4.8 Article

DVG-Face: Dual Variational Generation for Heterogeneous Face Recognition

出版社

IEEE COMPUTER SOC
DOI: 10.1109/TPAMI.2021.3052549

关键词

Face recognition; Learning systems; Databases; Generators; Gallium nitride; Image recognition; Training; Heterogeneous face recognition; cross-domain; dual generation; contrastive learning

资金

  1. Beijing Natural Science Foundation [JQ18017]
  2. National Natural Science Foundation of China [61721004, U20A20223]
  3. Youth Innovation PromotionAssociation CAS [Y201929]

向作者/读者索取更多资源

In this paper, a novel method called DVG-Face is proposed to address the challenges in heterogeneous face recognition (HFR) through dual generation and contrastive learning mechanism. The method achieves superior performances on multiple HFR tasks, demonstrating its effectiveness.
Heterogeneous face recognition (HFR) refers to matching cross-domain faces and plays a crucial role in public security. Nevertheless, HFR is confronted with challenges from large domain discrepancy and insufficient heterogeneous data. In this paper, we formulate HFR as a dual generation problem, and tackle it via a novel dual variational generation (DVG-Face) framework. Specifically, a dual variational generator is elaborately designed to learn the joint distribution of paired heterogeneous images. However, the small-scale paired heterogeneous training data may limit the identity diversity of sampling. In order to break through the limitation, we propose to integrate abundant identity information of large-scale visible data into the joint distribution. Furthermore, a pairwise identity preserving loss is imposed on the generated paired heterogeneous images to ensure their identity consistency. As a consequence, massive new diverse paired heterogeneous images with the same identity can be generated from noises. The identity consistency and identity diversity properties allow us to employ these generated images to train the HFR network via a contrastive learning mechanism, yielding both domain-invariant and discriminative embedding features. Concretely, the generated paired heterogeneous images are regarded as positive pairs, and the images obtained from different samplings are considered as negative pairs. Our method achieves superior performances over state-of-the-art methods on seven challenging databases belonging to five HFR tasks, including NIR-VIS, Sketch-Photo, Profile-Frontal Photo, Thermal-VIS, and ID-Camera.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.8
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据