A Prototypical Signature Approach for Writer-Independent Offline Signature Verification

2026-06-16

AI Summary

This work addresses the challenge in offline handwritten signature verification posed by the scarcity of genuine forgery samples, which often leads existing methods to use random signatures from other individuals as negative examples—resulting in data redundancy, limited diversity, and high computational costs. To overcome these limitations, the authors propose a prototype-based, data-driven strategy that constructs compact and non-identifiable signature feature summaries to generate informative and diverse negative samples. The approach is backbone-agnostic and can be integrated with a linear SVM in place of conventional RBF models. Evaluated across multiple deep architectures, the method significantly improves detection performance against highly skilled forgeries while reducing training complexity and enhancing scalability.

Abstract

Offline handwritten signature verification aims to distinguish genuine from forged signatures using static images. Since real forgeries are rarely available, negative samples are usually randomly drawn from genuine signatures of other users to create training data. However, this random selection often lacks diversity, increases redundancy, and escalates computational cost, leading to inefficient training. We propose a data-driven strategy to generate diverse, informative negative samples using prototypical signatures, which are compact, non-identifiable summaries of genuine signature features. Based on the experiments results, we conclude that (i) prototypical signatures yield more informative negative samples, improving the detection of skilled forgeries; (ii) the proposed approach is backbone-agnostic, showing robustness across architectures; and (iii) when combined with a primal-form linear SVM, it serves as an alternative to RBF-based models while significantly improving scalability and computational efficiency. Implementation of the method is available at https://github.com/kdmoura/proto_hsv.

Problem

• offline signature verification
• negative sample selection
• signature forgery detection
• data diversity
• computational efficiency

Innovation

• prototypical signatures
• offline signature verification
• negative sample generation
• writer-independent
• linear SVM