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.

DPE-MVS		97.83 2	98.13 2	97.48 3	98.83 22	99.19 2	98.99 1	96.70 1	96.05 18	94.39 9	98.30 1	99.47 2	97.02 6	97.75 5	97.02 12	98.98 2	99.10 7

APDe-MVS		97.79 3	97.96 4	97.60 1	99.20 2	99.10 4	98.88 2	96.68 2	96.81 5	94.64 5	97.84 2	98.02 9	97.24 3	97.74 6	97.02 12	98.97 3	99.16 4

MSP-MVS		97.93 1	98.23 1	97.58 2	99.05 6	99.31 1	98.64 4	96.62 3	97.56 1	95.08 4	96.61 12	99.64 1	97.32 1	97.91 2	97.31 6	98.77 11	99.26 1

APD-MVS		97.12 11	97.05 16	97.19 6	99.04 7	98.63 17	98.45 6	96.54 4	94.81 36	93.50 16	96.10 18	97.40 20	96.81 12	97.05 19	96.82 17	98.80 7	98.56 19
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
xxxxxxxxxxxxxcwj		95.62 30	94.35 44	97.10 8	98.95 15	98.51 26	97.51 28	96.48 5	96.17 14	94.64 5	97.32 4	76.98 133	96.23 25	96.78 25	96.15 36	98.79 9	98.55 24

SF-MVS		97.20 10	97.29 12	97.10 8	98.95 15	98.51 26	97.51 28	96.48 5	96.17 14	94.64 5	97.32 4	97.57 17	96.23 25	96.78 25	96.15 36	98.79 9	98.55 24

HPM-MVS++		97.22 9	97.40 10	97.01 11	99.08 4	98.55 22	98.19 13	96.48 5	96.02 19	93.28 20	96.26 16	98.71 6	96.76 16	97.30 13	96.25 33	98.30 48	98.68 13

DVP-MVS		97.70 4	98.09 3	97.24 5	99.00 11	99.17 3	98.76 3	96.41 8	96.91 3	93.88 14	97.72 3	99.04 5	96.93 10	97.29 14	97.31 6	98.45 30	99.23 2

NCCC		96.75 18	96.67 23	96.85 16	99.03 9	98.44 32	98.15 15	96.28 9	96.32 11	92.39 25	92.16 34	97.55 18	96.68 18	97.32 11	96.65 20	98.55 19	98.26 36

SMA-MVS		97.53 5	97.93 5	97.07 10	99.21 1	99.02 6	98.08 18	96.25 10	96.36 10	93.57 15	96.56 13	99.27 3	96.78 15	97.91 2	97.43 3	98.51 20	98.94 11

SD-MVS		97.35 6	97.73 6	96.90 14	97.35 44	98.66 12	97.85 24	96.25 10	96.86 4	94.54 8	96.75 10	99.13 4	96.99 7	96.94 22	96.58 21	98.39 38	99.20 3
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
CNVR-MVS		97.30 8	97.41 9	97.18 7	99.02 10	98.60 19	98.15 15	96.24 12	96.12 16	94.10 11	95.54 24	97.99 10	96.99 7	97.97 1	97.17 8	98.57 18	98.50 27

MCST-MVS		96.83 17	97.06 15	96.57 19	98.88 20	98.47 30	98.02 20	96.16 13	95.58 23	90.96 33	95.78 22	97.84 12	96.46 21	97.00 21	96.17 35	98.94 5	98.55 24

ACMMP_NAP		96.93 15	97.27 13	96.53 23	99.06 5	98.95 7	98.24 12	96.06 14	95.66 21	90.96 33	95.63 23	97.71 14	96.53 19	97.66 8	96.68 18	98.30 48	98.61 18

zzz-MVS		96.98 14	96.68 22	97.33 4	99.09 3	98.71 11	98.43 7	96.01 15	96.11 17	95.19 3	92.89 32	97.32 21	96.84 11	97.20 15	96.09 39	98.44 31	98.46 31

SR-MVS					98.93 18			96.00 16				97.75 13

SteuartSystems-ACMMP		97.10 13	97.49 8	96.65 18	98.97 13	98.95 7	98.43 7	95.96 17	95.12 28	91.46 28	96.85 8	97.60 16	96.37 23	97.76 4	97.16 9	98.68 12	98.97 9
Skip Steuart: Steuart Systems R&D Blog.
DeepC-MVS	92.10 3	95.22 34	94.77 38	95.75 30	97.77 37	98.54 23	97.63 27	95.96 17	95.07 31	88.85 46	85.35 72	91.85 53	95.82 28	96.88 24	97.10 10	98.44 31	98.63 15
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
AdaColmap		95.02 37	93.71 48	96.54 22	98.51 26	97.76 53	96.69 39	95.94 19	93.72 45	93.50 16	89.01 51	90.53 64	96.49 20	94.51 78	93.76 77	98.07 72	96.69 94

DeepC-MVS_fast	93.32 1	96.48 21	96.42 26	96.56 20	98.70 25	98.31 36	97.97 21	95.76 20	96.31 12	92.01 27	91.43 39	95.42 39	96.46 21	97.65 9	97.69 1	98.49 24	98.12 45
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
HFP-MVS		97.11 12	97.19 14	97.00 12	98.97 13	98.73 10	98.37 10	95.69 21	96.60 7	93.28 20	96.87 7	96.64 27	97.27 2	96.64 30	96.33 31	98.44 31	98.56 19

LS3D		91.97 65	90.98 83	93.12 58	97.03 51	97.09 72	95.33 57	95.59 22	92.47 55	79.26 110	81.60 101	82.77 96	94.39 43	94.28 80	94.23 66	97.14 122	94.45 144

MP-MVS		96.56 20	96.72 21	96.37 24	98.93 18	98.48 28	98.04 19	95.55 23	94.32 40	90.95 35	95.88 21	97.02 24	96.29 24	96.77 27	96.01 41	98.47 25	98.56 19
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
DPM-MVS		95.07 35	94.84 37	95.34 34	97.44 43	97.49 61	97.76 25	95.52 24	94.88 34	88.92 45	87.25 57	96.44 29	94.41 41	95.78 50	96.11 38	97.99 80	95.95 118

ACMMPR		96.92 16	96.96 17	96.87 15	98.99 12	98.78 9	98.38 9	95.52 24	96.57 8	92.81 24	96.06 19	95.90 35	97.07 5	96.60 32	96.34 30	98.46 27	98.42 32

CP-MVS		96.68 19	96.59 25	96.77 17	98.85 21	98.58 20	98.18 14	95.51 26	95.34 25	92.94 23	95.21 27	96.25 30	96.79 14	96.44 37	95.77 44	98.35 40	98.56 19

TSAR-MVS + ACMM		96.19 22	97.39 11	94.78 37	97.70 39	98.41 33	97.72 26	95.49 27	96.47 9	86.66 64	96.35 14	97.85 11	93.99 49	97.19 17	96.37 26	97.12 123	99.13 5

TSAR-MVS + MP.		97.31 7	97.64 7	96.92 13	97.28 46	98.56 21	98.61 5	95.48 28	96.72 6	94.03 13	96.73 11	98.29 7	97.15 4	97.61 10	96.42 24	98.96 4	99.13 5
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
PCF-MVS	90.19 8	92.98 54	92.07 69	94.04 43	96.39 57	97.87 48	96.03 47	95.47 29	87.16 110	85.09 85	84.81 76	93.21 47	93.46 59	91.98 124	91.98 120	97.78 91	97.51 69
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
X-MVS		96.07 25	96.33 27	95.77 29	98.94 17	98.66 12	97.94 22	95.41 30	95.12 28	88.03 50	93.00 31	96.06 31	95.85 27	96.65 29	96.35 27	98.47 25	98.48 28

train_agg		96.15 24	96.64 24	95.58 33	98.44 27	98.03 44	98.14 17	95.40 31	93.90 44	87.72 54	96.26 16	98.10 8	95.75 29	96.25 42	95.45 49	98.01 78	98.47 29

CPTT-MVS		95.54 31	95.07 35	96.10 25	97.88 35	97.98 47	97.92 23	94.86 32	94.56 39	92.16 26	91.01 41	95.71 36	96.97 9	94.56 76	93.50 84	96.81 146	98.14 43

ACMMP		95.54 31	95.49 33	95.61 32	98.27 30	98.53 24	97.16 34	94.86 32	94.88 34	89.34 41	95.36 26	91.74 54	95.50 32	95.51 54	94.16 68	98.50 22	98.22 38
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
MSLP-MVS++		96.05 26	95.63 30	96.55 21	98.33 29	98.17 39	96.94 36	94.61 34	94.70 38	94.37 10	89.20 50	95.96 34	96.81 12	95.57 53	97.33 5	98.24 56	98.47 29

PLC	90.69 4	94.32 44	92.99 55	95.87 28	97.91 33	96.49 85	95.95 50	94.12 35	94.94 32	94.09 12	85.90 68	90.77 61	95.58 31	94.52 77	93.32 90	97.55 106	95.00 138
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
DeepPCF-MVS	92.65 2	95.50 33	96.96 17	93.79 50	96.44 56	98.21 37	93.51 91	94.08 36	96.94 2	89.29 42	93.08 30	96.77 26	93.82 53	97.68 7	97.40 4	95.59 169	98.65 14

CSCG		95.68 29	95.46 34	95.93 27	98.71 24	99.07 5	97.13 35	93.55 37	95.48 24	93.35 19	90.61 44	93.82 45	95.16 34	94.60 75	95.57 47	97.70 98	99.08 8

CNLPA		93.69 50	92.50 61	95.06 36	97.11 49	97.36 63	93.88 81	93.30 38	95.64 22	93.44 18	80.32 106	90.73 62	94.99 36	93.58 94	93.33 88	97.67 101	96.57 99

CDPH-MVS		94.80 41	95.50 32	93.98 46	98.34 28	98.06 42	97.41 30	93.23 39	92.81 51	82.98 91	92.51 33	94.82 41	93.53 57	96.08 45	96.30 32	98.42 34	97.94 52

PHI-MVS		95.86 27	96.93 20	94.61 41	97.60 41	98.65 16	96.49 40	93.13 40	94.07 42	87.91 53	97.12 6	97.17 23	93.90 52	96.46 35	96.93 15	98.64 14	98.10 47

MSDG		90.42 87	88.25 106	92.94 61	96.67 55	94.41 111	93.96 76	92.91 41	89.59 91	86.26 66	76.74 123	80.92 110	90.43 89	92.60 112	92.08 117	97.44 111	91.41 170

PGM-MVS		96.16 23	96.33 27	95.95 26	99.04 7	98.63 17	98.32 11	92.76 42	93.42 47	90.49 38	96.30 15	95.31 40	96.71 17	96.46 35	96.02 40	98.38 39	98.19 40

OPM-MVS		91.08 77	89.34 96	93.11 59	96.18 59	96.13 94	96.39 42	92.39 43	82.97 148	81.74 94	82.55 95	80.20 113	93.97 51	94.62 73	93.23 91	98.00 79	95.73 122

OMC-MVS		94.49 43	94.36 43	94.64 40	97.17 48	97.73 55	95.49 55	92.25 44	96.18 13	90.34 39	88.51 52	92.88 49	94.90 37	94.92 63	94.17 67	97.69 99	96.15 113

EPNet		93.92 47	94.40 42	93.36 53	97.89 34	96.55 83	96.08 46	92.14 45	91.65 62	89.16 43	94.07 29	90.17 68	87.78 116	95.24 57	94.97 57	97.09 125	98.15 42
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
HQP-MVS		92.39 61	92.49 62	92.29 69	95.65 64	95.94 96	95.64 54	92.12 46	92.46 56	79.65 108	91.97 36	82.68 97	92.92 65	93.47 99	92.77 103	97.74 94	98.12 45

TSAR-MVS + COLMAP		92.39 61	92.31 66	92.47 65	95.35 72	96.46 87	96.13 45	92.04 47	95.33 26	80.11 106	94.95 28	77.35 131	94.05 48	94.49 79	93.08 96	97.15 120	94.53 142

CANet		94.85 38	94.92 36	94.78 37	97.25 47	98.52 25	97.20 32	91.81 48	93.25 48	91.06 32	86.29 64	94.46 43	92.99 63	97.02 20	96.68 18	98.34 42	98.20 39

abl_6				94.78 37	97.46 42	97.99 46	95.76 51	91.80 49	93.72 45	91.25 30	91.33 40	96.47 28	94.28 46			98.14 65	97.39 73

TSAR-MVS + GP.		95.86 27	96.95 19	94.60 42	94.07 82	98.11 41	96.30 43	91.76 50	95.67 20	91.07 31	96.82 9	97.69 15	95.71 30	95.96 47	95.75 45	98.68 12	98.63 15

TAPA-MVS	90.35 6	93.69 50	93.52 49	93.90 47	96.89 52	97.62 58	96.15 44	91.67 51	94.94 32	85.97 68	87.72 56	91.96 52	94.40 42	93.76 92	93.06 98	98.30 48	95.58 126
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
3Dnovator+	90.56 5	95.06 36	94.56 41	95.65 31	98.11 31	98.15 40	97.19 33	91.59 52	95.11 30	93.23 22	81.99 98	94.71 42	95.43 33	96.48 34	96.88 16	98.35 40	98.63 15

ACMM	88.76 10	91.70 72	90.43 86	93.19 56	95.56 65	95.14 102	93.35 94	91.48 53	92.26 57	87.12 59	84.02 80	79.34 116	93.99 49	94.07 86	92.68 104	97.62 105	95.50 127
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
MVS_0304		94.30 45	94.68 39	93.86 49	96.33 58	98.48 28	97.41 30	91.20 54	92.75 52	86.96 61	86.03 67	93.81 46	92.64 67	96.89 23	96.54 23	98.61 16	98.24 37

MVS_111021_HR		94.84 39	95.91 29	93.60 51	97.35 44	98.46 31	95.08 58	91.19 55	94.18 41	85.97 68	95.38 25	92.56 50	93.61 56	96.61 31	96.25 33	98.40 36	97.92 54

MVS_111021_LR		94.84 39	95.57 31	94.00 44	97.11 49	97.72 57	94.88 61	91.16 56	95.24 27	88.74 47	96.03 20	91.52 57	94.33 45	95.96 47	95.01 56	97.79 90	97.49 70

DELS-MVS		93.71 49	93.47 50	94.00 44	96.82 53	98.39 34	96.80 38	91.07 57	89.51 92	89.94 40	83.80 82	89.29 69	90.95 83	97.32 11	97.65 2	98.42 34	98.32 35
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
3Dnovator	90.28 7	94.70 42	94.34 45	95.11 35	98.06 32	98.21 37	96.89 37	91.03 58	94.72 37	91.45 29	82.87 89	93.10 48	94.61 38	96.24 43	97.08 11	98.63 15	98.16 41

QAPM		94.13 46	94.33 46	93.90 47	97.82 36	98.37 35	96.47 41	90.89 59	92.73 54	85.63 75	85.35 72	93.87 44	94.17 47	95.71 52	95.90 42	98.40 36	98.42 32

PVSNet_BlendedMVS		92.80 55	92.44 63	93.23 54	96.02 60	97.83 51	93.74 85	90.58 60	91.86 59	90.69 36	85.87 70	82.04 103	90.01 90	96.39 38	95.26 52	98.34 42	97.81 59

PVSNet_Blended		92.80 55	92.44 63	93.23 54	96.02 60	97.83 51	93.74 85	90.58 60	91.86 59	90.69 36	85.87 70	82.04 103	90.01 90	96.39 38	95.26 52	98.34 42	97.81 59

OpenMVS	88.18 11	92.51 59	91.61 76	93.55 52	97.74 38	98.02 45	95.66 53	90.46 62	89.14 95	86.50 65	75.80 130	90.38 67	92.69 66	94.99 60	95.30 51	98.27 52	97.63 64

EPNet_dtu		88.32 109	90.61 85	85.64 138	96.79 54	92.27 166	92.03 113	90.31 63	89.05 96	65.44 181	89.43 48	85.90 80	74.22 189	92.76 107	92.09 116	95.02 179	92.76 164
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
COLMAP_ROB	84.39 15	87.61 113	86.03 132	89.46 98	95.54 67	94.48 108	91.77 117	90.14 64	87.16 110	75.50 122	73.41 142	76.86 135	87.33 122	90.05 156	89.76 166	96.48 150	90.46 179
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
ACMP	89.13 9	92.03 64	91.70 75	92.41 67	94.92 73	96.44 89	93.95 77	89.96 65	91.81 61	85.48 80	90.97 42	79.12 117	92.42 69	93.28 104	92.55 107	97.76 92	97.74 62
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
LGP-MVS_train		91.83 68	92.04 70	91.58 75	95.46 68	96.18 93	95.97 49	89.85 66	90.45 75	77.76 113	91.92 37	80.07 114	92.34 71	94.27 81	93.47 85	98.11 69	97.90 57

thres400		89.40 100	87.58 119	91.53 77	94.06 83	97.21 68	94.19 74	89.83 67	85.69 123	81.08 100	75.50 132	69.76 155	91.80 74	94.79 70	93.51 81	98.20 60	96.60 97

tfpn200view9		89.55 98	87.86 112	91.53 77	93.90 89	97.26 65	94.31 70	89.74 68	85.87 121	81.15 98	76.46 125	70.38 151	91.76 76	94.92 63	93.51 81	98.28 51	96.61 96

thres600view7		89.28 103	87.47 122	91.39 80	94.12 80	97.25 66	93.94 79	89.74 68	85.62 126	80.63 104	75.24 134	69.33 156	91.66 78	94.92 63	93.23 91	98.27 52	96.72 93

thres200		89.49 99	87.72 114	91.55 76	93.95 86	97.25 66	94.34 68	89.74 68	85.66 124	81.18 97	76.12 129	70.19 154	91.80 74	94.92 63	93.51 81	98.27 52	96.40 103

baseline1		90.81 81	90.29 87	91.42 79	93.67 95	95.86 97	93.94 79	89.69 71	89.29 94	82.85 92	82.91 88	80.30 112	89.60 93	95.05 59	94.79 60	98.80 7	93.82 153

thres100view900		89.36 101	87.61 117	91.39 80	93.90 89	96.86 79	94.35 67	89.66 72	85.87 121	81.15 98	76.46 125	70.38 151	91.17 80	94.09 85	93.43 87	98.13 66	96.16 112

PVSNet_Blended_VisFu		91.92 66	92.39 65	91.36 83	95.45 70	97.85 50	92.25 106	89.54 73	88.53 102	87.47 57	79.82 108	90.53 64	85.47 141	96.31 41	95.16 55	97.99 80	98.56 19

casdiffmvs		91.72 71	91.16 81	92.38 68	93.16 100	97.15 69	93.95 77	89.49 74	91.58 65	86.03 67	80.75 105	80.95 109	93.16 61	95.25 56	95.22 54	98.50 22	97.23 79

Anonymous202405211			88.00 109		93.16 100	96.38 90	93.58 89	89.34 75	87.92 106		65.04 179	83.03 93	92.07 72	92.67 109	93.33 88	96.96 133	97.63 64

Anonymous20231211		89.82 94	88.18 107	91.74 73	92.52 112	96.09 95	93.38 93	89.30 76	88.95 97	85.90 71	64.55 183	84.39 86	92.41 70	92.24 119	93.06 98	96.93 138	97.95 51

UGNet		91.52 73	93.41 51	89.32 100	94.13 79	97.15 69	91.83 116	89.01 77	90.62 72	85.86 72	86.83 58	91.73 55	77.40 181	94.68 72	94.43 63	97.71 96	98.40 34
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
FC-MVSNet-train		90.55 85	90.19 89	90.97 86	93.78 93	95.16 101	92.11 111	88.85 78	87.64 107	83.38 90	84.36 79	78.41 122	89.53 94	94.69 71	93.15 95	98.15 63	97.92 54

diffmvs		91.37 74	91.09 82	91.70 74	92.71 110	96.47 86	94.03 75	88.78 79	92.74 53	85.43 82	83.63 84	80.37 111	91.76 76	93.39 101	93.78 76	97.50 108	97.23 79

CLD-MVS		92.50 60	91.96 71	93.13 57	93.93 88	96.24 91	95.69 52	88.77 80	92.92 49	89.01 44	88.19 55	81.74 106	93.13 62	93.63 93	93.08 96	98.23 57	97.91 56
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
ETV-MVS		93.80 48	94.57 40	92.91 63	93.98 84	97.50 60	93.62 88	88.70 81	91.95 58	87.57 55	90.21 46	90.79 60	94.56 39	97.20 15	96.35 27	99.02 1	97.98 49

PatchMatch-RL		90.30 88	88.93 100	91.89 71	95.41 71	95.68 98	90.94 119	88.67 82	89.80 88	86.95 62	85.90 68	72.51 142	92.46 68	93.56 96	92.18 113	96.93 138	92.89 163

EIA-MVS		92.72 57	92.96 56	92.44 66	93.86 91	97.76 53	93.13 96	88.65 83	89.78 89	86.68 63	86.69 61	87.57 70	93.74 54	96.07 46	95.32 50	98.58 17	97.53 68

UniMVSNet_NR-MVSNet		86.80 119	85.86 137	87.89 117	88.17 157	94.07 117	90.15 131	88.51 84	84.20 138	73.45 130	72.38 147	70.30 153	88.95 107	90.25 150	92.21 112	98.12 67	97.62 66

DI_MVS_plusplus_trai		91.05 78	90.15 90	92.11 70	92.67 111	96.61 81	96.03 47	88.44 85	90.25 78	85.92 70	73.73 137	84.89 85	91.92 73	94.17 84	94.07 72	97.68 100	97.31 77

TranMVSNet+NR-MVSNet		85.57 135	84.41 148	86.92 125	87.67 167	93.34 135	90.31 127	88.43 86	83.07 147	70.11 151	69.99 157	65.28 175	86.96 125	89.73 159	92.27 110	98.06 74	97.17 83

UA-Net		90.81 81	92.58 60	88.74 106	94.87 74	97.44 62	92.61 100	88.22 87	82.35 151	78.93 111	85.20 74	95.61 37	79.56 176	96.52 33	96.57 22	98.23 57	94.37 145

IB-MVS	85.10 14	87.98 110	87.97 111	87.99 114	94.55 75	96.86 79	84.52 185	88.21 88	86.48 119	88.54 49	74.41 136	77.74 128	74.10 191	89.65 162	92.85 102	98.06 74	97.80 61
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
IS_MVSNet		91.87 67	93.35 52	90.14 95	94.09 81	97.73 55	93.09 97	88.12 89	88.71 99	79.98 107	84.49 77	90.63 63	87.49 120	97.07 18	96.96 14	98.07 72	97.88 58

CS-MVS		93.68 52	94.33 46	92.93 62	94.15 78	98.04 43	94.43 63	87.99 90	91.64 63	87.54 56	88.22 54	92.09 51	94.56 39	96.77 27	95.85 43	98.88 6	97.71 63

ACMH	85.51 13	87.31 116	86.59 126	88.14 112	93.96 85	94.51 107	89.00 155	87.99 90	81.58 153	70.15 150	78.41 114	71.78 147	90.60 87	91.30 133	91.99 119	97.17 119	96.58 98
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
MAR-MVS		92.71 58	92.63 59	92.79 64	97.70 39	97.15 69	93.75 84	87.98 92	90.71 69	85.76 73	86.28 65	86.38 75	94.35 44	94.95 61	95.49 48	97.22 116	97.44 71
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
Vis-MVSNet (Re-imp)		90.54 86	92.76 58	87.94 115	93.73 94	96.94 77	92.17 109	87.91 93	88.77 98	76.12 121	83.68 83	90.80 59	79.49 177	96.34 40	96.35 27	98.21 59	96.46 101

MVSTER		91.73 70	91.61 76	91.86 72	93.18 99	94.56 105	94.37 66	87.90 94	90.16 83	88.69 48	89.23 49	81.28 108	88.92 109	95.75 51	93.95 74	98.12 67	96.37 104

EPP-MVSNet		92.13 63	93.06 54	91.05 85	93.66 96	97.30 64	92.18 107	87.90 94	90.24 79	83.63 88	86.14 66	90.52 66	90.76 85	94.82 68	94.38 64	98.18 62	97.98 49

DU-MVS		86.12 127	84.81 145	87.66 118	87.77 164	93.78 122	90.15 131	87.87 96	84.40 132	73.45 130	70.59 151	64.82 180	88.95 107	90.14 151	92.33 109	97.76 92	97.62 66

NR-MVSNet		85.46 137	84.54 147	86.52 131	88.33 156	93.78 122	90.45 124	87.87 96	84.40 132	71.61 138	70.59 151	62.09 189	82.79 160	91.75 126	91.75 124	98.10 70	97.44 71

canonicalmvs		93.08 53	93.09 53	93.07 60	94.24 77	97.86 49	95.45 56	87.86 98	94.00 43	87.47 57	88.32 53	82.37 101	95.13 35	93.96 91	96.41 25	98.27 52	98.73 12

TDRefinement		84.97 143	83.39 158	86.81 127	92.97 104	94.12 115	92.18 107	87.77 99	82.78 149	71.31 142	68.43 160	68.07 162	81.10 172	89.70 161	89.03 173	95.55 171	91.62 168

Baseline_NR-MVSNet		85.28 139	83.42 157	87.46 122	87.77 164	90.80 189	89.90 141	87.69 100	83.93 142	74.16 126	64.72 181	66.43 170	87.48 121	90.14 151	90.83 135	97.73 95	97.11 84

RPSCF		89.68 96	89.24 97	90.20 92	92.97 104	92.93 150	92.30 104	87.69 100	90.44 76	85.12 84	91.68 38	85.84 81	90.69 86	87.34 179	86.07 181	92.46 190	90.37 180

ACMH+	85.75 12	87.19 117	86.02 133	88.56 107	93.42 97	94.41 111	89.91 139	87.66 102	83.45 145	72.25 137	76.42 127	71.99 146	90.78 84	89.86 157	90.94 134	97.32 112	95.11 137

thisisatest0530		91.04 79	91.74 73	90.21 91	92.93 106	97.00 74	92.06 112	87.63 103	90.74 68	81.51 95	86.81 59	82.48 98	89.23 101	94.81 69	93.03 100	97.90 85	97.33 76

tttt0517		91.01 80	91.71 74	90.19 93	92.98 102	97.07 73	91.96 115	87.63 103	90.61 73	81.42 96	86.76 60	82.26 102	89.23 101	94.86 67	93.03 100	97.90 85	97.36 74

tfpnnormal		83.80 160	81.26 182	86.77 128	89.60 142	93.26 140	89.72 144	87.60 105	72.78 192	70.44 148	60.53 192	61.15 193	85.55 139	92.72 108	91.44 129	97.71 96	96.92 89

TransMVSNet (Re)		82.67 174	80.93 185	84.69 149	88.71 149	91.50 181	87.90 164	87.15 106	71.54 197	68.24 164	63.69 185	64.67 182	78.51 180	91.65 128	90.73 141	97.64 103	92.73 166

UniMVSNet (Re)		86.22 125	85.46 142	87.11 123	88.34 155	94.42 110	89.65 145	87.10 107	84.39 134	74.61 124	70.41 154	68.10 161	85.10 144	91.17 136	91.79 123	97.84 88	97.94 52

DCV-MVSNet		91.24 75	91.26 79	91.22 84	92.84 107	93.44 132	93.82 82	86.75 108	91.33 67	85.61 76	84.00 81	85.46 82	91.27 79	92.91 106	93.62 79	97.02 129	98.05 48

ET-MVSNet_ETH3D		89.93 92	90.84 84	88.87 104	79.60 202	96.19 92	94.43 63	86.56 109	90.63 71	80.75 103	90.71 43	77.78 127	93.73 55	91.36 132	93.45 86	98.15 63	95.77 121

UniMVSNet_ETH3D		84.57 146	81.40 180	88.28 110	89.34 145	94.38 113	90.33 125	86.50 110	74.74 190	77.52 115	59.90 193	62.04 190	88.78 112	88.82 172	92.65 105	97.22 116	97.24 78

MVS_Test		91.81 69	92.19 67	91.37 82	93.24 98	96.95 76	94.43 63	86.25 111	91.45 66	83.45 89	86.31 63	85.15 83	92.93 64	93.99 87	94.71 61	97.92 84	96.77 92

CDS-MVSNet		88.34 108	88.71 101	87.90 116	90.70 135	94.54 106	92.38 102	86.02 112	80.37 159	79.42 109	79.30 109	83.43 90	82.04 164	93.39 101	94.01 73	96.86 144	95.93 119
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
Vis-MVSNet		89.36 101	91.49 78	86.88 126	92.10 115	97.60 59	92.16 110	85.89 113	84.21 137	75.20 123	82.58 93	87.13 71	77.40 181	95.90 49	95.63 46	98.51 20	97.36 74
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
CHOSEN 1792x2688		88.57 106	87.82 113	89.44 99	95.46 68	96.89 78	93.74 85	85.87 114	89.63 90	77.42 116	61.38 189	83.31 91	88.80 111	93.44 100	93.16 94	95.37 174	96.95 88

GBi-Net		90.21 89	90.11 91	90.32 89	88.66 151	93.65 128	94.25 71	85.78 115	90.03 84	85.56 77	77.38 116	86.13 76	89.38 97	93.97 88	94.16 68	98.31 45	95.47 128

test1		90.21 89	90.11 91	90.32 89	88.66 151	93.65 128	94.25 71	85.78 115	90.03 84	85.56 77	77.38 116	86.13 76	89.38 97	93.97 88	94.16 68	98.31 45	95.47 128

FMVSNet3		90.19 91	90.06 93	90.34 88	88.69 150	93.85 120	94.58 62	85.78 115	90.03 84	85.56 77	77.38 116	86.13 76	89.22 103	93.29 103	94.36 65	98.20 60	95.40 132

v148		83.61 162	82.10 171	85.37 139	87.34 171	92.94 149	87.48 167	85.72 118	78.92 168	73.87 128	65.71 175	64.69 181	81.78 168	87.82 175	89.35 170	96.01 158	95.26 134

FMVSNet2		89.61 97	89.14 98	90.16 94	88.66 151	93.65 128	94.25 71	85.44 119	88.57 101	84.96 86	73.53 139	83.82 88	89.38 97	94.23 82	94.68 62	98.31 45	95.47 128

v2v482		84.51 148	83.05 164	86.20 133	87.25 173	93.28 138	90.22 129	85.40 120	79.94 165	69.78 153	67.74 162	65.15 177	87.57 118	89.12 168	90.55 145	96.97 131	95.60 125

pm-mvs1		84.55 147	83.46 154	85.82 134	88.16 158	93.39 134	89.05 154	85.36 121	74.03 191	72.43 136	65.08 178	71.11 148	82.30 163	93.48 98	91.70 125	97.64 103	95.43 131

baseline2		88.97 104	89.50 95	88.36 108	91.14 127	95.30 99	90.13 133	85.17 122	87.24 109	80.80 102	84.46 78	78.44 121	85.60 138	93.54 97	91.87 121	97.31 113	95.66 123

pmmvs4		86.00 130	84.28 149	88.00 113	87.80 162	92.01 173	89.94 138	84.91 123	86.79 113	80.98 101	73.41 142	66.34 171	88.12 114	89.31 165	88.90 174	96.24 156	93.20 161

FC-MVSNet-test		86.15 126	89.10 99	82.71 176	89.83 139	93.18 142	87.88 165	84.69 124	86.54 116	62.18 191	82.39 96	83.31 91	74.18 190	92.52 114	91.86 122	97.50 108	93.88 152

CANet_DTU		90.74 84	92.93 57	88.19 111	94.36 76	96.61 81	94.34 68	84.66 125	90.66 70	68.75 160	90.41 45	86.89 73	89.78 92	95.46 55	94.87 58	97.25 115	95.62 124

GA-MVS		85.08 141	85.65 139	84.42 153	89.77 140	94.25 114	89.26 149	84.62 126	81.19 156	62.25 190	75.72 131	68.44 160	84.14 152	93.57 95	91.68 127	96.49 149	94.71 141

USDC		86.73 121	85.96 135	87.63 120	91.64 119	93.97 118	92.76 99	84.58 127	88.19 103	70.67 147	80.10 107	67.86 163	89.43 95	91.81 125	89.77 165	96.69 148	90.05 183

baseline		91.19 76	91.89 72	90.38 87	92.76 108	95.04 103	93.55 90	84.54 128	92.92 49	85.71 74	86.68 62	86.96 72	89.28 100	92.00 123	92.62 106	96.46 151	96.99 86

FMVSNet1		87.33 115	86.00 134	88.89 103	87.13 177	92.83 153	93.08 98	84.46 129	81.35 155	82.20 93	66.33 170	77.96 125	88.96 106	93.97 88	94.16 68	97.54 107	95.38 133

Effi-MVS+		89.79 95	89.83 94	89.74 96	92.98 102	96.45 88	93.48 92	84.24 130	87.62 108	76.45 119	81.76 99	77.56 130	93.48 58	94.61 74	93.59 80	97.82 89	97.22 81

WR-MVS		83.14 168	83.38 159	82.87 174	87.55 168	93.29 137	86.36 178	84.21 131	80.05 163	66.41 175	66.91 166	66.92 168	75.66 187	88.96 170	90.56 144	97.05 127	96.96 87

HyFIR lowres test		87.87 111	86.42 128	89.57 97	95.56 65	96.99 75	92.37 103	84.15 132	86.64 114	77.17 117	57.65 195	83.97 87	91.08 82	92.09 122	92.44 108	97.09 125	95.16 135

thisisatest0515		85.70 132	87.00 123	84.19 156	88.16 158	93.67 127	84.20 187	84.14 133	83.39 146	72.91 132	76.79 122	74.75 140	78.82 179	92.57 113	91.26 132	96.94 135	96.56 100

TinyColmap		84.04 156	82.01 173	86.42 132	90.87 131	91.84 175	88.89 157	84.07 134	82.11 152	69.89 152	71.08 149	60.81 194	89.04 105	90.52 147	89.19 171	95.76 161	88.50 190

V42		84.48 150	83.36 160	85.79 136	87.14 176	93.28 138	90.03 134	83.98 135	80.30 160	71.20 143	66.90 167	67.17 165	85.55 139	89.35 163	90.27 150	96.82 145	96.27 110

PEN-MVS		82.49 176	81.58 177	83.56 164	86.93 180	92.05 172	86.71 175	83.84 136	76.94 180	64.68 184	67.24 163	60.11 197	81.17 171	87.78 176	90.70 142	98.02 77	96.21 111

DTE-MVSNet		81.76 181	81.04 183	82.60 178	86.63 184	91.48 183	85.97 181	83.70 137	76.45 184	62.44 189	67.16 164	59.98 198	78.98 178	87.15 180	89.93 162	97.88 87	95.12 136

test-LLR		86.88 118	88.28 104	85.24 142	91.22 125	92.07 170	87.41 168	83.62 138	84.58 130	69.33 156	83.00 86	82.79 94	84.24 149	92.26 117	89.81 163	95.64 167	93.44 156

test0.0.03 1		85.58 134	87.69 116	83.11 169	91.22 125	92.54 161	85.60 184	83.62 138	85.66 124	67.84 167	82.79 91	79.70 115	73.51 193	91.15 137	90.79 136	96.88 142	91.23 173

IterMVS-LS		88.60 105	88.45 102	88.78 105	92.02 116	92.44 164	92.00 114	83.57 140	86.52 117	78.90 112	78.61 113	81.34 107	89.12 104	90.68 145	93.18 93	97.10 124	96.35 105
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
CP-MVSNet		83.11 170	82.15 170	84.23 155	87.20 174	92.70 155	86.42 177	83.53 141	77.83 175	67.67 168	66.89 168	60.53 196	82.47 161	89.23 167	90.65 143	98.08 71	97.20 82

PS-CasMVS		82.53 175	81.54 178	83.68 162	87.08 179	92.54 161	86.20 179	83.46 142	76.46 183	65.73 180	65.71 175	59.41 201	81.61 169	89.06 169	90.55 145	98.03 76	97.07 85

v1144		84.03 157	82.88 165	85.37 139	87.17 175	93.15 145	90.18 130	83.31 143	78.83 169	67.85 166	65.99 172	64.99 178	86.79 127	90.75 142	90.33 149	96.90 140	96.15 113

CVMVSNet		83.83 159	85.53 140	81.85 183	89.60 142	90.92 185	87.81 166	83.21 144	80.11 162	60.16 193	76.47 124	78.57 120	76.79 183	89.76 158	90.13 153	93.51 182	92.75 165

WR-MVS_H		82.86 173	82.66 167	83.10 170	87.44 170	93.33 136	85.71 183	83.20 145	77.36 177	68.20 165	66.37 169	65.23 176	76.05 186	89.35 163	90.13 153	97.99 80	96.89 90

v1192		83.56 163	82.35 168	84.98 144	86.84 182	92.84 151	90.01 136	82.70 146	78.54 170	66.48 174	64.88 180	62.91 184	86.91 126	90.72 143	90.25 151	96.94 135	96.32 107

Fast-Effi-MVS+		88.56 107	87.99 110	89.22 101	91.56 122	95.21 100	92.29 105	82.69 147	86.82 112	77.73 114	76.24 128	73.39 141	93.36 60	94.22 83	93.64 78	97.65 102	96.43 102

Effi-MVS+-dtu		87.51 114	88.13 108	86.77 128	91.10 128	94.90 104	90.91 120	82.67 148	83.47 144	71.55 139	81.11 104	77.04 132	89.41 96	92.65 111	91.68 127	95.00 180	96.09 115

MDA-MVSNet-bldmvs		73.81 194	72.56 198	75.28 194	72.52 207	88.87 193	74.95 202	82.67 148	71.57 195	55.02 200	65.96 173	42.84 212	76.11 185	70.61 204	81.47 197	90.38 201	86.59 194

pmmvs6		80.90 183	78.77 188	83.38 167	85.84 188	91.61 179	86.01 180	82.54 150	64.17 204	70.43 149	54.14 202	67.06 167	80.73 173	90.50 148	89.17 172	94.74 181	94.75 140

v144192		83.48 164	82.23 169	84.94 145	86.65 183	92.84 151	89.63 146	82.48 151	77.87 174	67.36 170	65.33 177	63.50 183	86.51 129	89.72 160	89.99 161	97.03 128	96.35 105

v8		84.45 152	83.30 161	85.80 135	87.53 169	92.95 148	90.31 127	82.46 152	80.46 158	71.43 140	66.99 165	67.16 166	86.14 135	89.26 166	90.22 152	96.94 135	96.06 116

SixPastTwentyTwo		83.12 169	83.44 156	82.74 175	87.71 166	93.11 146	82.30 192	82.33 153	79.24 167	64.33 185	78.77 112	62.75 185	84.11 153	88.11 174	87.89 176	95.70 165	94.21 148

LTVRE_ROB	81.71 16	82.44 177	81.84 175	83.13 168	89.01 146	92.99 147	88.90 156	82.32 154	66.26 203	54.02 203	74.68 135	59.62 200	88.87 110	90.71 144	92.02 118	95.68 166	96.62 95
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
v1921920		83.30 166	82.09 172	84.70 148	86.59 185	92.67 157	89.82 142	82.23 155	78.32 171	65.76 179	64.64 182	62.35 187	86.78 128	90.34 149	90.02 159	97.02 129	96.31 108

pmmvs5		83.37 165	82.68 166	84.18 157	87.13 177	93.18 142	86.74 174	82.08 156	76.48 182	67.28 171	71.26 148	62.70 186	84.71 146	90.77 141	90.12 156	97.15 120	94.24 146

PMMVS		89.88 93	91.19 80	88.35 109	89.73 141	91.97 174	90.62 122	81.92 157	90.57 74	80.58 105	92.16 34	86.85 74	91.17 80	92.31 116	91.35 131	96.11 157	93.11 162

N_pmnet		77.55 192	76.68 195	78.56 190	85.43 192	87.30 199	78.84 197	81.88 158	78.30 172	60.61 192	61.46 188	62.15 188	74.03 192	82.04 197	80.69 199	90.59 200	84.81 200

v1240		82.88 172	81.66 176	84.29 154	86.46 186	92.52 163	89.06 153	81.82 159	77.16 178	65.09 183	64.17 184	61.50 191	86.36 130	90.12 153	90.13 153	96.95 134	96.04 117

testgi		81.94 179	84.09 150	79.43 188	89.53 144	90.83 188	82.49 191	81.75 160	80.59 157	59.46 195	82.82 90	65.75 172	67.97 195	90.10 154	89.52 168	95.39 173	89.03 186

CMPMVS	61.19 17	79.86 186	77.46 194	82.66 177	91.54 123	91.82 176	83.25 188	81.57 161	70.51 199	68.64 161	59.89 194	66.77 169	79.63 175	84.00 194	84.30 190	91.34 195	84.89 199
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
MS-PatchMatch		87.63 112	87.61 117	87.65 119	93.95 86	94.09 116	92.60 101	81.52 162	86.64 114	76.41 120	73.46 141	85.94 79	85.01 145	92.23 120	90.00 160	96.43 153	90.93 176

v10		84.18 153	83.17 163	85.37 139	87.34 171	92.68 156	90.32 126	81.33 163	79.93 166	69.23 158	66.33 170	65.74 173	87.03 124	90.84 140	90.38 147	96.97 131	96.29 109

pmmvs-eth3d		79.78 187	77.58 192	82.34 180	81.57 200	87.46 198	82.92 189	81.28 164	75.33 189	71.34 141	61.88 187	52.41 205	81.59 170	87.56 177	86.90 179	95.36 175	91.48 169

TAMVS		84.94 144	84.95 143	84.93 146	88.82 147	93.18 142	88.44 161	81.28 164	77.16 178	73.76 129	75.43 133	76.57 136	82.04 164	90.59 146	90.79 136	95.22 176	90.94 175

Anonymous20231206		78.09 190	78.11 191	78.07 191	85.19 193	89.17 192	80.99 194	81.24 166	75.46 188	58.25 197	54.78 201	59.90 199	66.73 198	88.94 171	88.26 175	96.01 158	90.25 181

v7n		82.25 178	81.54 178	83.07 171	85.55 191	92.58 159	86.68 176	81.10 167	76.54 181	65.97 178	62.91 186	60.56 195	82.36 162	91.07 138	90.35 148	96.77 147	96.80 91

test20.03		76.41 193	78.49 190	73.98 195	85.64 190	87.50 197	75.89 200	80.71 168	70.84 198	51.07 206	68.06 161	61.40 192	54.99 204	88.28 173	87.20 178	95.58 170	86.15 195

anonymousdsp		84.51 148	85.85 138	82.95 173	86.30 187	93.51 131	85.77 182	80.38 169	78.25 173	63.42 188	73.51 140	72.20 144	84.64 147	93.21 105	92.16 114	97.19 118	98.14 43

CHOSEN 280x420		90.77 83	92.14 68	89.17 102	93.86 91	92.81 154	93.16 95	80.22 170	90.21 80	84.67 87	89.89 47	91.38 58	90.57 88	94.94 62	92.11 115	92.52 189	93.65 155

MIMVSNet1		73.19 195	73.70 196	72.60 198	65.42 210	86.69 201	75.56 201	79.65 171	67.87 202	55.30 199	45.24 205	56.41 203	63.79 200	86.98 181	87.66 177	95.85 160	85.04 198

EU-MVSNet		78.43 188	80.25 186	76.30 193	83.81 196	87.27 200	80.99 194	79.52 172	76.01 185	54.12 202	70.44 153	64.87 179	67.40 197	86.23 185	85.54 185	91.95 194	91.41 170

FMVSNet5		84.47 151	84.72 146	84.18 157	83.30 197	88.43 194	88.09 163	79.42 173	84.25 136	74.14 127	73.15 144	78.74 118	83.65 155	91.19 135	91.19 133	96.46 151	86.07 196

PM-MVS		80.29 185	79.30 187	81.45 185	81.91 199	88.23 195	82.61 190	79.01 174	79.99 164	67.15 172	69.07 158	51.39 206	82.92 159	87.55 178	85.59 183	95.08 177	93.28 159

EG-PatchMatch MVS		81.70 182	81.31 181	82.15 181	88.75 148	93.81 121	87.14 171	78.89 175	71.57 195	64.12 187	61.20 191	68.46 159	76.73 184	91.48 129	90.77 138	97.28 114	91.90 167

IterMVS-SCA-FT		85.44 138	86.71 124	83.97 160	90.59 136	90.84 187	89.73 143	78.34 176	84.07 141	66.40 176	77.27 121	78.66 119	83.06 157	91.20 134	90.10 158	95.72 164	94.78 139

Fast-Effi-MVS+-dtu		86.25 123	87.70 115	84.56 151	90.37 138	93.70 125	90.54 123	78.14 177	83.50 143	65.37 182	81.59 102	75.83 139	86.09 137	91.70 127	91.70 125	96.88 142	95.84 120

IterMVS		85.25 140	86.49 127	83.80 161	90.42 137	90.77 190	90.02 135	78.04 178	84.10 139	66.27 177	77.28 120	78.41 122	83.01 158	90.88 139	89.72 167	95.04 178	94.24 146
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
MDTV_nov1_ep13		86.64 122	87.50 121	85.65 137	90.73 133	93.69 126	89.96 137	78.03 179	89.48 93	76.85 118	84.92 75	82.42 100	86.14 135	86.85 183	86.15 180	92.17 191	88.97 188

new-patchmatchnet		72.32 196	71.09 199	73.74 196	81.17 201	84.86 203	72.21 205	77.48 180	68.32 201	54.89 201	55.10 199	49.31 209	63.68 201	79.30 200	76.46 202	93.03 186	84.32 201

TESTMET0.1,1		86.11 128	88.28 104	83.59 163	87.80 162	92.07 170	87.41 168	77.12 181	84.58 130	69.33 156	83.00 86	82.79 94	84.24 149	92.26 117	89.81 163	95.64 167	93.44 156

test-mter		86.09 129	88.38 103	83.43 166	87.89 161	92.61 158	86.89 173	77.11 182	84.30 135	68.62 162	82.57 94	82.45 99	84.34 148	92.40 115	90.11 157	95.74 162	94.21 148

FPMVS		69.87 199	67.10 202	73.10 197	84.09 195	78.35 207	79.40 196	76.41 183	71.92 193	57.71 198	54.06 203	50.04 207	56.72 202	71.19 203	68.70 204	84.25 205	75.43 205

CostFormer		86.78 120	86.05 131	87.62 121	92.15 114	93.20 141	91.55 118	75.83 184	88.11 105	85.29 83	81.76 99	76.22 137	87.80 115	84.45 191	85.21 187	93.12 185	93.42 158

tpm cat1		84.13 154	81.99 174	86.63 130	91.74 118	91.50 181	90.68 121	75.69 185	86.12 120	85.44 81	72.39 146	70.72 149	85.16 143	80.89 199	81.56 196	91.07 197	90.71 177

dps		85.00 142	83.21 162	87.08 124	90.73 133	92.55 160	89.34 147	75.29 186	84.94 127	87.01 60	79.27 110	67.69 164	87.27 123	84.22 192	83.56 192	92.83 187	90.25 181

gm-plane-assit		77.65 191	78.50 189	76.66 192	87.96 160	85.43 202	64.70 208	74.50 187	64.15 205	51.26 205	61.32 190	58.17 202	84.11 153	95.16 58	93.83 75	97.45 110	91.41 170

PMVS	56.77 18	61.27 201	58.64 203	64.35 201	75.66 203	54.60 211	53.62 210	74.23 188	53.69 208	58.37 196	44.27 206	49.38 208	44.16 207	69.51 205	65.35 206	80.07 207	73.66 206
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
tpm		83.16 167	83.64 152	82.60 178	90.75 132	91.05 184	88.49 160	73.99 189	82.36 150	67.08 173	78.10 115	68.79 157	84.17 151	85.95 187	85.96 182	91.09 196	93.23 160

MDTV_nov1_ep13_2view		80.43 184	80.94 184	79.84 186	84.82 194	90.87 186	84.23 186	73.80 190	80.28 161	64.33 185	70.05 156	68.77 158	79.67 174	84.83 190	83.50 193	92.17 191	88.25 193

PatchmatchNet		85.70 132	86.65 125	84.60 150	91.79 117	93.40 133	89.27 148	73.62 191	90.19 81	72.63 135	82.74 92	81.93 105	87.64 117	84.99 189	84.29 191	92.64 188	89.00 187
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
EPMVS		85.77 131	86.24 130	85.23 143	92.76 108	93.78 122	89.91 139	73.60 192	90.19 81	74.22 125	82.18 97	78.06 124	87.55 119	85.61 188	85.38 186	93.32 184	88.48 191

tpmrst		83.72 161	83.45 155	84.03 159	92.21 113	91.66 178	88.74 158	73.58 193	88.14 104	72.67 134	77.37 119	72.11 145	86.34 131	82.94 196	82.05 195	90.63 199	89.86 184

CR-MVSNet		85.48 136	86.29 129	84.53 152	91.08 130	92.10 168	89.18 150	73.30 194	84.75 128	71.08 144	73.12 145	77.91 126	86.27 133	91.48 129	90.75 139	96.27 155	93.94 150

Patchmtry						92.39 165	89.18 150	73.30 194		71.08 144

SCA		86.25 123	87.52 120	84.77 147	91.59 120	93.90 119	89.11 152	73.25 196	90.38 77	72.84 133	83.26 85	83.79 89	88.49 113	86.07 186	85.56 184	93.33 183	89.67 185

MIMVSNet		82.97 171	84.00 151	81.77 184	82.23 198	92.25 167	87.40 170	72.73 197	81.48 154	69.55 154	68.79 159	72.42 143	81.82 167	92.23 120	92.25 111	96.89 141	88.61 189

PatchT		83.86 158	85.51 141	81.94 182	88.41 154	91.56 180	78.79 198	71.57 198	84.08 140	71.08 144	70.62 150	76.13 138	86.27 133	91.48 129	90.75 139	95.52 172	93.94 150

RPMNet		84.82 145	85.90 136	83.56 164	91.10 128	92.10 168	88.73 159	71.11 199	84.75 128	68.79 159	73.56 138	77.62 129	85.33 142	90.08 155	89.43 169	96.32 154	93.77 154

ADS-MVSNet		84.08 155	84.95 143	83.05 172	91.53 124	91.75 177	88.16 162	70.70 200	89.96 87	69.51 155	78.83 111	76.97 134	86.29 132	84.08 193	84.60 189	92.13 193	88.48 191

MVS-HIRNet		78.16 189	77.57 193	78.83 189	85.83 189	87.76 196	76.67 199	70.22 201	75.82 187	67.39 169	55.61 197	70.52 150	81.96 166	86.67 184	85.06 188	90.93 198	81.58 202

new_pmnet		72.29 197	73.25 197	71.16 200	75.35 204	81.38 204	73.72 204	69.27 202	75.97 186	49.84 207	56.27 196	56.12 204	69.08 194	81.73 198	80.86 198	89.72 203	80.44 203

gg-mvs-nofinetune		81.83 180	83.58 153	79.80 187	91.57 121	96.54 84	93.79 83	68.80 203	62.71 206	43.01 210	55.28 198	85.06 84	83.65 155	96.13 44	94.86 59	97.98 83	94.46 143

Gipuma		58.52 202	56.17 204	61.27 202	67.14 209	58.06 210	52.16 211	68.40 204	69.00 200	45.02 209	22.79 208	20.57 215	55.11 203	76.27 201	79.33 201	79.80 208	67.16 208
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
pmmvs3		71.13 198	71.06 200	71.21 199	73.54 206	80.19 205	71.69 206	64.86 205	62.04 207	52.10 204	54.92 200	48.00 210	75.03 188	83.75 195	83.24 194	90.04 202	85.27 197

E-PMN		40.00 204	35.74 207	44.98 205	57.69 213	39.15 215	28.05 213	62.70 206	35.52 211	17.78 213	20.90 209	14.36 217	44.47 206	35.89 209	47.86 208	59.15 211	56.47 210

EMVS		39.04 206	34.32 208	44.54 206	58.25 212	39.35 214	27.61 214	62.55 207	35.99 210	16.40 214	20.04 211	14.77 216	44.80 205	33.12 210	44.10 209	57.61 212	52.89 211

PMMVS2		53.68 203	55.72 205	51.30 203	58.84 211	67.02 209	54.23 209	60.97 208	47.50 209	19.42 212	34.81 207	31.97 213	30.88 209	65.84 206	69.99 203	83.47 206	72.92 207

MVE	39.81 19	39.52 205	41.58 206	37.11 207	33.93 214	49.06 212	26.45 215	54.22 209	29.46 212	24.15 211	20.77 210	10.60 218	34.42 208	51.12 208	65.27 207	49.49 213	64.81 209
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
DeepMVS_CX						71.82 208	68.37 207	48.05 210	77.38 176	46.88 208	65.77 174	47.03 211	67.48 196	64.27 207		76.89 209	76.72 204

tmp_tt				50.24 204	68.55 208	46.86 213	48.90 212	18.28 211	86.51 118	68.32 163	70.19 155	65.33 174	26.69 210	74.37 202	66.80 205	70.72 210

testmvs		4.35 207	6.54 209	1.79 209	0.60 215	1.82 216	3.06 217	0.95 212	7.22 213	0.88 216	12.38 212	1.25 219	3.87 212	6.09 211	5.58 210	1.40 214	11.42 213

test123		3.48 208	5.31 210	1.34 210	0.20 217	1.52 217	2.17 218	0.58 213	6.13 214	0.31 217	9.85 213	0.31 220	3.90 211	2.65 212	5.28 211	0.87 215	11.46 212

GG-mvs-BLEND		62.84 200	90.21 88	30.91 208	0.57 216	94.45 109	86.99 172	0.34 214	88.71 99	0.98 215	81.55 103	91.58 56	0.86 213	92.66 110	91.43 130	95.73 163	91.11 174

uanet_test		0.00 209	0.00 211	0.00 211	0.00 218	0.00 218	0.00 219	0.00 215	0.00 215	0.00 218	0.00 214	0.00 221	0.00 214	0.00 213	0.00 212	0.00 216	0.00 214

sosnet-low-res		0.00 209	0.00 211	0.00 211	0.00 218	0.00 218	0.00 219	0.00 215	0.00 215	0.00 218	0.00 214	0.00 221	0.00 214	0.00 213	0.00 212	0.00 216	0.00 214

sosnet		0.00 209	0.00 211	0.00 211	0.00 218	0.00 218	0.00 219	0.00 215	0.00 215	0.00 218	0.00 214	0.00 221	0.00 214	0.00 213	0.00 212	0.00 216	0.00 214

9.14												97.28 22

our_test_3					86.93 180	89.77 191	81.61 193

test_part1																	98.96 10

ambc			67.96 201		73.69 205	79.79 206	73.82 203		71.61 194	59.80 194	46.00 204	20.79 214	66.15 199	86.92 182	80.11 200	89.13 204	90.50 178

MTAPA										95.36 2		97.46 19

MTMP										95.70 1		96.90 25

Patchmatch-RL test							18.47 216

XVS					95.68 62	98.66 12	94.96 59			88.03 50		96.06 31				98.46 27

X-MVStestdata					95.68 62	98.66 12	94.96 59			88.03 50		96.06 31				98.46 27

mPP-MVS					98.76 23							95.49 38

NP-MVS									91.63 64

High-resolution multi-view benchmark