Deep Perceptual Mapping for Thermal to Visible Face Recognition

TL;DR

This paper introduces a deep neural network approach to significantly improve thermal-to-visible face recognition by learning a non-linear mapping that bridges the large modality gap, enhancing identification accuracy in challenging conditions.

Contribution

The paper presents a novel deep perceptual mapping method that effectively models the non-linear relationship between thermal and visible face images, substantially improving recognition performance.

Findings

Over 10% improvement in Rank-1 identification accuracy.

Reduced modality gap performance drop by more than 40%.

Substantive performance gains on a difficult dataset.

Abstract

Cross modal face matching between the thermal and visible spectrum is a much de- sired capability for night-time surveillance and security applications. Due to a very large modality gap, thermal-to-visible face recognition is one of the most challenging face matching problem. In this paper, we present an approach to bridge this modality gap by a significant margin. Our approach captures the highly non-linear relationship be- tween the two modalities by using a deep neural network. Our model attempts to learn a non-linear mapping from visible to thermal spectrum while preserving the identity in- formation. We show substantive performance improvement on a difficult thermal-visible face dataset. The presented approach improves the state-of-the-art by more than 10% in terms of Rank-1 identification and bridge the drop in performance due to the modality gap by more than 40%.