Altmetrics
Downloads
69
Views
25
Comments
0
A peer-reviewed article of this preprint also exists.
This version is not peer-reviewed
EER(%) | |||||||||
FV_USM | SDUMLA | MMCBNU _6000 | HKPU_FV | THU_FVD | SCUT_RIFV | UTFVP | PLUSVein | VERA | |
FV_CNN [2] | - | 6.42 | - | 4.67 | - | - | - | - | - |
Fvras-net [3] | 0.95 | 1.71 | 1.11 | - | - | - | - | - | - |
FV code [29] | - | - | - | 3.33 | - | - | - | - | - |
L-CNN [30] | - | 1.13 | - | 0.67 | - | - | - | - | - |
ArcVein [31] | 0.25 | 1.53 | - | 1.3 | - | - | - | - | - |
FVSR-Net[32] | - | 5.27 | - | - | - | - | - | - | - |
S-CNN [33] | - | 2.29 | 0.47 | - | - | - | - | - | - |
FVT [6] | 0.44 | 1.5 | 0.92 | 2.37 | 3.6 | 1.65 | 1.97 | 2.08 | 4.55 |
FVFSNet [34] | 0.20 | 1.10 | 0.18 | 0.81 | 2.15 | 0.83 | 2.08 | 1.32 | 6.82 |
Let-Net(ours) | 0.04 | 0.15 | 0.12 | 1.54 | 2.13 | 1.12 | 1.58 | 1.12 | 3.87 |
Method | FV_USM EER(%) | SDUMLA EER(%) | Parameters(M) | |
Kernel Size | 99.57 | 99.1 | 0.72 | |
99.66 | 99.35 | 0.81 | ||
99.77 | 99.42 | 0.89 | ||
99.68 | 99.34 | 1.08 | ||
99.66 | 99.33 | 1.67 | ||
Components of Let-Net |
No Stem | 98.25 | 97.86 | 0.51 |
No LK | 96.65 | 96.27 | 0.78 | |
No NAM | 95.76 | 95.17 | 0.66 | |
No Stem&Lk | 94.71 | 94.16 | 0.52 | |
No Stem&NAM | 93.64 | 93.11 | 0.27 | |
No LK&NAM | 88.12 | 87.76 | 0.55 | |
Stem&LK&NAM | 99.77 | 99.5 | 0.89 | |
Large Kernel Architecture |
Direct Connection | 96.32 | 96.01 | 0.88 |
Parallel Connection | 98.46 | 97.26 | 0.89 | |
Funnel Connection | 98.26 | 97.49 | 0.89 | |
Taper Connection | 99.77 | 99.5 | 0.89 |
Model | Params(M) | FLOPs(G) | EER(%)* | ACC(%)* |
ResNet50V2 | 23.63 | 6.99 | 3.04 | 93.28 |
DensNet121 | 7.07 | 5.70 | 2.57 | 92.79 |
Xception | 2.09 | 16.8 | 1.95 | 93.77 |
Let-Net(ours) | 0.89 | 0.25 | 1.26 | 94.84 |
Training(s) | Prediction(s) | Total(s) | Single batch time(ms) | |
VGG16 | 10 | 5 | 15 | 48 |
VGG19 | 12 | 6 | 18 | 55 |
Resnet50V2 | 11 | 6 | 17 | 53 |
InceptionV3 | 16 | 9 | 25 | 78 |
DensNet121 | 22 | 11 | 33 | 102 |
Xception | 19 | 6 | 25 | 77 |
RepLKNet | 96 | 6 | 120 | 373 |
Let-Net(ours) | 3 | 3 | 6 | 17 |
Training(s) | Prediction(s) | Total(s) | Single batch time(ms) | |
VGG16 | 13 | 7 | 20 | 48 |
VGG19 | 15 | 8 | 23 | 55 |
Resnet50V2 | 14 | 9 | 23 | 54 |
InceptionV3 | 20 | 12 | 32 | 77 |
DensNet121 | 26 | 15 | 41 | 97 |
Xception | 25 | 7 | 32 | 78 |
RepLKNet | 126 | 32 | 158 | 379 |
Let-Net(ours) | 4 | 3 | 7 | 18 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
Submitted:
10 January 2024
Posted:
10 January 2024
You are already at the latest version
A peer-reviewed article of this preprint also exists.
This version is not peer-reviewed
Submitted:
10 January 2024
Posted:
10 January 2024
You are already at the latest version
EER(%) | |||||||||
FV_USM | SDUMLA | MMCBNU _6000 | HKPU_FV | THU_FVD | SCUT_RIFV | UTFVP | PLUSVein | VERA | |
FV_CNN [2] | - | 6.42 | - | 4.67 | - | - | - | - | - |
Fvras-net [3] | 0.95 | 1.71 | 1.11 | - | - | - | - | - | - |
FV code [29] | - | - | - | 3.33 | - | - | - | - | - |
L-CNN [30] | - | 1.13 | - | 0.67 | - | - | - | - | - |
ArcVein [31] | 0.25 | 1.53 | - | 1.3 | - | - | - | - | - |
FVSR-Net[32] | - | 5.27 | - | - | - | - | - | - | - |
S-CNN [33] | - | 2.29 | 0.47 | - | - | - | - | - | - |
FVT [6] | 0.44 | 1.5 | 0.92 | 2.37 | 3.6 | 1.65 | 1.97 | 2.08 | 4.55 |
FVFSNet [34] | 0.20 | 1.10 | 0.18 | 0.81 | 2.15 | 0.83 | 2.08 | 1.32 | 6.82 |
Let-Net(ours) | 0.04 | 0.15 | 0.12 | 1.54 | 2.13 | 1.12 | 1.58 | 1.12 | 3.87 |
Method | FV_USM EER(%) | SDUMLA EER(%) | Parameters(M) | |
Kernel Size | 99.57 | 99.1 | 0.72 | |
99.66 | 99.35 | 0.81 | ||
99.77 | 99.42 | 0.89 | ||
99.68 | 99.34 | 1.08 | ||
99.66 | 99.33 | 1.67 | ||
Components of Let-Net |
No Stem | 98.25 | 97.86 | 0.51 |
No LK | 96.65 | 96.27 | 0.78 | |
No NAM | 95.76 | 95.17 | 0.66 | |
No Stem&Lk | 94.71 | 94.16 | 0.52 | |
No Stem&NAM | 93.64 | 93.11 | 0.27 | |
No LK&NAM | 88.12 | 87.76 | 0.55 | |
Stem&LK&NAM | 99.77 | 99.5 | 0.89 | |
Large Kernel Architecture |
Direct Connection | 96.32 | 96.01 | 0.88 |
Parallel Connection | 98.46 | 97.26 | 0.89 | |
Funnel Connection | 98.26 | 97.49 | 0.89 | |
Taper Connection | 99.77 | 99.5 | 0.89 |
Model | Params(M) | FLOPs(G) | EER(%)* | ACC(%)* |
ResNet50V2 | 23.63 | 6.99 | 3.04 | 93.28 |
DensNet121 | 7.07 | 5.70 | 2.57 | 92.79 |
Xception | 2.09 | 16.8 | 1.95 | 93.77 |
Let-Net(ours) | 0.89 | 0.25 | 1.26 | 94.84 |
Training(s) | Prediction(s) | Total(s) | Single batch time(ms) | |
VGG16 | 10 | 5 | 15 | 48 |
VGG19 | 12 | 6 | 18 | 55 |
Resnet50V2 | 11 | 6 | 17 | 53 |
InceptionV3 | 16 | 9 | 25 | 78 |
DensNet121 | 22 | 11 | 33 | 102 |
Xception | 19 | 6 | 25 | 77 |
RepLKNet | 96 | 6 | 120 | 373 |
Let-Net(ours) | 3 | 3 | 6 | 17 |
Training(s) | Prediction(s) | Total(s) | Single batch time(ms) | |
VGG16 | 13 | 7 | 20 | 48 |
VGG19 | 15 | 8 | 23 | 55 |
Resnet50V2 | 14 | 9 | 23 | 54 |
InceptionV3 | 20 | 12 | 32 | 77 |
DensNet121 | 26 | 15 | 41 | 97 |
Xception | 25 | 7 | 32 | 78 |
RepLKNet | 126 | 32 | 158 | 379 |
Let-Net(ours) | 4 | 3 | 7 | 18 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 MDPI (Basel, Switzerland) unless otherwise stated