GenPalm: Contactless Palmprint Generation with Diffusion Models

TL;DR

This paper presents GenPalm, a diffusion model-based framework for generating realistic synthetic palmprints to overcome data scarcity in contactless palmprint recognition, improving recognition accuracy across multiple databases.

Contribution

It introduces a novel diffusion model approach for palmprint synthesis, providing a stable and effective alternative to GANs, and demonstrates its utility in enhancing recognition performance.

Findings

Generated palmprints are realistic and diverse.

Synthetic data improves recognition accuracy.

Method outperforms GAN-based approaches.

Abstract

The scarcity of large-scale palmprint databases poses a significant bottleneck to advancements in contactless palmprint recognition. To address this, researchers have turned to synthetic data generation. While Generative Adversarial Networks (GANs) have been widely used, they suffer from instability and mode collapse. Recently, diffusion probabilistic models have emerged as a promising alternative, offering stable training and better distribution coverage. This paper introduces a novel palmprint generation method using diffusion probabilistic models, develops an end-to-end framework for synthesizing multiple palm identities, and validates the realism and utility of the generated palmprints. Experimental results demonstrate the effectiveness of our approach in generating palmprint images which enhance contactless palmprint recognition performance across several test databases utilizing challenging cross-database and time-separated evaluation protocols.