xEdgeFace: Efficient Cross-Spectral Face Recognition for Edge Devices
Anjith George, Sebastien Marcel
TL;DR
xEdgeFace tackles cross-spectral face recognition with a focus on edge-device efficiency. It fine-tunes a pretrained EdgeFace RGB backbone by selectively adapting LayerNorm layers and early blocks, guided by a contrastive cross-modal loss and a self-distillation term to prevent forgetting the RGB task. The approach achieves strong HFR performance across multiple benchmarks while maintaining a light footprint (low GFLOPs and few parameters), outperforming or matching state-of-the-art methods. This enables practical cross-modal FR in constrained environments and demonstrates robust generalization with minimal paired data, making it suitable for real-world edge deployments.
Abstract
Heterogeneous Face Recognition (HFR) addresses the challenge of matching face images across different sensing modalities, such as thermal to visible or near-infrared to visible, expanding the applicability of face recognition systems in real-world, unconstrained environments. While recent HFR methods have shown promising results, many rely on computation-intensive architectures, limiting their practicality for deployment on resource-constrained edge devices. In this work, we present a lightweight yet effective HFR framework by adapting a hybrid CNN-Transformer architecture originally designed for face recognition. Our approach enables efficient end-to-end training with minimal paired heterogeneous data while preserving strong performance on standard RGB face recognition tasks. This makes it a compelling solution for both homogeneous and heterogeneous scenarios. Extensive experiments across multiple challenging HFR and face recognition benchmarks demonstrate that our method consistently outperforms state-of-the-art approaches while maintaining a low computational overhead.