Abstract | ||
---|---|---|
Deep learning-based domain-invariant feature learning methods are advancing in near-infrared and visible (NIR-VIS) heterogeneous face recognition. However, these methods are prone to overfitting due to the large intra-class variation and the lack of NIR images for training. In this paper, we introduce Domain-Based Label Face (DBLFace), a learning approach based on the assumption that a subject is not represented by a single label but by a set of labels. Each label represents images of a specific domain. In particular, a set of two labels per subject, one for the NIR images and one for the VIS images, are used for training a NIR-VIS face recognition model. The classification of images into different domains reduces the intra-class variation and lessens the negative impact of data imbalance in training. To train a network with sets of labels, we introduce a domain-based angular margin loss and a maximum angular loss to maintain the inter-class discrepancy and to enforce the close relationship of labels in a set. Quantitative experiments confirm that DBLFace significantly improves the rank-1 identification rate by 6.7% on the EDGE20 dataset and achieves state-of-the-art performance on the CASIA NIR-VIS 2.0 dataset. |
Year | DOI | Venue |
---|---|---|
2020 | 10.1109/IJCB48548.2020.9304884 | 2020 IEEE International Joint Conference on Biometrics (IJCB) |
Keywords | DocType | ISSN |
DBLFace,NIR images,VIS images,domain-based angular margin loss,NIR-VIS heterogeneous face recognition,deep learning,maximum angular loss,CASIA NIR-VIS 2.0 dataset,near-infrared and visible heterogeneous face recognition,domain-based label face,image classification,EDGE20 dataset,image representation,domain-invariant feature learning | Conference | 2474-9680 |
ISBN | Citations | PageRank |
978-1-7281-9187-4 | 0 | 0.34 |
References | Authors | |
0 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ha Le | 1 | 2 | 1.06 |
Ioannis A. Kakadiaris | 2 | 1910 | 203.66 |