Cross-Domain Identification for Thermal-to-Visible Face Recognition

TL;DR

This paper introduces a novel domain adaptation framework for thermal-to-visible face recognition that reduces the need for image co-registration by using new feature mapping and loss functions, improving recognition across varied conditions.

Contribution

It proposes a new domain adaptation method combining feature mapping and specialized loss functions, enhancing thermal-to-visible face recognition without requiring precise image alignment.

Findings

Outperforms state-of-the-art models on challenging datasets

Effective across varying ranges, poses, and expressions

Shows promise for non-frontal thermal-to-visible recognition

Abstract

Recent advances in domain adaptation, especially those applied to heterogeneous facial recognition, typically rely upon restrictive Euclidean loss functions (e.g., $L_2$ norm) which perform best when images from two different domains (e.g., visible and thermal) are co-registered and temporally synchronized. This paper proposes a novel domain adaptation framework that combines a new feature mapping sub-network with existing deep feature models, which are based on modified network architectures (e.g., VGG16 or Resnet50). This framework is optimized by introducing new cross-domain identity and domain invariance loss functions for thermal-to-visible face recognition, which alleviates the requirement for precisely co-registered and synchronized imagery. We provide extensive analysis of both features and loss functions used, and compare the proposed domain adaptation framework with state-of-the-art feature based domain adaptation models on a difficult dataset containing facial imagery collected at varying ranges, poses, and expressions. Moreover, we analyze the viability of the proposed framework for more challenging tasks, such as non-frontal thermal-to-visible face recognition.