FedPalm: A General Federated Learning Framework for Closed- and Open-Set Palmprint Verification

TL;DR

This work addresses privacy concerns in palmprint verification by proposing FedPalm, a federated learning framework that combines personalized closed-set texture-experts with a global open-set texture-expert, connected via a Textural Expert Interaction Module (TEIM). It also establishes a comprehensive benchmark for FL-based palmprint recognition, defining closed-set and open-set tasks under strict privacy constraints. Empirical results on IITD, Tongji, and PolyU show FedPalm achieving superior performance over both generic and personalized FL baselines, with TEIM providing robust cross-texture integration. The approach offers a practical path toward privacy-preserving, generalizable palmprint verification suitable for real-world, large-scale deployments.

Abstract

Current deep learning (DL)-based palmprint verification models rely on centralized training with large datasets, which raises significant privacy concerns due to biometric data's sensitive and immutable nature. Federated learning~(FL), a privacy-preserving distributed learning paradigm, offers a compelling alternative by enabling collaborative model training without the need for data sharing. However, FL-based palmprint verification faces critical challenges, including data heterogeneity from diverse identities and the absence of standardized evaluation benchmarks. This paper addresses these gaps by establishing a comprehensive benchmark for FL-based palmprint verification, which explicitly defines and evaluates two practical scenarios: closed-set and open-set verification. We propose FedPalm, a unified FL framework that balances local adaptability with global generalization. Each client trains a personalized textural expert tailored to local data and collaboratively contributes to a shared global textural expert for extracting generalized features. To further enhance verification performance, we introduce a Textural Expert Interaction Module that dynamically routes textural features among experts to generate refined side textural features. Learnable parameters are employed to model relationships between original and side features, fostering cross-texture-expert interaction and improving feature discrimination. Extensive experiments validate the effectiveness of FedPalm, demonstrating robust performance across both scenarios and providing a promising foundation for advancing FL-based palmprint verification research.

Figures (2)

• Figure 1: Overview of the proposed FedPalm. In the proposed TEIM, $N$ local textural experts and one shared global textural expert collaboratively extract features from the input image. TEIM routes features from other experts to generate enhanced features for each expert, which are then used for subsequent embedding.
• Figure 2: The ROC curves of different methods. (a)-(c) denote the Tongji, IITD, and PolyU results, respectively.