Lightweight Cross-Spectral Face Recognition via Contrastive Alignment and Distillation

• URL: http://arxiv.org/abs/2605.04769v1
• Date: Wed, 06 May 2026 11:16:08 GMT
• Title: Lightweight Cross-Spectral Face Recognition via Contrastive Alignment and Distillation
• Abstract summary: Heterogeneous Face Recognition (HFR) aims at matching face images captured across different sensing modalities.<n>We introduce a lightweight yet effective HFR framework by adapting a hybrid CNN-Transformer model.<n>Our approach enables efficient end-to-end training with only a small amount of paired heterogeneous data.
• Score: 6.690084812573466
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Heterogeneous Face Recognition (HFR) aims at matching face images captured across different sensing modalities, such as thermal-to-visible or near-infrared-to-visible, enhancing the usability of face recognition systems in challenging real-world conditions. Although recent HFR methods have achieved significant improvements in performance, many rely on computationally expensive models, making them impractical for deployment on resource-limited edge devices. In this work, we introduce a lightweight yet effective HFR framework by adapting a hybrid CNN-Transformer model originally developed for RGB homogeneous face recognition. Our approach enables efficient end-to-end training with only a small amount of paired heterogeneous data, while still maintaining strong performance on standard RGB face recognition benchmarks. This makes it suitable for both homogeneous and heterogeneous settings. Comprehensive experiments on several challenging HFR and face recognition benchmarks show that our method achieves state-of-the-art or competitive performance while keeping computational requirements low.

Related papers

• SynThermFace: Amplifying Limited Paired Data for Visible-Thermal Face Recognition via Synthetic Data Generation [5.952563476824616]
  Cross-spectral face recognition improves performance in low-light, nighttime, and unconstrained conditions.<n>We propose SynThermFace, a framework that amplifies limited real visible-thermal supervision into larger paired adaptation datasets.
  arXiv  Detail & Related papers  (2026-09-09T15:15:04Z)
• xEdgeFace: Efficient Cross-Spectral Face Recognition for Edge Devices [4.910937238451485]
  Heterogeneous Face Recognition (HFR) addresses the challenge of matching face images across different sensing modalities.<n>We present a lightweight yet effective HFR framework by adapting a hybrid CNN-Transformer architecture.<n>Our approach enables efficient end-to-end training with minimal paired heterogeneous data while preserving strong performance on standard RGB face recognition tasks.
  arXiv  Detail & Related papers  (2025-04-28T10:03:11Z)
• Diffusion Prior Interpolation for Flexibility Real-World Face Super-Resolution [48.34173818491552]
  Diffusion Prior Interpolation (DPI) can balance consistency and diversity and can be seamlessly integrated into pre-trained models.<n>In extensive experiments conducted on synthetic and real datasets, DPI demonstrates superiority over SOTA FSR methods.
  arXiv  Detail & Related papers  (2024-12-21T09:28:44Z)
• Heterogeneous Face Recognition Using Domain Invariant Units [4.910937238451485]
  We leverage a pretrained face recognition model as a teacher network to learn domaininvariant network layers called Domain-Invariant Units (DIU) The proposed DIU can be trained effectively even with a limited amount of paired training data, in a contrastive distillation framework. This proposed approach has the potential to enhance pretrained models, making them more adaptable to a wider range of variations in data.
  arXiv  Detail & Related papers  (2024-04-22T16:58:37Z)
• Confidence-Aware RGB-D Face Recognition via Virtual Depth Synthesis [48.59382455101753]
  2D face recognition encounters challenges in unconstrained environments due to varying illumination, occlusion, and pose. Recent studies focus on RGB-D face recognition to improve robustness by incorporating depth information. In this work, we first construct a diverse depth dataset generated by 3D Morphable Models for depth model pre-training. Then, we propose a domain-independent pre-training framework that utilizes readily available pre-trained RGB and depth models to separately perform face recognition without needing additional paired data for retraining.
  arXiv  Detail & Related papers  (2024-03-11T09:12:24Z)
• Rethinking the Domain Gap in Near-infrared Face Recognition [65.7871950460781]
  Heterogeneous face recognition (HFR) involves the intricate task of matching face images across the visual domains of visible (VIS) and near-infrared (NIR) Much of the existing literature on HFR identifies the domain gap as a primary challenge and directs efforts towards bridging it at either the input or feature level. We observe that large neural networks, unlike their smaller counterparts, when pre-trained on large scale homogeneous VIS data, demonstrate exceptional zero-shot performance in HFR.
  arXiv  Detail & Related papers  (2023-12-01T14:43:28Z)
• Physically-Based Face Rendering for NIR-VIS Face Recognition [165.54414962403555]
  Near infrared (NIR) to Visible (VIS) face matching is challenging due to the significant domain gaps. We propose a novel method for paired NIR-VIS facial image generation. To facilitate the identity feature learning, we propose an IDentity-based Maximum Mean Discrepancy (ID-MMD) loss.
  arXiv  Detail & Related papers  (2022-11-11T18:48:16Z)
• Heterogeneous Face Recognition via Face Synthesis with Identity-Attribute Disentanglement [33.42679052386639]
  Heterogeneous Face Recognition (HFR) aims to match faces across different domains. We propose a new HFR method named Face Synthesis with Identity-Attribute Disentanglement (FSIAD) FSIAD decouples face images into identity-related representations and identity-unrelated representations (called attributes)
  arXiv  Detail & Related papers  (2022-06-10T03:01:33Z)
• Blind Face Restoration: Benchmark Datasets and a Baseline Model [63.053331687284064]
  Blind Face Restoration (BFR) aims to construct a high-quality (HQ) face image from its corresponding low-quality (LQ) input. We first synthesize two blind face restoration benchmark datasets called EDFace-Celeb-1M (BFR128) and EDFace-Celeb-150K (BFR512) State-of-the-art methods are benchmarked on them under five settings including blur, noise, low resolution, JPEG compression artifacts, and the combination of them (full degradation)
  arXiv  Detail & Related papers  (2022-06-08T06:34:24Z)
• Escaping Data Scarcity for High-Resolution Heterogeneous Face Hallucination [68.78903256687697]
  In Heterogeneous Face Recognition (HFR), the objective is to match faces across two different domains such as visible and thermal. Recent methods attempting to fill the gap via synthesis have achieved promising results, but their performance is still limited by the scarcity of paired training data. In this paper, we propose a new face hallucination paradigm for HFR, which not only enables data-efficient synthesis but also allows to scale up model training without breaking any privacy policy.
  arXiv  Detail & Related papers  (2022-03-30T20:44:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.