Dual Variational Generation for Low-Shot Heterogeneous Face Recognition

25 Mar 2019  ·  Chaoyou Fu, Xiang Wu, Yibo Hu, Huaibo Huang, Ran He ·

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.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Face Verification BUAA-VisNir LightCNN-29 + DVG TAR @ FAR=0.001 97.3 # 1
TAR @ FAR=0.01 98.5 # 1
Face Verification CASIA NIR-VIS 2.0 LightCNN-29 + DVG TAR @ FAR=0.001 99.8 # 1
Face Verification IIIT-D Viewed Sketch LightCNN-29 + DVG TAR @ FAR=0.01 97.86 # 1
Face Verification Oulu-CASIA NIR-VIS LightCNN-29 + DVG TAR @ FAR=0.001 92.9 # 1
TAR @ FAR=0.01 98.5 # 1