Online signature verification improves using new math and AI model
Online Signature Verification Using Augmented Path Signature and T-Mamba
Machine LearningComputer Vision and Pattern Recognition
Summary
Verifying handwritten signatures is important for security in places like banks. The researchers found that common AI methods sometimes miss important details or connections in signature data. They combined a special mathematical tool called augmented path signature, which captures shape and patterns well, with a new AI model called T-Mamba that looks at both short and long parts of the signature sequence. This approach worked better than past methods, especially when there were fewer example signatures to learn from. Their method was tested on three popular signature datasets and showed improved accuracy.
handwritten signature verificationaugmented path signaturetime seriesstate space modelstemporal convolutional networkssequence modellingerror ratesauthentication
Authors
Ruiling Li, Danyu Yang
Abstract
Handwritten signature verification is vital for personal authentication across commercial and financial applications. Although deep learning methods are widely adopted for online signature verification (OSV), they often struggle with capturing highly discriminative features and modelling long-range dependencies. To address these issues, we propose a novel framework that integrates the augmented path signature (APS) descriptor with the T-Mamba model. The APS descriptor first applies time and basepoint augmentations, then computes sliding-window path signatures. The path signature is a non-parametric feature map from rough path theory that effectively captures geometric structures and nonlinear inter-channel interactions. Inspired by the efficacy of state space models (SSMs) in sequence modelling, our T-Mamba model employs a hybrid design combining two temporal convolutional network (TCN) blocks with a time-scanning Mamba. This design enables the model to learn both local temporal patterns and global long-range dependencies, substantially improving verification accuracy. Our framework achieves state-of-the-art EERs on three public benchmark datasets (MCYT-100, SVC-2004 Task 2, DeepSignDB), validating its effectiveness and robustness, especially when the training data is limited. Our code is publicly available at https://github.com/DLRL04/OSV-using-APS-and-T-Mamba.