Not Afraid of the Dark: NIR-VIS Face Recognition via Cross-spectral Hallucination and Low-rank Embedding

TL;DR

This paper introduces a novel approach combining cross-spectral hallucination and low-rank embedding to enable near-infrared to visible spectrum face recognition without retraining existing deep models, achieving state-of-the-art results.

Contribution

It proposes a method to adapt existing VIS face recognition models for NIR images using hallucination and low-rank embedding, avoiding retraining.

Findings

Achieves state-of-the-art accuracy on CASIA NIR-VIS v2.0 benchmark.

Effective cross-spectral recognition without retraining deep models.

Combining hallucination and low-rank embedding yields significant improvements.

Abstract

Surveillance cameras today often capture NIR (near infrared) images in low-light environments. However, most face datasets accessible for training and verification are only collected in the VIS (visible light) spectrum. It remains a challenging problem to match NIR to VIS face images due to the different light spectrum. Recently, breakthroughs have been made for VIS face recognition by applying deep learning on a huge amount of labeled VIS face samples. The same deep learning approach cannot be simply applied to NIR face recognition for two main reasons: First, much limited NIR face images are available for training compared to the VIS spectrum. Second, face galleries to be matched are mostly available only in the VIS spectrum. In this paper, we propose an approach to extend the deep learning breakthrough for VIS face recognition to the NIR spectrum, without retraining the underlying deep models that see only VIS faces. Our approach consists of two core components, cross-spectral hallucination and low-rank embedding, to optimize respectively input and output of a VIS deep model for cross-spectral face recognition. Cross-spectral hallucination produces VIS faces from NIR images through a deep learning approach. Low-rank embedding restores a low-rank structure for faces deep features across both NIR and VIS spectrum. We observe that it is often equally effective to perform hallucination to input NIR images or low-rank embedding to output deep features for a VIS deep model for cross-spectral recognition. When hallucination and low-rank embedding are deployed together, we observe significant further improvement; we obtain state-of-the-art accuracy on the CASIA NIR-VIS v2.0 benchmark, without the need at all to re-train the recognition system.