New framework enhances online signature verification with path signatures and T-Mamba

Researchers have developed a new framework for online signature verification that combines the augmented path signature (APS) descriptor with a T-Mamba model. The APS descriptor captures geometric structures and nonlinear interactions by computing sliding-window path signatures after time and basepoint augmentations. The T-Mamba model, a hybrid design incorporating temporal convolutional network blocks and a time-scanning Mamba, is capable of learning both local temporal patterns and global long-range dependencies. This integrated approach has demonstrated state-of-the-art performance on benchmark datasets, particularly in scenarios with limited training data. AI

IMPACT This research could improve the accuracy and robustness of signature verification systems, especially in low-data scenarios.

RANK_REASON This is a research paper detailing a new method and model for a specific task. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• Augmented Path Signature
• DeepSignDB
• MCYT-100
• State Space Models
• SVC-2004 Task 2
• Temporal Convolutional Network
• T-Mamba

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →