GMoE-FTI：基于真实 MoE 机制的通用人脸模板反演方法
GMoE-FTI: A Universal Facial Template Inversion Method Adopting Genuine Mixture-of-Experts Mechanism
AI 导读
论文提出 GMoE-FTI，一种采用真实 MoE 机制的通用人脸模板反演（FTI）框架，可从主流系统的存储模板重建人脸，威胁生物特征隐私。该框架将模板提取器分为五类分布，训练五个映射专家与匹配模块，生成兼容预训练 StyleGAN3 的隐码，并做端到端联合微调。在 LFW 基准上，对 AdaFace、CosFace 等七种未见模板类型平均 TAR 提升 40%，FAPD 与 FAPC 各降 2%。
正文 · 原文
Abstract:Face recognition systems have been ubiquitously deployed across daily scenarios. To protect user privacy, mainstream systems store facial templates rather than raw face images for identity verification. However, facial template inversion (FTI) technologies can reconstruct faces from templates, which endangers biometric privacy. Existing FTI methods only fit single type of templates, failing to reconstruct identity-consistent faces from unseen types. To tackle this limitation, we survey mainstream template extractors which can be classified into five categories according the template distribution: unit-sphere, angle-linear, strict unit-sphere, cone-shaped, anuniform hypersphere. Inspired by this observation, this paper proposes GMoE-FTI, a universal FTI framework powered by a genuine MoE mechanism that differs from load-balanced multi-branch alternatives. It follows a three-stage training strategy: (1) independently train five mapping experts (MEs) and a matching module (MM); (2) conduct collaborative inference by weighting each ME's output with MM's output, so as to generate unified latent codes compatible with pre-trained StyleGAN3; (3) perform end-to-end joint fine-tuning over all modules to identity consistency and boost cross-expert coordination. Extensive evaluations on the LFW benchmark validate the superiority of GMoE-FTI against SOTA methods. Evaluated on seven prevailing unseen template types including AdaFace, CosFace, CurricularFace, ElasticFace, SphereFace2, SphereFace-R, and TransFace, GMoE-FTI achieves an average 40% TAR improvement, with average FAPD and FAPC each reduced by 2%, verifying its universal generalization across heterogeneous templates.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.33055 [cs.CV] |
| (or arXiv:2610.33055v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.33055 arXiv-issued DOI via DataCite |
Submission history
From: Zixuan Shen [view email]
[v1]
Mon, 19 Oct 2026 07:31:52 UTC (1,859 KB)
来源：arXiv cs.AI · arxiv.org