Dual Variational Generation for Low-Shot Heterogeneous Face Recognition
This addresses the challenge of recognizing faces across different domains (e.g., sketches vs. photos) with limited data, offering a novel solution for biometric security applications.
The paper tackles the problem of Heterogeneous Face Recognition (HFR) by proposing a Dual Variational Generation (DVG) framework that generates paired heterogeneous images to reduce domain discrepancy, significantly improving state-of-the-art results on four databases.
Heterogeneous Face Recognition (HFR) is a challenging issue because of the large domain discrepancy and a lack of heterogeneous data. This paper considers HFR as a dual generation problem, and proposes a novel Dual Variational Generation (DVG) framework. It generates large-scale new paired heterogeneous images with the same identity from noise, for the sake of reducing the domain gap of HFR. Specifically, we first introduce a dual variational autoencoder to represent a joint distribution of paired heterogeneous images. Then, in order to ensure the identity consistency of the generated paired heterogeneous images, we impose a distribution alignment in the latent space and a pairwise identity preserving in the image space. Moreover, the HFR network reduces the domain discrepancy by constraining the pairwise feature distances between the generated paired heterogeneous images. Extensive experiments on four HFR databases show that our method can significantly improve state-of-the-art results. The related code is available at https://github.com/BradyFU/DVG.