		83.03 7	85.67 25	79.95 5	80.69 17	81.09 15	92.40 10	63.06 6	79.38 56	80.21 9	80.31 71	91.44 38	71.75 12	90.46 15	88.53 15	91.57 9	88.50 20

WR-MVS_H		77.56 43	85.88 21	67.86 62	80.54 18	74.32 52	83.23 52	61.78 13	83.47 25	47.46 134	91.81 5	95.84 8	50.50 113	90.44 16	84.37 33	83.63 68	80.89 59

SteuartSystems-ACMMP		82.16 15	85.55 27	78.21 11	80.48 19	79.28 30	92.65 7	61.03 20	80.55 50	77.00 23	81.80 62	90.71 53	68.73 26	90.25 17	87.94 18	89.36 28	88.30 22
Skip Steuart: Steuart Systems R&D Blog.
XVS					80.47 20	81.29 12	93.33 3			77.45 20		90.19 61				91.52 11

X-MVStestdata					80.47 20	81.29 12	93.33 3			77.45 20		90.19 61				91.52 11

SD-MVS		82.13 17	86.80 8	76.67 20	80.36 22	80.66 22	89.48 26	56.93 35	82.50 34	67.55 64	87.05 19	91.40 40	72.84 7	88.66 28	88.32 16	92.85 2	89.04 10
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
X-MVS		81.61 21	84.73 35	77.97 12	80.31 23	81.29 12	93.53 2	62.50 11	81.41 43	77.45 20	72.04 118	90.19 61	62.50 51	90.57 13	88.87 13	91.54 10	88.73 15

SMA-MVS		82.15 16	85.93 20	77.74 16	80.13 24	80.25 26	91.01 16	60.61 24	85.54 10	78.61 15	83.21 46	86.96 95	65.95 34	88.10 32	87.59 20	90.11 22	89.83 4

ACMMP		83.17 6	86.75 10	79.01 8	80.11 25	82.01 7	92.29 11	60.35 26	82.20 36	78.32 16	80.59 69	93.14 21	70.67 15	91.30 8	89.36 10	92.30 4	88.62 17
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
ACMM	71.24 7	82.85 9	86.59 11	78.50 10	80.10 26	78.59 31	91.77 12	60.76 23	84.43 16	76.49 25	81.58 64	93.50 16	70.45 19	91.38 7	89.42 9	91.42 16	87.22 29
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
HFP-MVS		82.37 11	86.28 14	77.81 15	79.94 27	80.96 18	91.13 15	63.30 4	84.04 20	71.81 39	82.39 54	89.59 68	69.16 24	89.08 26	88.83 14	91.49 13	89.10 9

ACMP	70.35 9	82.17 14	86.45 13	77.18 18	79.33 28	81.00 17	89.27 28	58.63 32	81.35 44	75.46 30	82.97 49	95.08 12	68.90 25	90.49 14	87.43 22	91.48 14	86.84 31
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
OPM-MVS		81.44 22	85.68 24	76.49 21	79.27 29	78.21 34	89.84 23	58.67 31	85.25 11	76.26 26	85.28 31	92.88 25	66.03 33	87.20 38	85.40 28	88.86 32	85.58 41

APDe-MVS		81.08 25	86.12 17	75.20 27	79.25 30	80.91 19	90.38 21	57.05 34	85.83 9	66.07 73	87.34 18	91.27 41	69.45 20	85.99 44	82.55 45	88.98 31	88.95 12

NCCC		77.82 41	80.72 56	74.43 30	79.24 31	75.72 45	88.06 34	56.36 39	79.61 54	73.22 35	67.75 131	87.05 91	63.09 47	85.62 45	84.00 37	86.62 47	85.30 44

ACMMP_NAP		81.79 20	85.72 23	77.21 17	79.15 32	79.68 28	91.62 14	59.66 28	83.55 24	77.74 19	83.72 41	87.34 88	65.36 35	88.61 29	87.56 21	89.73 27	89.58 5

MSP-MVS		79.65 32	84.28 39	74.25 31	78.92 33	81.86 9	89.07 29	60.49 25	83.85 22	70.05 53	85.12 32	90.92 49	62.99 48	81.15 73	81.64 54	83.99 64	85.42 43

LGP-MVS_train		82.91 8	86.50 12	78.72 9	78.72 34	81.03 16	89.78 24	61.16 18	80.15 52	80.44 6	84.83 34	94.19 14	70.52 18	90.70 11	87.19 23	91.71 8	87.37 27

DeepC-MVS	73.80 3	82.34 12	86.87 7	77.06 19	78.62 35	84.34 1	90.30 22	63.54 2	83.10 29	71.30 44	86.91 21	90.54 58	67.12 30	87.81 35	87.05 24	91.46 15	88.37 21
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
DPE-MVS		81.01 26	85.18 29	76.15 22	78.58 36	80.64 23	89.77 25	57.92 33	81.66 41	73.45 33	86.84 22	89.80 66	69.33 22	85.40 46	82.91 44	87.87 39	89.01 11

CNVR-MVS		77.79 42	81.57 51	73.38 37	78.37 37	75.91 43	87.97 35	55.11 49	79.41 55	70.98 47	74.70 106	86.43 99	61.77 54	85.10 51	83.73 39	86.10 53	85.68 40

TSAR-MVS + ACMM		81.20 24	86.92 6	74.52 29	77.60 38	82.29 5	84.41 47	62.95 8	82.99 30	64.03 77	87.71 15	89.17 70	71.98 11	88.19 31	88.10 17	86.18 51	89.95 3

HQP-MVS		75.81 55	78.91 63	72.18 42	77.41 39	75.38 47	84.75 44	53.35 57	76.12 67	73.32 34	69.48 123	88.07 77	57.76 75	79.42 84	78.44 65	86.48 48	85.50 42

CDPH-MVS		77.22 48	81.05 55	72.75 40	77.29 40	77.46 37	86.36 41	54.02 55	73.00 84	69.75 55	77.78 90	88.90 73	61.31 58	84.09 56	82.54 46	87.79 41	83.57 46

TSAR-MVS + GP.		73.42 65	76.31 75	70.05 52	77.15 41	71.13 70	81.59 63	54.11 54	69.84 108	58.65 88	66.20 143	78.77 134	65.29 37	83.65 58	83.14 42	83.54 70	81.47 56

PMVS	70.37 8	81.82 19	87.08 5	75.68 25	77.06 42	77.23 38	87.77 37	56.25 41	83.33 27	67.18 69	89.48 12	87.94 79	77.70 1	93.02 2	92.57 2	88.13 37	86.00 37
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
AdaColmap		74.73 60	77.57 69	71.40 44	76.90 43	75.76 44	84.54 46	53.08 59	76.20 66	66.64 72	66.06 144	78.16 137	61.32 57	85.37 48	82.20 51	85.95 55	79.27 64

UniMVSNet_ETH3D		71.84 69	81.36 52	60.74 103	76.46 44	66.01 83	66.49 136	60.24 27	86.58 7	41.87 158	90.04 9	96.02 5	43.72 147	85.14 50	77.30 74	75.64 125	68.40 128

DeepC-MVS_fast	71.40 6	78.48 37	82.92 45	73.31 38	76.44 45	82.23 6	87.59 38	56.56 38	77.79 60	68.91 60	77.00 92	87.32 89	61.90 53	85.40 46	84.37 33	88.46 34	86.33 35
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
DVP-MVS		80.31 29	85.60 26	74.15 33	76.23 46	78.39 32	86.62 39	55.79 44	86.47 8	71.32 43	90.96 6	89.02 72	69.28 23	84.62 54	81.64 54	85.66 56	88.09 23

WR-MVS		78.32 38	86.09 19	69.25 58	76.22 47	72.33 67	85.71 43	59.02 29	86.66 6	51.41 114	92.91 1	96.76 1	53.09 102	90.21 18	85.30 29	90.05 24	78.46 72

COLMAP_ROB	75.87 2	84.34 2	89.80 2	77.97 12	75.52 48	82.76 4	90.39 20	54.21 53	89.37 2	83.18 2	89.90 10	95.58 11	72.34 10	92.31 4	90.04 5	92.17 5	88.61 18
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
LS3D		79.33 34	84.03 41	73.84 34	75.37 49	78.09 35	83.30 51	52.94 60	84.42 17	76.01 27	84.16 37	87.44 87	65.34 36	86.30 41	82.08 52	90.09 23	85.70 39

TDRefinement		86.29 1	90.77 1	81.06 1	75.10 50	83.76 2	93.79 1	61.08 19	89.57 1	86.19 1	90.06 8	93.01 23	76.72 2	94.71 1	92.72 1	93.47 1	91.56 1

DPM-MVS		71.35 73	73.50 99	68.84 59	74.93 51	73.35 57	84.07 48	50.56 65	71.91 88	67.06 70	61.21 166	77.02 142	52.64 105	74.15 107	75.14 86	83.79 67	81.74 54

ambc			79.96 59		74.57 52	75.48 46	73.75 106		80.32 51	72.34 38	78.46 85	92.41 28	59.05 71	80.24 82	73.95 91	75.41 127	78.85 68

MVS_0304		73.74 64	77.16 72	69.74 55	74.24 53	73.47 56	84.70 45	49.62 68	62.26 135	67.27 67	75.87 97	87.57 84	57.49 78	81.20 72	79.50 62	85.10 58	80.27 60

DeepPCF-MVS	71.57 5	79.49 33	84.05 40	74.17 32	74.14 54	80.88 20	89.33 27	56.24 42	82.41 35	71.58 41	82.27 55	86.47 98	66.47 32	84.80 52	84.16 36	87.26 43	87.34 28

CLD-MVS		71.24 75	78.12 66	63.20 92	74.03 55	71.60 68	82.82 56	32.91 180	74.23 77	69.32 57	79.65 78	91.54 36	47.02 134	81.22 71	79.01 63	73.09 140	69.63 117
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
EPNet		66.87 102	68.89 126	64.53 86	73.97 56	61.13 116	78.46 74	61.03 20	56.78 158	53.41 105	66.91 138	70.91 158	43.49 148	76.08 101	76.68 79	76.81 116	73.73 92
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
ACMH+	67.97 10	80.15 30	86.16 15	73.14 39	73.82 57	76.41 41	83.59 50	54.82 51	87.35 5	70.86 48	86.98 20	96.27 4	66.50 31	89.17 25	83.39 40	89.26 29	83.56 47

MCST-MVS		73.84 63	77.44 70	69.63 56	73.75 58	74.73 50	81.38 64	48.58 74	74.77 71	69.16 58	71.97 119	86.20 102	59.50 68	78.51 89	74.06 90	85.42 57	81.85 52

CSCG		76.95 51	82.08 49	70.97 45	73.32 59	78.35 33	81.08 65	47.19 83	83.47 25	69.82 54	80.44 70	87.19 90	64.59 41	81.01 76	77.26 75	89.83 26	86.84 31

OMC-MVS		79.95 31	85.28 28	73.74 36	72.95 60	80.10 27	87.87 36	48.13 78	84.62 14	79.42 13	80.27 72	92.49 27	64.14 42	87.25 37	85.11 31	89.92 25	87.10 30

CANet		71.07 76	75.09 84	66.39 75	72.57 61	71.53 69	82.38 58	47.10 84	59.81 141	59.81 84	74.97 102	84.37 114	54.25 96	79.89 83	77.64 73	82.25 84	77.40 76

TranMVSNet+NR-MVSNet		71.66 70	79.23 62	62.83 95	72.54 62	65.64 85	74.77 96	55.27 47	75.91 68	45.50 146	89.55 11	94.25 13	45.96 138	82.74 65	77.03 76	82.96 76	69.48 121

UniMVSNet_NR-MVSNet		70.82 80	77.44 70	63.11 93	71.75 63	66.02 82	73.93 103	55.00 50	70.90 94	46.77 140	86.68 23	91.54 36	46.91 135	81.07 75	76.32 81	84.28 62	69.54 118

PCF-MVS	65.25 14	73.99 62	76.74 74	70.79 47	71.61 64	75.33 48	83.76 49	50.40 67	74.88 70	74.50 32	67.60 132	85.36 108	58.30 73	78.61 88	74.25 89	86.15 52	81.13 58
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
TSAR-MVS + COLMAP		75.85 54	81.06 53	69.77 54	71.15 65	76.90 39	82.93 54	52.43 62	79.25 57	70.13 51	82.78 50	87.00 93	60.02 63	80.30 80	79.61 61	81.95 87	81.61 55

PHI-MVS		75.17 58	78.37 64	71.43 43	71.13 66	72.46 65	82.28 59	50.55 66	73.39 82	79.05 14	73.65 110	87.50 86	61.98 52	81.10 74	78.48 64	83.60 69	81.99 51

DU-MVS		71.03 78	77.92 67	62.98 94	70.81 67	65.48 87	73.93 103	56.76 36	69.95 106	46.77 140	85.70 29	93.49 17	46.91 135	83.47 59	77.82 71	82.72 79	69.54 118

Baseline_NR-MVSNet		68.15 99	75.12 83	60.02 106	70.81 67	55.67 146	75.88 88	53.40 56	71.25 91	43.96 151	85.88 27	92.68 26	45.76 139	83.47 59	68.34 128	70.34 157	68.58 126

CNLPA		76.67 52	81.72 50	70.77 48	70.75 69	76.68 40	86.14 42	46.11 93	81.82 39	74.68 31	76.37 94	86.23 101	62.92 50	85.28 49	83.29 41	84.02 63	82.40 50

EG-PatchMatch MVS		71.50 72	76.82 73	65.30 82	70.74 70	66.50 80	74.23 100	43.25 115	72.02 87	59.11 86	79.85 76	86.88 96	63.95 43	80.29 81	75.25 85	80.51 93	76.98 79

3Dnovator+	72.94 4	78.78 35	83.05 44	73.80 35	70.70 71	81.34 11	88.33 33	56.01 43	81.33 45	72.87 37	78.06 88	81.15 122	63.83 44	87.39 36	85.82 26	91.06 18	86.28 36

MSLP-MVS++		76.66 53	82.32 48	70.06 51	70.51 72	80.27 25	79.77 71	55.58 46	77.79 60	63.09 78	67.25 135	89.50 69	71.01 14	88.10 32	85.74 27	80.39 94	87.56 26

PLC	64.88 15	75.76 56	80.22 58	70.57 50	70.46 73	77.75 36	82.01 60	48.84 73	80.74 49	70.85 49	71.32 120	84.82 111	63.69 45	84.73 53	82.35 49	87.54 42	79.80 62
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
UniMVSNet (Re)		71.29 74	78.14 65	63.30 90	70.29 74	66.57 79	75.98 86	54.74 52	70.20 101	46.20 144	85.08 33	93.21 19	48.19 125	82.50 67	78.33 67	84.40 61	71.08 109

Effi-MVS+-dtu		70.10 87	73.76 97	65.82 77	70.23 75	74.92 49	79.47 73	44.49 107	56.98 156	54.34 98	64.26 150	84.78 112	59.97 64	80.96 77	80.38 57	86.44 49	74.05 91

MAR-MVS		70.00 88	72.28 110	67.34 67	69.89 76	72.57 64	80.09 70	49.49 70	60.28 140	69.03 59	59.29 176	80.79 124	54.68 94	78.39 91	76.00 82	80.87 91	78.67 70
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
ACMH	66.19 11	78.12 39	84.55 37	70.63 49	69.62 77	72.40 66	80.77 67	46.43 91	89.24 3	77.99 17	87.42 17	95.83 9	62.95 49	86.27 42	78.24 68	86.00 54	82.46 49
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
abl_6				65.41 80	69.37 78	74.02 54	82.50 57	47.39 82	66.39 121	56.63 94	60.61 169	82.76 119	53.68 99			82.92 77	78.39 73

EIA-MVS		63.24 122	64.16 146	62.16 96	69.30 79	63.20 105	72.40 108	40.82 139	48.31 191	51.50 112	59.63 174	62.23 178	57.33 79	78.00 92	71.94 102	81.59 90	65.82 140

FPMVS		70.46 84	74.89 85	65.28 83	69.09 80	61.42 114	77.07 83	46.92 87	76.73 65	53.53 104	67.33 133	75.07 148	67.23 29	83.41 61	81.54 56	77.86 112	78.73 69

IS_MVSNet		68.20 98	74.41 93	60.96 101	68.55 81	64.36 100	71.47 111	48.33 76	70.11 103	43.30 154	80.90 67	74.54 150	47.19 133	81.25 70	77.97 70	86.94 45	71.76 107

MVS_111021_HR		72.37 67	76.12 78	68.00 61	68.55 81	64.30 102	82.93 54	48.98 72	74.25 76	65.39 74	73.59 111	84.11 115	59.48 69	82.61 66	78.38 66	82.66 80	75.59 84

CS-MVS		64.89 113	64.54 143	65.31 81	68.39 83	61.63 112	75.90 87	43.28 114	54.33 164	63.06 79	55.59 188	62.41 177	56.76 81	77.69 93	70.80 111	84.46 60	69.33 122

EPP-MVSNet		69.51 90	76.17 76	61.74 100	68.38 84	66.60 78	71.77 109	46.98 85	73.60 81	41.79 159	82.06 58	69.65 163	52.51 106	83.41 61	79.94 60	89.02 30	77.94 74

ETV-MVS		63.96 118	65.51 136	62.16 96	67.69 85	61.58 113	75.47 91	42.28 120	56.52 159	48.60 129	61.95 162	56.87 188	55.63 86	78.78 87	72.67 98	83.29 73	70.17 112

TAPA-MVS	66.11 12	75.37 57	79.24 61	70.86 46	67.63 86	74.09 53	83.17 53	44.75 106	81.82 39	80.83 5	65.61 146	88.04 78	61.58 55	83.21 63	80.12 58	87.17 44	81.82 53
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
MVS_111021_LR		71.60 71	75.21 82	67.38 66	67.42 87	62.44 108	81.73 61	46.24 92	70.89 95	66.80 71	73.19 113	84.98 109	60.09 62	81.94 69	77.77 72	82.00 86	75.29 85

Effi-MVS+		69.04 92	73.01 103	64.40 87	67.20 88	64.83 95	74.87 95	43.97 110	63.33 131	60.90 82	73.06 114	85.79 105	55.61 87	73.58 113	76.41 80	83.84 66	74.09 90

MSDG		68.98 94	73.31 102	63.92 88	67.08 89	68.27 74	75.41 92	40.77 140	67.61 115	64.89 75	75.75 98	78.96 131	53.70 98	76.72 97	73.95 91	81.71 89	71.93 106

v7n		74.47 61	81.06 53	66.77 70	66.98 90	67.10 76	76.76 84	45.88 95	81.98 38	67.43 66	88.38 13	95.67 10	61.38 56	80.76 78	73.49 93	82.21 85	80.06 61

v1240		70.94 79	74.52 90	66.76 71	66.54 91	64.40 98	77.76 78	45.29 98	70.05 104	71.45 42	83.36 45	90.96 47	60.37 60	70.50 143	68.68 120	79.14 103	73.68 93

v1192		71.06 77	74.87 86	66.61 72	66.38 92	65.80 84	78.27 75	45.28 99	70.19 102	70.79 50	83.37 43	91.79 34	58.76 72	70.86 140	69.02 119	80.16 96	73.08 96

v1921920		70.82 80	74.46 92	66.58 73	66.33 93	64.35 101	77.72 79	45.07 101	70.39 98	71.18 45	83.15 47	90.62 56	59.97 64	70.90 138	68.43 127	79.19 102	73.39 94

Anonymous202405211			72.22 111		66.19 94	61.09 118	62.23 155	45.87 96	71.25 91		79.33 83	86.16 104	37.36 166	73.54 114	69.84 117	75.45 126	64.32 146

TinyColmap		71.85 68	76.11 79	66.87 69	66.07 95	65.34 89	74.35 98	49.30 71	79.93 53	75.93 28	75.66 99	87.74 81	54.72 93	80.66 79	70.42 113	80.85 92	73.02 98

gm-plane-assit		56.76 153	57.64 172	55.73 131	66.01 96	55.45 149	74.96 94	30.54 190	73.71 80	56.04 96	81.81 61	30.91 214	43.83 145	58.77 180	54.71 178	63.02 172	48.13 188

v144192		70.68 83	74.40 94	66.34 76	65.94 97	64.38 99	77.63 80	45.18 100	69.97 105	70.11 52	82.70 52	90.77 50	59.84 66	71.43 136	68.46 123	79.31 101	73.08 96

canonicalmvs		66.37 103	74.37 95	57.04 124	65.89 98	65.06 90	62.58 153	42.55 117	76.82 64	46.87 139	67.33 133	86.38 100	45.49 141	76.77 96	71.85 103	78.87 106	76.35 80

v1144		70.45 85	74.50 91	65.73 78	65.74 99	64.88 93	77.33 82	44.16 108	70.59 97	69.63 56	83.15 47	91.42 39	57.79 74	71.29 137	68.53 122	79.72 99	71.63 108

NR-MVSNet		68.66 96	76.15 77	59.93 107	65.49 100	65.48 87	74.42 97	56.76 36	69.95 106	45.38 147	85.70 29	91.13 42	34.68 176	74.52 106	76.75 78	82.83 78	69.49 120

IterMVS-LS		65.76 107	70.85 120	59.81 109	65.33 101	57.78 131	64.63 146	48.02 79	65.65 124	51.05 117	81.31 65	77.47 140	54.94 92	69.46 151	69.36 118	74.90 129	74.95 86
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
v10		70.25 86	74.59 89	65.19 84	65.32 102	66.46 81	76.60 85	44.84 104	67.38 117	67.21 68	82.75 51	90.56 57	57.70 76	71.69 133	68.63 121	79.44 100	74.67 88

FC-MVSNet-train		64.87 114	74.76 88	53.33 137	65.24 103	58.05 128	69.69 121	41.92 127	70.99 93	32.62 181	85.75 28	88.23 75	32.10 186	77.61 94	74.41 88	78.43 107	68.25 129

RPSCF		77.56 43	84.51 38	69.46 57	65.17 104	74.36 51	79.74 72	47.45 81	84.01 21	72.89 36	77.89 89	90.67 54	65.14 38	88.25 30	89.74 7	86.38 50	86.64 33

SixPastTwentyTwo		77.24 47	83.65 43	69.78 53	65.14 105	64.85 94	77.44 81	47.74 80	82.76 32	68.52 61	87.65 16	93.31 18	71.68 13	89.49 24	82.41 48	88.14 36	85.05 45

test20.03		53.49 166	60.95 163	44.78 179	64.73 106	47.25 177	61.58 157	43.30 113	65.86 123	22.85 202	66.87 140	79.85 127	22.99 198	62.38 172	56.95 173	53.25 189	47.46 190

Vis-MVSNet (Re-imp)		60.99 138	67.78 129	53.06 139	64.66 107	53.49 159	67.40 131	49.52 69	68.55 113	28.00 194	79.53 82	71.41 157	33.08 184	75.30 103	71.28 105	75.69 124	54.91 172

TransMVSNet (Re)		63.49 120	73.86 96	51.39 149	64.26 108	56.07 143	61.17 158	42.23 121	78.81 58	34.80 172	85.94 25	90.63 55	34.35 180	72.73 123	67.98 134	71.50 147	64.84 144

v2v482		69.01 93	73.39 101	63.89 89	63.86 109	62.99 106	75.26 93	42.05 124	70.22 100	68.46 62	82.64 53	91.61 35	55.38 88	70.89 139	66.93 141	78.30 108	68.48 127

Fast-Effi-MVS+		67.71 100	72.54 107	62.07 98	63.83 110	63.68 103	75.74 90	39.94 146	60.89 139	54.29 99	73.00 115	86.19 103	56.85 80	78.46 90	73.23 94	81.74 88	72.36 102

DCV-MVSNet		64.34 117	72.84 105	54.42 135	63.79 111	62.09 110	62.50 154	42.72 116	74.32 75	41.34 161	66.96 136	88.57 74	39.18 159	75.20 104	70.35 114	77.01 115	72.37 101

v8		68.77 95	73.50 99	63.26 91	63.74 112	64.47 97	74.22 101	42.07 123	67.30 118	64.89 75	82.08 57	90.23 60	56.50 84	71.85 132	66.57 142	78.14 109	72.02 104

Anonymous20231211		63.69 119	72.86 104	53.00 140	63.72 113	60.25 123	60.33 160	40.96 136	72.49 85	38.91 164	81.77 63	88.17 76	37.60 165	73.30 115	68.01 133	76.47 121	66.06 139

Gipuma		73.40 66	79.27 60	66.55 74	63.64 114	59.35 124	70.28 118	45.92 94	83.79 23	71.78 40	84.04 39	93.07 22	68.69 27	87.90 34	76.76 77	78.98 105	69.96 114
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
LTVRE_ROB	75.99 1	82.04 18	87.16 4	76.07 23	63.57 115	70.27 72	86.48 40	62.99 7	89.00 4	80.32 7	86.25 24	91.04 45	74.66 4	92.58 3	90.29 4	88.42 35	90.72 2
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
PVSNet_Blended_VisFu		70.70 82	73.62 98	67.28 68	63.53 116	72.96 58	77.97 76	52.10 63	63.65 129	62.66 80	71.14 121	73.46 152	63.55 46	79.35 86	75.34 84	83.90 65	79.43 63

USDC		68.53 97	71.82 115	64.68 85	63.53 116	61.87 111	70.12 119	46.98 85	77.89 59	76.58 24	68.55 128	86.88 96	50.50 113	73.73 110	65.62 144	80.39 94	68.21 130

gg-mvs-nofinetune		56.45 158	61.04 162	51.10 152	63.42 118	49.40 173	53.71 185	52.52 61	74.77 71	46.93 137	77.31 91	53.88 190	26.42 196	62.51 171	57.81 171	63.60 171	51.57 180

pmmvs-eth3d		65.36 111	70.09 123	59.85 108	63.05 119	53.61 158	74.29 99	46.45 90	68.14 114	51.45 113	78.83 84	85.78 106	49.87 118	70.44 144	70.45 112	74.00 133	63.38 150

casdiffmvs		66.19 105	72.34 108	59.02 111	62.75 120	60.61 122	69.06 124	41.38 132	69.49 109	54.11 100	84.00 40	89.74 67	49.12 120	70.74 142	62.70 159	77.70 114	69.14 124

pmmvs6		64.78 115	75.82 80	51.89 143	62.41 121	57.13 136	60.24 161	45.59 97	82.90 31	34.69 174	84.83 34	93.18 20	36.22 170	76.43 98	71.13 107	72.21 144	65.12 143

DI_MVS_plusplus_trai		63.43 121	67.54 130	58.63 113	62.34 122	58.06 127	65.75 142	42.15 122	63.05 132	53.28 106	75.88 96	75.92 145	50.18 115	68.04 154	64.20 150	78.07 110	67.65 131

CDS-MVSNet		57.90 150	63.57 150	51.28 150	62.30 123	53.17 160	64.70 145	51.61 64	57.41 152	32.75 180	63.73 152	70.53 159	27.12 192	72.49 124	73.02 96	69.22 161	54.68 173
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
tfpnnormal		61.41 133	71.33 118	49.83 158	61.73 124	54.90 154	58.52 169	41.24 133	75.20 69	32.00 186	82.13 56	87.87 80	35.63 174	72.75 122	66.30 143	69.87 158	60.14 158

OpenMVS	60.79 16	64.42 116	69.72 124	58.23 120	61.63 125	62.17 109	64.11 148	37.54 164	67.17 119	55.71 97	65.89 145	74.89 149	52.67 104	72.20 128	68.29 130	77.73 113	77.39 77

thres600view7		58.87 147	65.91 134	50.66 155	61.27 126	56.32 140	59.88 164	40.63 142	64.88 126	32.10 185	64.82 147	69.83 161	36.72 168	72.99 119	72.55 99	73.34 136	59.97 159

Fast-Effi-MVS+-dtu		63.22 123	65.55 135	60.49 105	61.24 127	64.70 96	74.15 102	53.24 58	51.46 171	49.67 125	58.03 182	78.42 135	48.05 127	72.03 129	71.14 106	76.60 120	63.09 151

pmmvs4		61.12 136	64.61 142	57.04 124	60.88 128	52.15 163	70.59 117	44.82 105	61.35 138	46.91 138	72.08 117	73.27 153	46.79 137	65.06 160	67.76 137	72.28 142	60.58 157

3Dnovator	65.69 13	69.43 91	75.74 81	62.06 99	60.78 129	70.50 71	75.85 89	39.57 149	74.44 73	57.41 91	75.91 95	77.73 139	55.34 89	76.86 95	75.61 83	83.44 71	79.14 66

Vis-MVSNet		69.95 89	77.69 68	60.91 102	60.67 130	66.71 77	77.94 77	48.58 74	69.10 111	45.78 145	80.21 73	83.58 117	53.41 101	82.92 64	80.11 59	79.08 104	81.21 57
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
pm-mvs1		61.97 128	72.01 114	50.25 157	60.64 131	55.23 150	58.67 168	42.44 118	74.40 74	33.63 178	81.03 66	89.86 64	34.87 175	72.93 121	67.95 135	71.28 148	62.65 152

MS-PatchMatch		56.31 160	60.22 167	51.73 146	60.53 132	55.53 148	63.41 150	37.18 165	51.34 174	37.44 166	60.53 170	62.19 179	45.52 140	64.25 166	63.17 157	66.33 166	64.56 145

thres400		57.25 152	63.73 148	49.70 160	60.19 133	54.95 153	58.16 170	39.60 148	62.42 133	31.98 188	62.33 159	69.20 164	35.96 172	70.07 148	68.03 132	72.28 142	59.12 162

CANet_DTU		61.22 135	67.07 132	54.40 136	59.89 134	63.62 104	70.98 114	36.77 167	50.49 178	47.15 135	62.45 158	80.81 123	37.90 164	71.87 131	70.09 116	73.69 134	70.19 111

v148		64.92 112	70.58 122	58.32 117	59.89 134	57.09 137	66.04 138	35.27 170	69.11 110	60.66 83	79.57 79	90.93 48	53.91 97	69.81 150	62.22 160	74.14 131	65.31 142

DELS-MVS		65.54 110	71.79 116	58.24 119	59.68 136	65.55 86	70.99 113	38.69 155	62.29 134	49.27 127	75.03 101	81.42 121	50.93 109	73.71 112	71.35 104	79.90 98	73.20 95
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
QAPM		66.36 104	72.76 106	58.90 112	59.57 137	65.01 91	64.05 149	41.17 135	73.09 83	56.82 93	69.42 124	77.78 138	55.07 91	73.00 118	72.07 101	76.71 117	78.96 67

MVS_Test		62.58 126	67.46 131	56.89 126	59.52 138	55.90 145	64.94 144	38.83 152	57.08 155	56.55 95	76.53 93	84.49 113	47.45 129	66.95 157	62.01 161	74.04 132	69.27 123

test-LLR		46.01 187	45.06 203	47.11 174	59.39 139	36.72 197	51.28 190	40.95 137	36.41 208	34.45 175	46.14 202	47.02 203	38.00 162	51.78 196	48.53 191	58.60 176	48.84 185

test0.0.03 1		45.40 189	49.55 191	40.57 188	59.39 139	44.36 185	53.37 186	40.95 137	47.14 192	19.23 205	45.49 204	60.24 184	19.24 202	54.82 191	51.98 185	51.21 193	42.82 197

GA-MVS		60.73 140	64.24 145	56.64 128	59.38 141	57.45 135	65.07 143	36.65 169	57.39 153	58.17 89	73.43 112	69.10 166	47.38 130	64.47 165	63.63 154	73.19 138	64.22 147

PatchMatch-RL		60.96 139	63.00 155	58.57 115	59.16 142	52.18 162	67.38 132	41.99 125	57.85 150	48.16 130	53.55 195	69.77 162	59.47 70	73.73 110	72.49 100	75.27 128	61.44 155

PM-MVS		65.66 108	71.25 119	59.14 110	58.92 143	54.88 155	73.66 107	38.55 156	66.12 122	49.91 124	69.87 122	86.97 94	60.61 59	76.30 99	74.75 87	73.19 138	69.83 115

thres200		56.35 159	62.36 156	49.34 162	58.87 144	56.32 140	55.91 178	40.63 142	58.51 144	31.34 189	58.81 180	67.31 168	35.96 172	72.99 119	65.51 145	73.34 136	57.07 165

baseline1		52.90 170	58.38 169	46.51 178	58.87 144	50.01 171	54.17 183	40.45 145	56.81 157	29.25 193	62.72 157	58.99 186	30.25 188	65.05 161	60.57 164	66.07 167	54.54 174

ET-MVSNet_ETH3D		60.33 142	62.10 158	58.27 118	58.61 146	58.05 128	68.06 128	41.20 134	51.40 172	51.10 116	64.06 151	49.42 196	50.61 112	74.72 105	70.29 115	80.05 97	66.74 135

thres100view900		53.88 163	59.19 168	47.68 172	58.57 147	52.74 161	54.45 182	38.07 159	53.23 165	31.01 190	56.41 185	66.40 173	32.80 185	65.03 162	64.43 149	71.18 151	56.10 168

tfpn200view9		56.07 161	61.85 159	49.34 162	58.57 147	56.48 139	58.01 172	40.72 141	53.23 165	31.01 190	56.41 185	66.40 173	34.18 181	73.02 117	68.06 131	73.53 135	59.35 161

testgi		51.94 173	61.37 161	40.94 186	58.38 149	47.03 179	65.88 139	30.49 192	70.87 96	22.64 203	57.53 184	87.59 83	18.30 204	63.01 168	54.32 180	49.93 195	49.27 183

FMVSNet1		61.92 129	71.36 117	50.90 153	57.67 150	59.29 125	59.48 166	44.14 109	70.24 99	34.72 173	75.45 100	84.94 110	36.75 167	72.33 125	68.45 124	72.66 141	68.83 125

IterMVS-SCA-FT		62.67 124	68.00 128	56.45 130	56.92 151	64.92 92	57.51 173	38.12 157	59.44 142	53.62 103	74.74 105	71.60 155	64.84 39	70.24 146	65.27 146	67.70 165	69.83 115

IB-MVS	57.02 17	61.37 134	65.39 137	56.69 127	56.65 152	60.85 121	70.70 116	37.90 162	49.37 185	45.37 148	48.75 200	79.14 130	53.55 100	76.26 100	70.85 110	75.97 122	72.50 100
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
thisisatest0515		66.95 101	72.29 109	60.72 104	56.37 153	56.05 144	71.08 112	38.81 153	67.59 116	53.26 107	78.21 86	79.79 129	60.11 61	75.69 102	73.02 96	84.69 59	75.66 83

PVSNet_BlendedMVS		61.75 130	65.07 140	57.87 122	56.27 154	60.99 119	65.81 140	43.75 111	51.27 175	54.08 101	62.12 160	78.84 132	50.67 110	71.49 134	63.91 152	76.64 118	66.86 133

PVSNet_Blended		61.75 130	65.07 140	57.87 122	56.27 154	60.99 119	65.81 140	43.75 111	51.27 175	54.08 101	62.12 160	78.84 132	50.67 110	71.49 134	63.91 152	76.64 118	66.86 133

diffmvs		62.64 125	69.66 125	54.46 134	56.19 156	55.06 152	67.36 133	36.74 168	64.18 128	50.58 118	79.54 80	87.55 85	45.13 143	68.04 154	63.20 155	70.78 152	70.02 113

GBi-Net		56.54 156	63.26 153	48.70 167	55.88 157	57.61 132	57.26 174	41.75 128	49.06 186	32.37 182	61.81 163	67.02 169	34.58 177	72.33 125	68.45 124	70.38 154	66.56 136

test1		56.54 156	63.26 153	48.70 167	55.88 157	57.61 132	57.26 174	41.75 128	49.06 186	32.37 182	61.81 163	67.02 169	34.58 177	72.33 125	68.45 124	70.38 154	66.56 136

FMVSNet2		57.80 151	65.39 137	48.94 166	55.88 157	57.61 132	57.26 174	42.37 119	58.21 145	33.19 179	68.36 129	75.55 147	34.58 177	66.91 158	64.55 148	70.38 154	66.56 136

tpm cat1		50.98 174	51.28 185	50.62 156	55.74 160	49.92 172	63.13 151	38.12 157	52.38 169	47.61 133	60.11 171	44.51 209	44.86 144	51.31 198	47.49 194	54.25 188	53.24 176

new-patchmatchnet		47.33 183	60.49 166	31.99 201	55.69 161	33.86 202	36.84 209	33.31 179	72.36 86	14.33 208	80.09 74	92.14 31	13.27 206	63.54 167	40.09 201	38.51 203	41.32 201

EPNet_dtu		56.63 155	60.77 164	51.80 144	55.47 162	44.63 183	69.83 120	38.74 154	50.27 179	47.64 132	58.01 183	72.27 154	33.71 182	68.60 153	67.72 138	65.39 168	63.86 148
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
Anonymous20231206		50.28 178	57.94 171	41.35 185	55.45 163	43.65 187	58.06 171	34.12 174	62.02 137	24.25 200	59.33 175	79.80 128	24.49 197	59.55 174	54.28 181	51.74 191	46.94 192

baseline2		53.55 165	55.19 176	51.62 147	55.27 164	51.95 164	60.89 159	34.23 173	46.69 195	42.47 155	53.56 194	50.01 193	45.33 142	64.63 164	61.22 163	71.56 145	58.28 164

tttt0517		61.44 132	63.85 147	58.62 114	55.20 165	55.61 147	68.80 125	38.02 160	55.70 161	50.01 123	66.93 137	48.90 197	56.69 82	73.84 109	71.10 108	82.99 75	74.89 87

CVMVSNet		60.45 141	63.72 149	56.63 129	54.82 166	53.75 156	68.41 127	41.95 126	55.07 163	48.03 131	58.08 181	68.67 167	55.09 90	69.14 152	68.34 128	71.51 146	72.97 99

thisisatest0530		61.02 137	63.44 152	58.19 121	54.75 167	55.09 151	68.03 130	38.02 160	55.45 162	49.06 128	66.58 141	48.69 198	56.69 82	73.07 116	71.10 108	82.60 81	74.14 89

FC-MVSNet-test		60.28 143	70.83 121	47.96 170	54.69 168	47.12 178	68.06 128	41.68 131	71.42 89	23.73 201	84.70 36	77.41 141	28.92 190	82.33 68	73.08 95	70.68 153	59.77 160

UGNet		65.61 109	74.79 87	54.91 133	54.54 169	68.20 75	70.97 115	48.21 77	67.14 120	41.67 160	74.15 107	80.65 125	36.10 171	79.39 85	77.99 69	77.95 111	76.01 81
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
FMVSNet3		54.77 162	60.73 165	47.81 171	54.29 170	56.88 138	55.89 179	41.75 128	49.06 186	32.37 182	61.81 163	67.02 169	33.67 183	62.88 169	61.96 162	68.88 162	65.53 141

our_test_3					52.72 171	53.66 157	69.11 123

CostFormer		52.59 171	55.14 177	49.61 161	52.72 171	50.40 167	66.28 137	33.78 176	52.85 167	43.43 152	66.30 142	51.37 192	41.78 155	54.92 190	51.18 187	59.68 175	58.98 163

HyFIR lowres test		59.15 146	62.28 157	55.49 132	52.42 173	62.59 107	71.76 110	39.74 147	50.25 180	41.92 157	62.88 156	69.16 165	55.85 85	62.77 170	67.18 140	71.27 149	61.11 156

dps		49.71 179	51.97 183	47.07 175	52.37 174	47.00 180	53.02 188	40.52 144	44.91 199	41.23 162	64.55 148	44.27 210	40.12 158	57.71 183	51.97 186	55.14 186	53.41 175

V42		65.79 106	72.11 113	58.42 116	51.89 175	58.69 126	73.80 105	34.50 171	65.40 125	57.10 92	79.54 80	89.09 71	57.51 77	71.98 130	67.83 136	75.70 123	72.26 103

DWT-MVSNet_training		48.57 182	47.93 195	49.31 164	51.79 176	48.05 176	61.84 156	34.33 172	41.94 202	43.42 153	50.35 197	34.74 213	47.30 131	52.62 195	52.08 184	57.20 181	55.74 169

CR-MVSNet		53.82 164	55.40 175	51.98 142	51.57 177	50.23 168	45.00 200	44.97 102	46.90 193	52.60 109	67.91 130	46.99 205	48.37 123	59.15 178	59.53 168	69.38 160	57.07 165

IterMVS		59.24 145	64.45 144	53.16 138	50.98 178	61.29 115	66.51 135	32.85 181	58.17 146	46.31 143	72.58 116	70.23 160	54.26 95	64.81 163	60.24 165	68.04 164	63.81 149
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
RPMNet		50.92 175	50.32 187	51.62 147	50.25 179	50.23 168	59.16 167	46.70 89	46.90 193	42.39 156	48.97 199	37.23 212	41.78 155	57.30 185	56.18 175	69.44 159	55.43 170

MDTV_nov1_ep13_2view		58.09 149	63.54 151	51.74 145	50.13 180	46.56 182	66.95 134	33.41 178	63.52 130	58.77 87	74.84 103	84.10 116	43.12 149	65.95 159	54.69 179	58.04 179	55.13 171

tpm		46.67 184	49.20 193	43.72 181	49.60 181	36.60 199	53.93 184	26.84 195	52.70 168	58.05 90	69.04 125	47.96 201	30.06 189	48.33 201	42.76 197	43.88 198	47.01 191

TAMVS		46.64 185	53.62 180	38.49 191	49.56 182	36.87 196	53.16 187	25.76 197	56.33 160	22.55 204	60.72 168	61.80 181	27.12 192	59.50 175	58.33 170	52.79 190	41.82 200

PatchmatchNet		48.67 181	50.10 188	46.99 177	48.29 183	41.00 190	55.54 180	38.94 150	51.38 173	45.15 149	63.22 154	48.45 200	42.83 150	53.80 194	48.50 193	51.19 194	44.37 194
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
CMPMVS	45.32 18	58.10 148	65.24 139	49.76 159	47.88 184	46.86 181	48.16 198	32.82 182	58.06 147	61.35 81	59.64 173	80.00 126	47.27 132	70.15 147	64.10 151	61.08 173	77.85 75
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
EU-MVSNet		59.77 144	66.07 133	52.42 141	47.81 185	51.86 165	62.98 152	32.28 185	62.08 136	47.10 136	59.94 172	83.42 118	53.08 103	70.06 149	63.19 156	71.26 150	71.96 105

CHOSEN 1792x2688		52.99 169	56.91 173	48.42 169	47.32 186	50.10 170	64.18 147	33.85 175	45.46 198	36.95 168	55.20 191	66.49 172	51.20 108	59.28 176	59.81 166	57.01 182	61.99 154

MVSTER		53.08 168	56.39 174	49.21 165	47.19 187	51.08 166	60.14 163	31.74 187	40.63 204	38.97 163	55.78 187	46.74 206	42.47 152	67.29 156	62.99 158	74.73 130	70.23 110

tpmrst		43.31 195	46.14 199	40.02 190	47.05 188	36.48 200	48.01 199	32.17 186	49.50 184	37.26 167	63.66 153	47.04 202	31.98 187	42.00 207	40.55 200	43.64 199	43.75 196

pmmvs5		50.64 176	58.01 170	42.05 184	47.01 189	43.67 186	49.27 196	29.43 193	50.77 177	33.83 177	68.69 127	76.16 144	27.82 191	57.53 184	57.07 172	64.95 169	52.18 178

SCA		52.47 172	53.97 179	50.71 154	46.95 190	57.79 130	60.18 162	46.89 88	51.92 170	46.71 142	60.73 167	49.97 194	47.69 128	56.39 187	52.98 183	55.82 184	48.03 189

MIMVSNet1		56.72 154	68.69 127	42.76 183	46.70 191	42.81 188	69.13 122	30.52 191	81.01 46	32.00 186	74.82 104	91.10 44	26.83 194	73.98 108	64.72 147	51.40 192	52.38 177

MDTV_nov1_ep13		49.60 180	51.57 184	47.31 173	46.28 192	44.61 184	59.82 165	30.96 188	48.80 190	50.20 121	59.26 177	52.38 191	38.56 160	56.20 188	49.70 190	58.04 179	50.01 181

E-PMN		43.83 194	49.81 190	36.84 195	46.09 193	31.86 205	42.77 204	25.85 196	57.76 151	25.53 196	55.50 189	62.47 176	43.77 146	70.78 141	39.51 202	37.04 204	30.79 206

baseline		53.46 167	61.55 160	44.01 180	45.83 194	48.77 174	57.26 174	28.75 194	49.99 181	38.85 165	68.78 126	75.65 146	38.30 161	60.80 173	59.78 167	55.10 187	67.07 132

anonymousdsp		74.76 59	82.59 46	65.63 79	45.61 195	61.13 116	89.06 30	32.58 183	74.11 78	59.55 85	84.06 38	94.12 15	75.24 3	88.94 27	86.95 25	91.74 7	88.81 13

EMVS		43.85 193	49.91 189	36.77 196	45.46 196	32.70 203	44.09 202	25.33 198	57.88 149	26.62 195	58.99 179	61.14 183	42.77 151	70.26 145	38.52 205	36.38 205	29.87 207

MIMVSNet		45.83 188	53.46 182	36.94 194	45.38 197	39.50 193	52.20 189	30.68 189	57.09 154	24.53 199	55.22 190	71.54 156	21.74 200	55.81 189	51.08 188	47.11 196	43.96 195

N_pmnet		39.50 201	51.01 186	26.09 203	44.48 198	25.59 209	40.20 207	21.49 201	64.20 127	7.98 211	73.86 109	76.67 143	13.66 205	50.17 199	36.69 207	28.71 208	29.86 208

EPMVS		40.11 199	44.96 205	34.44 198	41.55 199	32.65 204	41.74 205	32.39 184	49.89 183	24.83 197	64.44 149	46.38 207	26.57 195	44.75 205	39.47 203	39.59 202	37.16 203

MVS-HIRNet		44.56 191	45.52 201	43.44 182	40.98 200	31.03 206	39.52 208	36.96 166	42.80 201	44.37 150	53.80 193	60.04 185	41.85 154	47.97 203	41.08 199	56.99 183	41.95 199

ADS-MVSNet		40.61 197	46.31 197	33.96 199	40.70 201	30.42 207	40.42 206	33.44 177	58.01 148	30.87 192	63.05 155	54.48 189	22.67 199	44.35 206	39.23 204	35.64 206	34.64 204

FMVSNet5		39.83 200	45.08 202	33.71 200	39.24 202	39.56 192	48.77 197	23.55 200	39.45 205	24.55 198	33.73 207	44.57 208	20.97 201	58.27 182	54.23 182	45.16 197	45.77 193

new_pmnet		35.76 203	45.64 200	24.22 204	38.59 203	25.83 208	31.87 211	19.24 204	49.06 186	9.01 210	54.34 192	64.73 175	12.46 207	49.21 200	44.91 195	34.17 207	31.41 205

PatchT		50.55 177	53.55 181	47.05 176	37.59 204	42.26 189	50.55 193	37.56 163	46.37 196	52.60 109	66.91 138	43.54 211	48.37 123	59.15 178	59.53 168	55.62 185	57.07 165

MDA-MVSNet-bldmvs		62.46 127	72.13 112	51.19 151	34.32 205	56.10 142	68.65 126	38.85 151	69.05 112	49.50 126	78.17 87	85.43 107	51.32 107	86.67 39	67.40 139	64.46 170	62.08 153

test-mter		44.18 192	47.60 196	40.18 189	33.20 206	39.03 194	55.28 181	13.91 208	39.07 207	36.63 169	48.09 201	49.52 195	41.12 157	54.55 192	50.91 189	60.97 174	52.03 179

CHOSEN 280x420		40.24 198	44.14 206	35.69 197	32.36 207	23.58 210	50.30 194	21.21 202	40.94 203	18.84 206	32.75 208	48.65 199	48.13 126	59.16 177	55.31 177	43.28 200	48.62 187

TESTMET0.1,1		41.79 196	45.06 203	37.97 192	31.32 208	36.72 197	51.28 190	14.17 207	36.41 208	34.45 175	46.14 202	47.02 203	38.00 162	51.78 196	48.53 191	58.60 176	48.84 185

pmmvs3		46.64 185	54.13 178	37.90 193	31.23 209	40.68 191	49.83 195	15.34 206	46.31 197	36.34 170	53.15 196	74.40 151	36.36 169	58.43 181	56.64 174	58.32 178	49.29 182

PMMVS		45.37 190	49.29 192	40.79 187	27.75 210	35.07 201	50.88 192	19.88 203	39.27 206	35.78 171	50.11 198	61.29 182	42.04 153	54.13 193	55.95 176	68.43 163	49.19 184

PMMVS2		34.11 204	48.55 194	17.26 205	25.45 211	20.72 211	35.08 210	16.26 205	58.71 143	4.16 213	59.22 178	78.40 136	3.65 208	57.24 186	38.31 206	18.94 210	27.28 209

MVE	28.01 19	35.86 202	43.56 207	26.88 202	22.33 212	19.75 212	30.85 212	23.88 199	49.90 182	10.48 209	43.64 205	61.87 180	48.99 122	47.26 204	42.15 198	24.76 209	40.37 202
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
tmp_tt				7.47 207	8.89 213	3.32 214	4.35 214	1.14 210	15.58 211	15.76 207	8.50 210	5.90 216	2.00 209	20.02 208	21.51 208	12.70 211

test123		0.53 206	0.60 209	0.46 208	0.22 214	0.25 215	0.33 217	0.13 212	0.66 212	1.37 214	1.10 211	0.00 218	0.43 210	0.68 210	0.61 209	0.26 213	0.88 211

testmvs		0.47 207	0.69 208	0.21 209	0.17 215	0.17 216	0.35 216	0.16 211	0.66 212	0.18 216	1.05 212	0.99 217	0.27 211	0.62 211	0.54 210	0.15 214	0.77 212

GG-mvs-BLEND		31.54 205	46.27 198	14.37 206	0.07 216	48.65 175	42.97 203	0.08 213	44.04 200	1.21 215	39.77 206	57.94 187	0.15 212	48.19 202	42.82 196	41.70 201	42.46 198

sosnet-low-res		0.00 208	0.00 210	0.00 210	0.00 217	0.00 217	0.00 218	0.00 214	0.00 214	0.00 217	0.00 213	0.00 218	0.00 213	0.00 212	0.00 211	0.00 215	0.00 213

sosnet		0.00 208	0.00 210	0.00 210	0.00 217	0.00 217	0.00 218	0.00 214	0.00 214	0.00 217	0.00 213	0.00 218	0.00 213	0.00 212	0.00 211	0.00 215	0.00 213

test_part1																	88.77 14

MTAPA										80.26 8		90.53 59

MTMP										82.07 4		91.00 46

Patchmatch-RL test							2.05 215

NP-MVS									71.39 90

Patchmtry						37.73 195	45.00 200	44.97 102		52.60 109

DeepMVS_CX						8.52 213	9.75 213	3.19 209	16.70 210	5.02 212	23.06 209	19.33 215	18.69 203	13.75 209		11.34 212	25.07 210

High-resolution multi-view benchmark