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		95.53 2	96.13 2	94.82 1	96.81 21	98.05 2	97.42 1	93.09 1	94.31 7	91.49 5	97.12 1	95.03 2	93.27 3	95.55 4	94.58 10	96.86 3	98.25 2

APDe-MVS		95.23 3	95.69 4	94.70 3	97.12 10	97.81 4	97.19 2	92.83 2	95.06 4	90.98 7	96.47 2	92.77 9	93.38 2	95.34 7	94.21 14	96.68 6	98.17 3

DVP-MVS		95.56 1	96.26 1	94.73 2	96.93 16	98.19 1	96.62 5	92.81 3	96.15 1	91.73 4	95.01 4	95.31 1	93.41 1	95.95 1	94.77 6	96.90 2	98.46 1

APD-MVS		94.37 9	94.47 13	94.26 5	97.18 8	96.99 13	96.53 6	92.68 4	92.45 21	89.96 14	94.53 8	91.63 17	92.89 5	94.58 19	93.82 21	96.31 14	97.26 16
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
HPM-MVS++		94.60 6	94.91 8	94.24 6	97.86 1	96.53 29	96.14 7	92.51 5	93.87 12	90.76 9	93.45 15	93.84 3	92.62 7	95.11 10	94.08 17	95.58 47	97.48 13

NCCC		93.69 17	93.66 20	93.72 13	97.37 5	96.66 26	95.93 15	92.50 6	93.40 16	88.35 22	87.36 32	92.33 12	92.18 11	94.89 12	94.09 16	96.00 20	96.91 24

CNVR-MVS		94.37 9	94.65 9	94.04 8	97.29 6	97.11 9	96.00 9	92.43 7	93.45 13	89.85 16	90.92 22	93.04 7	92.59 8	95.77 3	94.82 4	96.11 18	97.42 15

MSP-MVS		95.12 4	95.83 3	94.30 4	96.82 20	97.94 3	96.98 3	92.37 8	95.40 2	90.59 10	96.16 3	93.71 4	92.70 6	94.80 14	94.77 6	96.37 11	97.99 7

SD-MVS		94.53 7	95.22 6	93.73 12	95.69 33	97.03 11	95.77 19	91.95 9	94.41 6	91.35 6	94.97 5	93.34 6	91.80 17	94.72 17	93.99 18	95.82 31	98.07 6
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
SMA-MVS		94.70 5	95.35 5	93.93 9	97.57 2	97.57 6	95.98 10	91.91 10	94.50 5	90.35 11	93.46 14	92.72 10	91.89 15	95.89 2	95.22 1	95.88 24	98.10 4

SteuartSystems-ACMMP		94.06 11	94.65 9	93.38 16	96.97 15	97.36 7	96.12 8	91.78 11	92.05 25	87.34 27	94.42 9	90.87 21	91.87 16	95.47 6	94.59 9	96.21 16	97.77 10
Skip Steuart: Steuart Systems R&D Blog.
DeepC-MVS	87.86 3	92.26 28	91.86 31	92.73 22	96.18 26	96.87 16	95.19 25	91.76 12	92.17 24	86.58 32	81.79 47	85.85 47	90.88 26	94.57 21	94.61 8	95.80 32	97.18 17
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
MCST-MVS		93.81 14	94.06 16	93.53 14	96.79 22	96.85 17	95.95 12	91.69 13	92.20 23	87.17 29	90.83 24	93.41 5	91.96 13	94.49 22	93.50 28	97.61 1	97.12 20

AdaColmap		90.29 40	88.38 54	92.53 23	96.10 28	95.19 50	92.98 43	91.40 14	89.08 44	88.65 20	78.35 68	81.44 66	91.30 25	90.81 79	90.21 73	94.72 85	93.59 84

ACMMP_NAP		93.94 13	94.49 12	93.30 17	97.03 13	97.31 8	95.96 11	91.30 15	93.41 15	88.55 21	93.00 16	90.33 25	91.43 23	95.53 5	94.41 12	95.53 49	97.47 14

DeepC-MVS_fast	88.76 1	93.10 20	93.02 27	93.19 19	97.13 9	96.51 30	95.35 23	91.19 16	93.14 18	88.14 23	85.26 38	89.49 32	91.45 20	95.17 8	95.07 2	95.85 29	96.48 32
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
zzz-MVS		93.80 15	93.45 23	94.20 7	97.53 3	96.43 33	95.88 16	91.12 17	94.09 10	92.74 3	87.68 30	90.77 22	92.04 12	94.74 16	93.56 27	95.91 23	96.85 25

DPM-MVS		91.72 32	91.48 32	92.00 28	95.53 34	95.75 42	95.94 13	91.07 18	91.20 31	85.58 38	81.63 51	90.74 23	88.40 42	93.40 35	93.75 23	95.45 53	93.85 78

SR-MVS					96.58 23			90.99 19				92.40 11

HFP-MVS		94.02 12	94.22 15	93.78 11	97.25 7	96.85 17	95.81 17	90.94 20	94.12 9	90.29 13	94.09 11	89.98 28	92.52 9	93.94 29	93.49 30	95.87 26	97.10 21

TSAR-MVS + MP.		94.48 8	94.97 7	93.90 10	95.53 34	97.01 12	96.69 4	90.71 21	94.24 8	90.92 8	94.97 5	92.19 13	93.03 4	94.83 13	93.60 25	96.51 10	97.97 8
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
MP-MVS		93.35 18	93.59 21	93.08 20	97.39 4	96.82 19	95.38 22	90.71 21	90.82 33	88.07 24	92.83 18	90.29 26	91.32 24	94.03 26	93.19 35	95.61 45	97.16 18
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
ACMMPR		93.72 16	93.94 17	93.48 15	97.07 11	96.93 14	95.78 18	90.66 23	93.88 11	89.24 18	93.53 13	89.08 35	92.24 10	93.89 31	93.50 28	95.88 24	96.73 29

CP-MVS		93.25 19	93.26 24	93.24 18	96.84 19	96.51 30	95.52 21	90.61 24	92.37 22	88.88 19	90.91 23	89.52 31	91.91 14	93.64 33	92.78 41	95.69 38	97.09 22

train_agg		92.87 22	93.53 22	92.09 27	96.88 18	95.38 46	95.94 13	90.59 25	90.65 35	83.65 48	94.31 10	91.87 16	90.30 28	93.38 36	92.42 43	95.17 66	96.73 29

X-MVS		92.36 27	92.75 28	91.90 30	96.89 17	96.70 22	95.25 24	90.48 26	91.50 30	83.95 45	88.20 28	88.82 37	89.11 34	93.75 32	93.43 31	95.75 36	96.83 27

TSAR-MVS + ACMM		92.97 21	94.51 11	91.16 34	95.88 31	96.59 27	95.09 26	90.45 27	93.42 14	83.01 50	94.68 7	90.74 23	88.74 37	94.75 15	93.78 22	93.82 125	97.63 11

PCF-MVS	84.60 6	88.66 51	87.75 64	89.73 45	93.06 59	96.02 37	93.22 41	90.00 28	82.44 73	80.02 69	77.96 71	85.16 50	87.36 55	88.54 106	88.54 109	94.72 85	95.61 49
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
DeepPCF-MVS	88.51 2	92.64 26	94.42 14	90.56 38	94.84 41	96.92 15	91.31 58	89.61 29	95.16 3	84.55 43	89.91 26	91.45 18	90.15 30	95.12 9	94.81 5	92.90 144	97.58 12

ACMMP		92.03 30	92.16 29	91.87 31	95.88 31	96.55 28	94.47 32	89.49 30	91.71 28	85.26 39	91.52 21	84.48 52	90.21 29	92.82 45	91.63 49	95.92 22	96.42 34
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++		92.02 31	91.40 34	92.75 21	96.01 29	95.88 41	93.73 37	89.00 31	89.89 41	90.31 12	81.28 53	88.85 36	91.45 20	92.88 44	94.24 13	96.00 20	96.76 28

LS3D		85.96 74	84.37 89	87.81 64	94.13 47	93.27 77	90.26 65	89.00 31	84.91 62	72.84 102	71.74 101	72.47 115	87.45 54	89.53 97	89.09 100	93.20 140	89.60 141

CPTT-MVS		91.39 34	90.95 37	91.91 29	95.06 36	95.24 48	95.02 27	88.98 33	91.02 32	86.71 31	84.89 40	88.58 40	91.60 19	90.82 78	89.67 89	94.08 112	96.45 33

CSCG		92.76 23	93.16 25	92.29 26	96.30 25	97.74 5	94.67 30	88.98 33	92.46 20	89.73 17	86.67 34	92.15 14	88.69 39	92.26 51	92.92 39	95.40 54	97.89 9

CDPH-MVS		91.14 36	92.01 30	90.11 40	96.18 26	96.18 36	94.89 28	88.80 35	88.76 45	77.88 81	89.18 27	87.71 44	87.29 57	93.13 39	93.31 34	95.62 43	95.84 43

OPM-MVS		87.56 64	85.80 80	89.62 47	93.90 50	94.09 66	94.12 33	88.18 36	75.40 124	77.30 84	76.41 76	77.93 88	88.79 36	92.20 53	90.82 60	95.40 54	93.72 82

EPNet		89.60 44	89.91 43	89.24 51	96.45 24	93.61 72	92.95 44	88.03 37	85.74 57	83.36 49	87.29 33	83.05 59	80.98 89	92.22 52	91.85 47	93.69 130	95.58 50
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
TSAR-MVS + GP.		92.71 25	93.91 18	91.30 32	91.96 69	96.00 38	93.43 38	87.94 38	92.53 19	86.27 37	93.57 12	91.94 15	91.44 22	93.29 37	92.89 40	96.78 4	97.15 19

PGM-MVS		92.76 23	93.03 26	92.45 25	97.03 13	96.67 25	95.73 20	87.92 39	90.15 40	86.53 33	92.97 17	88.33 41	91.69 18	93.62 34	93.03 36	95.83 30	96.41 35

3Dnovator+	86.06 4	91.60 33	90.86 39	92.47 24	96.00 30	96.50 32	94.70 29	87.83 40	90.49 36	89.92 15	74.68 87	89.35 33	90.66 27	94.02 27	94.14 15	95.67 40	96.85 25

PHI-MVS		92.05 29	93.74 19	90.08 41	94.96 38	97.06 10	93.11 42	87.71 41	90.71 34	80.78 64	92.40 19	91.03 19	87.68 50	94.32 24	94.48 11	96.21 16	96.16 38

HQP-MVS		89.13 48	89.58 46	88.60 58	93.53 53	93.67 70	93.29 40	87.58 42	88.53 46	75.50 85	87.60 31	80.32 71	87.07 58	90.66 84	89.95 81	94.62 91	96.35 37

CANet		91.33 35	91.46 33	91.18 33	95.01 37	96.71 21	93.77 35	87.39 43	87.72 48	87.26 28	81.77 48	89.73 29	87.32 56	94.43 23	93.86 20	96.31 14	96.02 41

PLC	83.76 9	88.61 53	86.83 70	90.70 36	94.22 46	92.63 88	91.50 56	87.19 44	89.16 43	86.87 30	75.51 82	80.87 68	89.98 31	90.01 89	89.20 98	94.41 104	90.45 138
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
abl_6				90.66 37	94.65 44	96.27 34	92.21 47	86.94 45	90.23 38	86.38 34	85.50 37	92.96 8	88.37 43			95.40 54	95.46 52

ACMM	83.27 10	87.68 63	86.09 76	89.54 48	93.26 55	92.19 94	91.43 57	86.74 46	86.02 55	82.85 51	75.63 81	75.14 98	88.41 41	90.68 83	89.99 78	94.59 92	92.97 91
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
MVS_0304		90.88 37	91.35 35	90.34 39	93.91 49	96.79 20	94.49 31	86.54 47	86.57 52	82.85 51	81.68 50	89.70 30	87.57 52	94.64 18	93.93 19	96.67 7	96.15 39

MVS_111021_HR		90.56 38	91.29 36	89.70 46	94.71 43	95.63 44	91.81 54	86.38 48	87.53 49	81.29 58	87.96 29	85.43 49	87.69 49	93.90 30	92.93 38	96.33 12	95.69 46

OMC-MVS		90.23 41	90.40 40	90.03 42	93.45 54	95.29 47	91.89 53	86.34 49	93.25 17	84.94 42	81.72 49	86.65 46	88.90 35	91.69 59	90.27 72	94.65 89	93.95 76

DELS-MVS		89.71 43	89.68 45	89.74 44	93.75 51	96.22 35	93.76 36	85.84 50	82.53 70	85.05 41	78.96 65	84.24 53	84.25 72	94.91 11	94.91 3	95.78 35	96.02 41
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
CNLPA		88.40 54	87.00 68	90.03 42	93.73 52	94.28 62	89.56 76	85.81 51	91.87 26	87.55 26	69.53 116	81.49 65	89.23 33	89.45 98	88.59 108	94.31 108	93.82 79

MSDG		83.87 90	81.02 112	87.19 69	92.17 68	89.80 123	89.15 80	85.72 52	80.61 92	79.24 71	66.66 129	68.75 130	82.69 77	87.95 113	87.44 118	94.19 110	85.92 171

TSAR-MVS + COLMAP		88.40 54	89.09 49	87.60 67	92.72 64	93.92 69	92.21 47	85.57 53	91.73 27	73.72 95	91.75 20	73.22 113	87.64 51	91.49 61	89.71 88	93.73 128	91.82 119

3Dnovator	85.17 5	90.48 39	89.90 44	91.16 34	94.88 40	95.74 43	93.82 34	85.36 54	89.28 42	87.81 25	74.34 89	87.40 45	88.56 40	93.07 40	93.74 24	96.53 9	95.71 45

ACMP	83.90 8	88.32 57	88.06 57	88.62 57	92.18 67	93.98 68	91.28 59	85.24 55	86.69 51	81.23 59	85.62 36	75.13 99	87.01 60	89.83 91	89.77 86	94.79 79	95.43 53
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
LGP-MVS_train		88.25 58	88.55 51	87.89 63	92.84 63	93.66 71	93.35 39	85.22 56	85.77 56	74.03 94	86.60 35	76.29 95	86.62 62	91.20 65	90.58 68	95.29 62	95.75 44

QAPM		89.49 45	89.58 46	89.38 49	94.73 42	95.94 39	92.35 46	85.00 57	85.69 58	80.03 68	76.97 75	87.81 43	87.87 47	92.18 55	92.10 45	96.33 12	96.40 36

TAPA-MVS	84.37 7	88.91 50	88.93 50	88.89 52	93.00 60	94.85 56	92.00 50	84.84 58	91.68 29	80.05 67	79.77 60	84.56 51	88.17 45	90.11 88	89.00 104	95.30 61	92.57 106
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
MVS_111021_LR		90.14 42	90.89 38	89.26 50	93.23 56	94.05 67	90.43 62	84.65 59	90.16 39	84.52 44	90.14 25	83.80 56	87.99 46	92.50 49	90.92 58	94.74 83	94.70 64

UA-Net		86.07 72	87.78 62	84.06 93	92.85 62	95.11 51	87.73 99	84.38 60	73.22 144	73.18 98	79.99 59	89.22 34	71.47 164	93.22 38	93.03 36	94.76 82	90.69 133

UniMVSNet_NR-MVSNet		81.87 107	81.33 108	82.50 110	85.31 140	91.30 99	85.70 128	84.25 61	75.89 120	64.21 141	66.95 128	64.65 142	80.22 102	87.07 120	89.18 99	95.27 64	94.29 69

ACMH	78.52 14	81.86 108	80.45 119	83.51 102	90.51 86	91.22 100	85.62 131	84.23 62	70.29 159	62.21 155	69.04 120	64.05 144	84.48 71	87.57 116	88.45 111	94.01 116	92.54 108
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
TranMVSNet+NR-MVSNet		80.52 119	79.84 128	81.33 124	84.92 149	90.39 108	85.53 133	84.22 63	74.27 133	60.68 170	64.93 142	59.96 165	77.48 131	86.75 128	89.28 96	95.12 71	93.29 86

Anonymous202405211			82.75 98		89.58 97	92.97 83	89.04 85	84.13 64	78.72 106		57.18 177	76.64 94	83.13 76	89.55 96	89.92 82	93.38 138	94.28 72

EPNet_dtu		81.98 106	83.82 91	79.83 140	94.10 48	85.97 165	87.29 106	84.08 65	80.61 92	59.96 172	81.62 52	77.19 92	62.91 187	87.21 118	86.38 137	90.66 169	87.77 158
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
casdiffmvs		87.45 65	87.15 67	87.79 66	90.15 93	94.22 63	89.96 69	83.93 66	85.08 60	80.91 61	75.81 80	77.88 89	86.08 64	91.86 58	90.86 59	95.74 37	94.37 67

PVSNet_BlendedMVS		88.19 59	88.00 59	88.42 59	92.71 65	94.82 57	89.08 82	83.81 67	84.91 62	86.38 34	79.14 62	78.11 86	82.66 78	93.05 41	91.10 53	95.86 27	94.86 60

PVSNet_Blended		88.19 59	88.00 59	88.42 59	92.71 65	94.82 57	89.08 82	83.81 67	84.91 62	86.38 34	79.14 62	78.11 86	82.66 78	93.05 41	91.10 53	95.86 27	94.86 60

baseline1		84.54 84	84.43 88	84.67 81	90.62 82	91.16 101	88.63 91	83.75 69	79.78 97	71.16 105	75.14 84	74.10 103	77.84 129	91.56 60	90.67 65	96.04 19	88.58 147

FC-MVSNet-train		85.18 79	85.31 83	85.03 79	90.67 81	91.62 98	87.66 100	83.61 70	79.75 98	74.37 92	78.69 66	71.21 119	78.91 121	91.23 63	89.96 80	94.96 74	94.69 65

UGNet		85.90 75	88.23 55	83.18 104	88.96 102	94.10 65	87.52 101	83.60 71	81.66 80	77.90 80	80.76 56	83.19 58	66.70 181	91.13 72	90.71 64	94.39 105	96.06 40
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
MAR-MVS		88.39 56	88.44 53	88.33 62	94.90 39	95.06 52	90.51 61	83.59 72	85.27 59	79.07 72	77.13 73	82.89 60	87.70 48	92.19 54	92.32 44	94.23 109	94.20 74
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
DU-MVS		81.20 116	80.30 120	82.25 113	84.98 147	90.94 103	85.70 128	83.58 73	75.74 121	64.21 141	65.30 138	59.60 170	80.22 102	86.89 123	89.31 95	94.77 81	94.29 69

NR-MVSNet		80.25 123	79.98 126	80.56 133	85.20 142	90.94 103	85.65 130	83.58 73	75.74 121	61.36 165	65.30 138	56.75 183	72.38 160	88.46 108	88.80 106	95.16 67	93.87 77

Baseline_NR-MVSNet		79.84 127	78.37 143	81.55 121	84.98 147	86.66 161	85.06 135	83.49 75	75.57 123	63.31 148	58.22 176	60.97 160	78.00 127	86.89 123	87.13 122	94.47 100	93.15 88

OpenMVS	82.53 11	87.71 62	86.84 69	88.73 55	94.42 45	95.06 52	91.02 60	83.49 75	82.50 72	82.24 55	67.62 126	85.48 48	85.56 67	91.19 66	91.30 52	95.67 40	94.75 62

ACMH+	79.08 13	81.84 109	80.06 124	83.91 95	89.92 95	90.62 105	86.21 122	83.48 77	73.88 137	65.75 132	66.38 130	65.30 140	84.63 70	85.90 141	87.25 121	93.45 136	91.13 131

PVSNet_Blended_VisFu		87.40 66	87.80 61	86.92 70	92.86 61	95.40 45	88.56 93	83.45 78	79.55 101	82.26 54	74.49 88	84.03 54	79.24 120	92.97 43	91.53 51	95.15 68	96.65 31

Anonymous20231211		84.42 88	83.02 94	86.05 73	88.85 103	92.70 86	88.92 88	83.40 79	79.99 95	78.31 76	55.83 181	78.92 83	83.33 75	89.06 101	89.76 87	93.50 135	94.90 58

COLMAP_ROB	76.78 15	80.50 121	78.49 139	82.85 106	90.96 79	89.65 129	86.20 123	83.40 79	77.15 114	66.54 127	62.27 151	65.62 139	77.89 128	85.23 151	84.70 158	92.11 151	84.83 175
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
ETV-MVS		89.45 46	90.00 41	88.82 53	91.22 74	94.29 61	90.21 66	83.36 81	86.48 53	81.94 56	83.65 41	81.98 64	89.46 32	94.58 19	93.37 33	96.73 5	95.62 48

UniMVSNet (Re)		81.22 115	81.08 111	81.39 122	85.35 139	91.76 97	84.93 137	82.88 82	76.13 119	65.02 138	64.94 141	63.09 148	75.17 144	87.71 115	89.04 102	94.97 73	94.88 59

EIA-MVS		87.94 61	88.05 58	87.81 64	91.46 70	95.00 54	88.67 89	82.81 83	82.53 70	80.81 63	80.04 58	80.20 72	87.48 53	92.58 48	91.61 50	95.63 42	94.36 68

thres100view900		82.55 102	81.01 114	84.34 85	90.30 91	92.27 92	89.04 85	82.77 84	75.14 125	69.56 112	65.72 133	63.13 146	79.62 115	89.97 90	89.26 97	94.73 84	91.61 126

tfpn200view9		82.86 97	81.46 105	84.48 83	90.30 91	93.09 79	89.05 84	82.71 85	75.14 125	69.56 112	65.72 133	63.13 146	80.38 101	91.15 69	89.51 92	94.91 75	92.50 110

thres400		82.68 100	81.15 110	84.47 84	90.52 84	92.89 84	88.95 87	82.71 85	74.33 132	69.22 117	65.31 137	62.61 152	80.63 96	90.96 76	89.50 93	94.79 79	92.45 112

thres600view7		82.53 103	81.02 112	84.28 88	90.61 83	93.05 80	88.57 92	82.67 87	74.12 135	68.56 120	65.09 140	62.13 157	80.40 100	91.15 69	89.02 103	94.88 76	92.59 104

thres200		82.77 99	81.25 109	84.54 82	90.38 88	93.05 80	89.13 81	82.67 87	74.40 131	69.53 114	65.69 135	63.03 149	80.63 96	91.15 69	89.42 94	94.88 76	92.04 116

DI_MVS_plusplus_trai		86.41 70	85.54 82	87.42 68	89.24 98	93.13 78	92.16 49	82.65 89	82.30 74	80.75 65	68.30 123	80.41 70	85.01 69	90.56 85	90.07 76	94.70 87	94.01 75

TDRefinement		79.05 139	77.05 158	81.39 122	88.45 106	89.00 142	86.92 114	82.65 89	74.21 134	64.41 140	59.17 169	59.16 173	74.52 150	85.23 151	85.09 153	91.37 161	87.51 159

canonicalmvs		89.36 47	89.92 42	88.70 56	91.38 72	95.92 40	91.81 54	82.61 91	90.37 37	82.73 53	82.09 45	79.28 81	88.30 44	91.17 67	93.59 26	95.36 57	97.04 23

IS_MVSNet		86.18 71	88.18 56	83.85 96	91.02 77	94.72 59	87.48 102	82.46 92	81.05 87	70.28 109	76.98 74	82.20 63	76.65 136	93.97 28	93.38 32	95.18 65	94.97 56

CS-MVS		88.97 49	89.44 48	88.41 61	91.45 71	95.24 48	90.03 67	82.43 93	84.08 65	81.16 60	81.02 55	83.83 55	88.74 37	94.25 25	92.73 42	96.67 7	94.95 57

EPP-MVSNet		86.55 68	87.76 63	85.15 78	90.52 84	94.41 60	87.24 108	82.32 94	81.79 79	73.60 96	78.57 67	82.41 61	82.07 83	91.23 63	90.39 70	95.14 69	95.48 51

CLD-MVS		88.66 51	88.52 52	88.82 53	91.37 73	94.22 63	92.82 45	82.08 95	88.27 47	85.14 40	81.86 46	78.53 85	85.93 66	91.17 67	90.61 66	95.55 48	95.00 55
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
Vis-MVSNet (Re-imp)		83.65 93	86.81 71	79.96 138	90.46 87	92.71 85	84.84 139	82.00 96	80.93 89	62.44 154	76.29 77	82.32 62	65.54 184	92.29 50	91.66 48	94.49 99	91.47 128

PatchMatch-RL		83.34 95	81.36 107	85.65 75	90.33 90	89.52 131	84.36 143	81.82 97	80.87 91	79.29 70	74.04 90	62.85 151	86.05 65	88.40 109	87.04 125	92.04 152	86.77 164

UniMVSNet_ETH3D		79.24 137	76.47 163	82.48 111	85.66 135	90.97 102	86.08 125	81.63 98	64.48 183	68.94 119	54.47 183	57.65 179	78.83 122	85.20 154	88.91 105	93.72 129	93.60 83

diffmvs		86.52 69	86.76 72	86.23 72	88.31 108	92.63 88	89.58 75	81.61 99	86.14 54	80.26 66	79.00 64	77.27 91	83.58 73	88.94 102	89.06 101	94.05 114	94.29 69

IB-MVS	79.09 12	82.60 101	82.19 100	83.07 105	91.08 76	93.55 73	80.90 170	81.35 100	76.56 116	80.87 62	64.81 143	69.97 124	68.87 171	85.64 144	90.06 77	95.36 57	94.74 63
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
tfpnnormal		77.46 155	74.86 180	80.49 134	86.34 128	88.92 143	84.33 144	81.26 101	61.39 191	61.70 162	51.99 190	53.66 194	74.84 147	88.63 105	87.38 120	94.50 97	92.08 114

TransMVSNet (Re)		76.57 162	75.16 179	78.22 154	85.60 136	87.24 157	82.46 155	81.23 102	59.80 194	59.05 178	57.07 178	59.14 174	66.60 182	88.09 111	86.82 127	94.37 106	87.95 156

Vis-MVSNet		84.38 89	86.68 73	81.70 118	87.65 115	94.89 55	88.14 95	80.90 103	74.48 130	68.23 121	77.53 72	80.72 69	69.98 168	92.68 46	91.90 46	95.33 60	94.58 66
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
DCV-MVSNet		85.88 76	86.17 74	85.54 77	89.10 101	89.85 121	89.34 78	80.70 104	83.04 68	78.08 79	76.19 78	79.00 82	82.42 81	89.67 94	90.30 71	93.63 133	95.12 54

WR-MVS		76.63 161	78.02 149	75.02 175	84.14 157	89.76 125	78.34 183	80.64 105	69.56 160	52.32 187	61.26 154	61.24 159	60.66 188	84.45 162	87.07 123	93.99 117	92.77 97

ET-MVSNet_ETH3D		84.65 83	85.58 81	83.56 100	74.99 201	92.62 90	90.29 64	80.38 106	82.16 75	73.01 101	83.41 42	71.10 120	87.05 59	87.77 114	90.17 74	95.62 43	91.82 119

GBi-Net		84.51 85	84.80 85	84.17 90	84.20 154	89.95 116	89.70 72	80.37 107	81.17 83	75.50 85	69.63 111	79.69 78	79.75 112	90.73 80	90.72 61	95.52 50	91.71 121

test1		84.51 85	84.80 85	84.17 90	84.20 154	89.95 116	89.70 72	80.37 107	81.17 83	75.50 85	69.63 111	79.69 78	79.75 112	90.73 80	90.72 61	95.52 50	91.71 121

FMVSNet3		84.44 87	84.64 87	84.21 89	84.32 153	90.13 114	89.85 71	80.37 107	81.17 83	75.50 85	69.63 111	79.69 78	79.62 115	89.72 93	90.52 69	95.59 46	91.58 127

CDS-MVSNet		81.63 113	82.09 101	81.09 127	87.21 120	90.28 110	87.46 104	80.33 110	69.06 163	70.66 106	71.30 102	73.87 105	67.99 174	89.58 95	89.87 83	92.87 145	90.69 133
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
Effi-MVS+		85.33 78	85.08 84	85.63 76	89.69 96	93.42 75	89.90 70	80.31 111	79.32 102	72.48 104	73.52 95	74.03 104	86.55 63	90.99 74	89.98 79	94.83 78	94.27 73

FMVSNet2		83.87 90	83.73 92	84.05 94	84.20 154	89.95 116	89.70 72	80.21 112	79.17 104	74.89 89	65.91 131	77.49 90	79.75 112	90.87 77	91.00 57	95.52 50	91.71 121

MVS_Test		86.93 67	87.24 66	86.56 71	90.10 94	93.47 74	90.31 63	80.12 113	83.55 67	78.12 77	79.58 61	79.80 76	85.45 68	90.17 87	90.59 67	95.29 62	93.53 85

thisisatest0530		85.15 80	85.86 78	84.33 86	89.19 100	92.57 91	87.22 109	80.11 114	82.15 76	74.41 91	78.15 69	73.80 107	79.90 108	90.99 74	89.58 90	95.13 70	93.75 81

tttt0517		85.11 81	85.81 79	84.30 87	89.24 98	92.68 87	87.12 113	80.11 114	81.98 77	74.31 93	78.08 70	73.57 109	79.90 108	91.01 73	89.58 90	95.11 72	93.77 80

DTE-MVSNet		75.14 178	75.44 177	74.80 177	83.18 166	87.19 158	78.25 185	80.11 114	66.05 175	48.31 194	60.88 159	54.67 190	64.54 185	82.57 174	86.17 140	94.43 103	90.53 137

PEN-MVS		76.02 171	76.07 167	75.95 170	83.17 167	87.97 151	79.65 174	80.07 117	66.57 173	51.45 189	60.94 158	55.47 188	66.81 180	82.72 172	86.80 128	94.59 92	92.03 117

MVSTER		86.03 73	86.12 75	85.93 74	88.62 105	89.93 119	89.33 79	79.91 118	81.87 78	81.35 57	81.07 54	74.91 100	80.66 95	92.13 56	90.10 75	95.68 39	92.80 96

v2v482		79.84 127	78.07 146	81.90 116	83.75 159	90.21 113	87.17 110	79.85 119	70.65 155	65.93 131	61.93 152	60.07 164	80.82 90	85.25 150	86.71 129	93.88 122	91.70 124

CP-MVSNet		76.36 168	76.41 164	76.32 167	82.73 176	88.64 145	79.39 177	79.62 120	67.21 169	53.70 183	60.72 160	55.22 189	67.91 176	83.52 168	86.34 138	94.55 95	93.19 87

FMVSNet1		81.64 112	80.61 117	82.84 107	82.36 179	89.20 137	88.67 89	79.58 121	70.79 154	72.63 103	58.95 172	72.26 116	79.34 118	90.73 80	90.72 61	94.47 100	91.62 125

PS-CasMVS		75.90 173	75.86 172	75.96 169	82.59 177	88.46 148	79.23 180	79.56 122	66.00 176	52.77 185	59.48 168	54.35 192	67.14 179	83.37 169	86.23 139	94.47 100	93.10 89

RPSCF		83.46 94	83.36 93	83.59 99	87.75 111	87.35 156	84.82 140	79.46 123	83.84 66	78.12 77	82.69 44	79.87 74	82.60 80	82.47 175	81.13 179	88.78 181	86.13 169

pm-mvs1		78.51 147	77.75 152	79.40 141	84.83 150	89.30 134	83.55 150	79.38 124	62.64 187	63.68 146	58.73 174	64.68 141	70.78 167	89.79 92	87.84 114	94.17 111	91.28 130

WR-MVS_H		75.84 174	76.93 160	74.57 180	82.86 173	89.50 132	78.34 183	79.36 125	66.90 171	52.51 186	60.20 164	59.71 167	59.73 189	83.61 167	85.77 147	94.65 89	92.84 94

v148		78.59 145	76.84 161	80.62 132	83.61 162	89.16 138	83.65 149	79.24 126	69.38 161	69.34 116	59.88 166	60.41 163	75.19 143	83.81 166	84.63 159	92.70 147	90.63 135

CHOSEN 1792x2688		82.16 104	80.91 115	83.61 98	91.14 75	92.01 95	89.55 77	79.15 127	79.87 96	70.29 108	52.51 189	72.56 114	81.39 85	88.87 104	88.17 112	90.15 173	92.37 113

baseline2		82.80 98	82.86 97	82.73 109	87.68 114	90.50 107	84.92 138	78.93 128	78.07 111	73.06 99	75.08 85	69.77 125	77.31 132	88.90 103	86.94 126	94.50 97	90.74 132

USDC		80.69 118	79.89 127	81.62 120	86.48 126	89.11 140	86.53 119	78.86 129	81.15 86	63.48 147	72.98 97	59.12 175	81.16 87	87.10 119	85.01 154	93.23 139	84.77 176

FC-MVSNet-test		76.53 164	81.62 104	70.58 188	84.99 146	85.73 168	74.81 191	78.85 130	77.00 115	39.13 205	75.90 79	73.50 110	54.08 195	86.54 133	85.99 145	91.65 157	86.68 165

pmmvs4		79.99 124	78.08 145	82.22 114	83.04 169	87.16 159	84.95 136	78.80 131	78.64 107	74.53 90	64.61 144	59.41 171	79.45 117	84.13 164	84.54 161	92.53 148	88.08 153

IterMVS-LS		83.28 96	82.95 96	83.65 97	88.39 107	88.63 146	86.80 117	78.64 132	76.56 116	73.43 97	72.52 100	75.35 97	80.81 92	86.43 136	88.51 110	93.84 124	92.66 101
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
GA-MVS		79.52 132	79.71 131	79.30 142	85.68 134	90.36 109	84.55 141	78.44 133	70.47 158	57.87 180	68.52 122	61.38 158	76.21 138	89.40 99	87.89 113	93.04 143	89.96 140

v1144		79.38 136	77.83 150	81.18 126	83.62 161	90.23 111	87.15 112	78.35 134	69.13 162	64.02 144	60.20 164	59.41 171	80.14 106	86.78 126	86.57 133	93.81 126	92.53 109

HyFIR lowres test		81.62 114	79.45 134	84.14 92	91.00 78	93.38 76	88.27 94	78.19 135	76.28 118	70.18 110	48.78 193	73.69 108	83.52 74	87.05 121	87.83 116	93.68 131	89.15 144

TinyColmap		76.73 159	73.95 183	79.96 138	85.16 144	85.64 170	82.34 158	78.19 135	70.63 156	62.06 157	60.69 161	49.61 199	80.81 92	85.12 155	83.69 166	91.22 165	82.27 184

Effi-MVS+-dtu		82.05 105	81.76 102	82.38 112	87.72 112	90.56 106	86.90 116	78.05 137	73.85 138	66.85 126	71.29 103	71.90 117	82.00 84	86.64 131	85.48 150	92.76 146	92.58 105

test-LLR		79.47 134	79.84 128	79.03 144	87.47 116	82.40 190	81.24 167	78.05 137	73.72 139	62.69 151	73.76 92	74.42 101	73.49 155	84.61 160	82.99 170	91.25 163	87.01 162

test0.0.03 1		76.03 170	78.51 138	73.12 185	87.47 116	85.13 176	76.32 188	78.05 137	73.19 146	50.98 192	70.64 105	69.28 128	55.53 191	85.33 149	84.38 162	90.39 171	81.63 187

v1192		78.94 140	77.33 154	80.82 129	83.25 165	89.90 120	86.91 115	77.72 140	68.63 166	62.61 153	59.17 169	57.53 180	80.62 98	86.89 123	86.47 135	93.79 127	92.75 99

Fast-Effi-MVS+		83.77 92	82.98 95	84.69 80	87.98 109	91.87 96	88.10 96	77.70 141	78.10 110	73.04 100	69.13 118	68.51 131	86.66 61	90.49 86	89.85 84	94.67 88	92.88 93

v144192		78.81 141	77.22 156	80.67 131	82.95 170	89.79 124	86.40 120	77.42 142	68.26 168	63.13 149	59.50 167	58.13 177	80.08 107	85.93 140	86.08 142	94.06 113	92.83 95

v8		79.90 126	78.39 142	81.66 119	83.97 158	89.81 122	87.16 111	77.40 143	71.49 150	67.71 122	61.24 155	62.49 153	79.83 111	85.48 148	86.17 140	93.89 121	92.02 118

LTVRE_ROB	74.41 16	75.78 175	74.72 181	77.02 161	85.88 130	89.22 136	82.44 157	77.17 144	50.57 204	45.45 198	65.44 136	52.29 196	81.25 86	85.50 147	87.42 119	89.94 175	92.62 102
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		78.57 146	76.99 159	80.41 136	82.93 171	89.63 130	86.38 121	77.14 145	68.31 167	61.80 161	58.89 173	56.79 182	80.19 105	86.50 135	86.05 144	94.02 115	92.76 98

pmmvs6		74.83 179	72.89 186	77.09 159	82.11 180	87.50 155	80.88 171	76.97 146	52.79 202	61.91 160	46.66 195	60.49 161	69.28 170	86.74 129	85.46 151	91.39 160	90.56 136

thisisatest0515		79.76 129	80.59 118	78.80 146	84.40 152	88.91 144	79.48 176	76.94 147	72.29 148	67.33 124	67.82 125	65.99 137	70.80 166	88.50 107	87.84 114	93.86 123	92.75 99

v10		79.62 130	78.19 144	81.28 125	83.73 160	89.69 127	87.27 107	76.86 148	70.50 157	65.46 133	60.58 162	60.47 162	80.44 99	86.91 122	86.63 132	93.93 118	92.55 107

v1240		78.15 148	76.53 162	80.04 137	82.85 174	89.48 133	85.61 132	76.77 149	67.05 170	61.18 168	58.37 175	56.16 186	79.89 110	86.11 139	86.08 142	93.92 119	92.47 111

V42		79.59 131	78.43 141	80.94 128	82.79 175	89.71 126	86.66 118	76.73 150	71.38 151	67.42 123	61.01 157	62.30 155	78.39 124	85.56 146	86.48 134	93.65 132	92.60 103

v7n		77.22 156	76.23 166	78.38 153	81.89 182	89.10 141	82.24 161	76.36 151	65.96 177	61.21 167	56.56 179	55.79 187	75.07 146	86.55 132	86.68 130	93.52 134	92.95 92

DWT-MVSNet_training		80.51 120	78.05 148	83.39 103	88.64 104	88.33 150	86.11 124	76.33 152	79.65 99	78.64 74	69.62 114	58.89 176	80.82 90	80.50 182	82.03 177	89.77 176	87.36 160

SixPastTwentyTwo		76.02 171	75.72 173	76.36 166	83.38 163	87.54 154	75.50 190	76.22 153	65.50 180	57.05 181	70.64 105	53.97 193	74.54 149	80.96 180	82.12 175	91.44 159	89.35 143

MDA-MVSNet-bldmvs		66.22 193	64.49 196	68.24 191	61.67 205	82.11 192	70.07 198	76.16 154	59.14 196	47.94 195	54.35 184	35.82 209	67.33 178	64.94 203	75.68 193	86.30 193	79.36 193

test20.03		68.31 191	70.05 192	66.28 195	82.41 178	80.84 194	67.35 200	76.11 155	58.44 197	40.80 204	53.77 186	54.54 191	42.28 202	83.07 170	81.96 178	88.73 182	77.76 197

pmmvs-eth3d		74.32 182	71.96 188	77.08 160	77.33 195	82.71 186	78.41 182	76.02 156	66.65 172	65.98 130	54.23 185	49.02 201	73.14 159	82.37 176	82.69 172	91.61 158	86.05 170

Anonymous20231206		70.80 187	70.59 191	71.04 187	81.60 184	82.49 189	74.64 192	75.87 157	64.17 184	49.27 193	44.85 199	53.59 195	54.68 194	83.07 170	82.34 174	90.17 172	83.65 179

baseline		84.89 82	86.06 77	83.52 101	87.25 119	89.67 128	87.76 98	75.68 158	84.92 61	78.40 75	80.10 57	80.98 67	80.20 104	86.69 130	87.05 124	91.86 155	92.99 90

CANet_DTU		85.43 77	87.72 65	82.76 108	90.95 80	93.01 82	89.99 68	75.46 159	82.67 69	64.91 139	83.14 43	80.09 73	80.68 94	92.03 57	91.03 55	94.57 94	92.08 114

MS-PatchMatch		81.79 110	81.44 106	82.19 115	90.35 89	89.29 135	88.08 97	75.36 160	77.60 112	69.00 118	64.37 146	78.87 84	77.14 135	88.03 112	85.70 148	93.19 141	86.24 168

CVMVSNet		76.70 160	78.46 140	74.64 179	83.34 164	84.48 178	81.83 163	74.58 161	68.88 164	51.23 191	69.77 110	70.05 123	67.49 177	84.27 163	83.81 164	89.38 178	87.96 155

testgi		71.92 186	74.20 182	69.27 190	84.58 151	83.06 182	73.40 193	74.39 162	64.04 185	46.17 197	68.90 121	57.15 181	48.89 199	84.07 165	83.08 169	88.18 184	79.09 195

pmmvs5		76.93 158	76.33 165	77.62 156	81.97 181	88.40 149	81.32 166	74.35 163	65.42 181	61.42 164	63.07 149	57.95 178	73.23 158	85.60 145	85.35 152	93.41 137	88.55 148

MIMVSNet1		65.00 194	66.24 195	63.55 197	58.41 208	80.01 196	69.00 199	74.03 164	55.81 200	41.88 202	36.81 204	49.48 200	47.89 200	81.32 179	82.40 173	90.08 174	77.88 196

EG-PatchMatch MVS		76.40 167	75.47 176	77.48 157	85.86 132	90.22 112	82.45 156	73.96 165	59.64 195	59.60 174	52.75 188	62.20 156	68.44 173	88.23 110	87.50 117	94.55 95	87.78 157

CMPMVS	56.49 17	73.84 184	71.73 189	76.31 168	85.20 142	85.67 169	75.80 189	73.23 166	62.26 188	65.40 134	53.40 187	59.70 168	71.77 163	80.25 184	79.56 182	86.45 192	81.28 188
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
anonymousdsp		77.94 150	79.00 135	76.71 163	79.03 190	87.83 152	79.58 175	72.87 167	65.80 178	58.86 179	65.82 132	62.48 154	75.99 139	86.77 127	88.66 107	93.92 119	95.68 47

Fast-Effi-MVS+-dtu		79.95 125	80.69 116	79.08 143	86.36 127	89.14 139	85.85 126	72.28 168	72.85 147	59.32 175	70.43 109	68.42 132	77.57 130	86.14 138	86.44 136	93.11 142	91.39 129

IterMVS-SCA-FT		79.41 135	80.20 122	78.49 151	85.88 130	86.26 163	83.95 146	71.94 169	73.55 142	61.94 158	70.48 108	70.50 121	75.23 142	85.81 143	84.61 160	91.99 154	90.18 139

PM-MVS		74.17 183	73.10 184	75.41 172	76.07 198	82.53 188	77.56 186	71.69 170	71.04 152	61.92 159	61.23 156	47.30 202	74.82 148	81.78 178	79.80 180	90.42 170	88.05 154

TAMVS		76.42 165	77.16 157	75.56 171	83.05 168	85.55 171	80.58 172	71.43 171	65.40 182	61.04 169	67.27 127	69.22 129	67.99 174	84.88 158	84.78 157	89.28 179	83.01 182

N_pmnet		66.85 192	66.63 193	67.11 194	78.73 191	74.66 202	70.53 197	71.07 172	66.46 174	46.54 196	51.68 191	51.91 197	55.48 192	74.68 197	72.38 198	80.29 203	74.65 200

new-patchmatchnet		63.80 195	63.31 197	64.37 196	76.49 196	75.99 200	63.73 203	70.99 173	57.27 198	43.08 200	45.86 197	43.80 203	45.13 201	73.20 198	70.68 201	86.80 191	76.34 199

CostFormer		80.94 117	80.21 121	81.79 117	87.69 113	88.58 147	87.47 103	70.66 174	80.02 94	77.88 81	73.03 96	71.40 118	78.24 125	79.96 185	79.63 181	88.82 180	88.84 145

tpm cat1		77.78 152	75.28 178	80.70 130	87.14 121	85.84 167	85.81 127	70.40 175	77.44 113	78.80 73	63.72 147	64.01 145	76.55 137	75.60 196	75.21 195	85.51 196	85.12 173

MDTV_nov1_ep13		79.14 138	79.49 133	78.74 148	85.40 138	86.89 160	84.32 145	70.29 176	78.85 105	69.42 115	75.37 83	73.29 112	75.64 141	80.61 181	79.48 183	87.36 187	81.91 185

FMVSNet5		75.50 177	76.07 167	74.83 176	76.16 197	81.19 193	81.34 165	70.21 177	73.20 145	61.59 163	58.97 171	68.33 133	68.50 172	85.87 142	85.85 146	91.18 166	79.11 194

dps		78.02 149	75.94 171	80.44 135	86.06 129	86.62 162	82.58 154	69.98 178	75.14 125	77.76 83	69.08 119	59.93 166	78.47 123	79.47 187	77.96 188	87.78 185	83.40 180

EU-MVSNet		69.98 189	72.30 187	67.28 193	75.67 200	79.39 197	73.12 194	69.94 179	63.59 186	42.80 201	62.93 150	56.71 184	55.07 193	79.13 190	78.55 186	87.06 190	85.82 172

IterMVS		78.79 142	79.71 131	77.71 155	85.26 141	85.91 166	84.54 142	69.84 180	73.38 143	61.25 166	70.53 107	70.35 122	74.43 151	85.21 153	83.80 165	90.95 167	88.77 146
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
PMVS	50.48 18	55.81 200	51.93 202	60.33 199	72.90 202	49.34 208	48.78 207	69.51 181	43.49 207	54.25 182	36.26 205	41.04 208	39.71 204	65.07 202	60.70 203	76.85 205	67.58 204
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
FPMVS		63.63 196	60.08 201	67.78 192	80.01 188	71.50 204	72.88 195	69.41 182	61.82 190	53.11 184	45.12 198	42.11 206	50.86 197	66.69 201	63.84 202	80.41 202	69.46 203

SCA		79.51 133	80.15 123	78.75 147	86.58 125	87.70 153	83.07 152	68.53 183	81.31 82	66.40 128	73.83 91	75.38 96	79.30 119	80.49 183	79.39 184	88.63 183	82.96 183

CR-MVSNet		78.71 143	78.86 136	78.55 150	85.85 133	85.15 174	82.30 159	68.23 184	74.71 128	65.37 135	64.39 145	69.59 127	77.18 133	85.10 156	84.87 155	92.34 150	88.21 151

Patchmtry						85.54 172	82.30 159	68.23 184		65.37 135

PatchmatchNet		78.67 144	78.85 137	78.46 152	86.85 124	86.03 164	83.77 148	68.11 186	80.88 90	66.19 129	72.90 98	73.40 111	78.06 126	79.25 189	77.71 189	87.75 186	81.75 186
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
MIMVSNet		74.69 180	75.60 175	73.62 182	76.02 199	85.31 173	81.21 169	67.43 187	71.02 153	59.07 177	54.48 182	64.07 143	66.14 183	86.52 134	86.64 131	91.83 156	81.17 189

PMMVS		81.65 111	84.05 90	78.86 145	78.56 192	82.63 187	83.10 151	67.22 188	81.39 81	70.11 111	84.91 39	79.74 77	82.12 82	87.31 117	85.70 148	92.03 153	86.67 167

tpm		76.30 169	76.05 169	76.59 164	86.97 122	83.01 184	83.83 147	67.06 189	71.83 149	63.87 145	69.56 115	62.88 150	73.41 157	79.79 186	78.59 185	84.41 197	86.68 165

CHOSEN 280x420		80.28 122	81.66 103	78.67 149	82.92 172	79.24 198	85.36 134	66.79 190	78.11 109	70.32 107	75.03 86	79.87 74	81.09 88	89.07 100	83.16 168	85.54 195	87.17 161

EPMVS		77.53 154	78.07 146	76.90 162	86.89 123	84.91 177	82.18 162	66.64 191	81.00 88	64.11 143	72.75 99	69.68 126	74.42 152	79.36 188	78.13 187	87.14 189	80.68 191

MDTV_nov1_ep13_2view		73.21 185	72.91 185	73.56 183	80.01 188	84.28 180	78.62 181	66.43 192	68.64 165	59.12 176	60.39 163	59.69 169	69.81 169	78.82 191	77.43 190	87.36 187	81.11 190

tpmrst		76.55 163	75.99 170	77.20 158	87.32 118	83.05 183	82.86 153	65.62 193	78.61 108	67.22 125	69.19 117	65.71 138	75.87 140	76.75 194	75.33 194	84.31 198	83.28 181

gm-plane-assit		70.29 188	70.65 190	69.88 189	85.03 145	78.50 199	58.41 206	65.47 194	50.39 205	40.88 203	49.60 192	50.11 198	75.14 145	91.43 62	89.78 85	94.32 107	84.73 177

RPMNet		77.07 157	77.63 153	76.42 165	85.56 137	85.15 174	81.37 164	65.27 195	74.71 128	60.29 171	63.71 148	66.59 136	73.64 154	82.71 173	82.12 175	92.38 149	88.39 149

MVS-HIRNet		68.83 190	66.39 194	71.68 186	77.58 194	75.52 201	66.45 201	65.05 196	62.16 189	62.84 150	44.76 200	56.60 185	71.96 162	78.04 192	75.06 196	86.18 194	72.56 201

PatchT		76.42 165	77.81 151	74.80 177	78.46 193	84.30 179	71.82 196	65.03 197	73.89 136	65.37 135	61.58 153	66.70 135	77.18 133	85.10 156	84.87 155	90.94 168	88.21 151

gg-mvs-nofinetune		75.64 176	77.26 155	73.76 181	87.92 110	92.20 93	87.32 105	64.67 198	51.92 203	35.35 206	46.44 196	77.05 93	71.97 161	92.64 47	91.02 56	95.34 59	89.53 142

Gipuma		49.17 201	47.05 203	51.65 201	59.67 207	48.39 209	41.98 209	63.47 199	55.64 201	33.33 208	14.90 207	13.78 213	41.34 203	69.31 200	72.30 199	70.11 206	55.00 207
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
ADS-MVSNet		74.53 181	75.69 174	73.17 184	81.57 185	80.71 195	79.27 179	63.03 200	79.27 103	59.94 173	67.86 124	68.32 134	71.08 165	77.33 193	76.83 191	84.12 200	79.53 192

TESTMET0.1,1		77.78 152	79.84 128	75.38 173	80.86 187	82.40 190	81.24 167	62.72 201	73.72 139	62.69 151	73.76 92	74.42 101	73.49 155	84.61 160	82.99 170	91.25 163	87.01 162

test-mter		77.79 151	80.02 125	75.18 174	81.18 186	82.85 185	80.52 173	62.03 202	73.62 141	62.16 156	73.55 94	73.83 106	73.81 153	84.67 159	83.34 167	91.37 161	88.31 150

new_pmnet		59.28 198	61.47 200	56.73 200	61.66 206	68.29 206	59.57 205	54.91 203	60.83 192	34.38 207	44.66 201	43.65 204	49.90 198	71.66 199	71.56 200	79.94 204	69.67 202

E-PMN		31.40 203	26.80 206	36.78 203	51.39 210	29.96 212	20.20 212	54.17 204	25.93 210	12.75 211	14.73 208	8.58 215	34.10 206	27.36 208	37.83 207	48.07 210	43.18 209

EMVS		30.49 205	25.44 207	36.39 204	51.47 209	29.89 213	20.17 213	54.00 205	26.49 209	12.02 212	13.94 210	8.84 214	34.37 205	25.04 209	34.37 208	46.29 211	39.53 210

pmmvs3		61.89 197	61.74 199	62.06 198	64.30 204	70.83 205	64.22 202	52.14 206	48.78 206	44.47 199	41.67 202	41.70 207	63.03 186	76.06 195	76.02 192	84.18 199	77.14 198

PMMVS2		41.68 202	44.74 204	38.10 202	46.97 211	52.32 207	40.63 210	48.08 207	35.51 208	7.36 213	26.86 206	24.64 211	16.72 208	55.24 205	59.03 204	68.85 207	59.59 206

MVE	30.17 19	30.88 204	33.52 205	27.80 207	23.78 213	39.16 211	18.69 214	46.90 208	21.88 211	15.39 210	14.37 209	7.31 216	24.41 207	41.63 207	56.22 205	37.64 212	54.07 208
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
DeepMVS_CX						48.31 210	48.03 208	26.08 209	56.42 199	25.77 209	47.51 194	31.31 210	51.30 196	48.49 206		53.61 208	61.52 205

tmp_tt				32.73 205	43.96 212	21.15 214	26.71 211	8.99 210	65.67 179	51.39 190	56.01 180	42.64 205	11.76 209	56.60 204	50.81 206	53.55 209

testmvs		1.03 206	1.63 208	0.34 208	0.09 215	0.35 215	0.61 216	0.16 211	1.49 212	0.10 215	3.15 211	0.15 217	0.86 211	1.32 210	1.18 209	0.20 213	3.76 212

GG-mvs-BLEND		57.56 199	82.61 99	28.34 206	0.22 214	90.10 115	79.37 178	0.14 212	79.56 100	0.40 214	71.25 104	83.40 57	0.30 212	86.27 137	83.87 163	89.59 177	83.83 178

test123		0.87 207	1.40 209	0.25 209	0.03 216	0.25 216	0.35 217	0.08 213	1.21 213	0.05 216	2.84 212	0.03 218	0.89 210	0.43 211	1.16 210	0.13 214	3.87 211

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	0.00 219	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	0.00 219	0.00 213	0.00 212	0.00 211	0.00 215	0.00 213

our_test_3					81.81 183	83.96 181	76.61 187

test_part1																	98.10 4

ambc			61.92 198		70.98 203	73.54 203	63.64 204		60.06 193	52.23 188	38.44 203	19.17 212	57.12 190	82.33 177	75.03 197	83.21 201	84.89 174

MTAPA										92.97 2		91.03 19

MTMP										93.14 1		90.21 27

Patchmatch-RL test							8.55 215

XVS					93.11 57	96.70 22	91.91 51			83.95 45		88.82 37				95.79 33

X-MVStestdata					93.11 57	96.70 22	91.91 51			83.95 45		88.82 37				95.79 33

mPP-MVS					97.06 12							88.08 42

NP-MVS									87.47 50

High-resolution multi-view benchmark