# Learning Features for Offline Handwritten Signature Verification using Deep Convolutional Neural Networks

## TL;DR

This paper introduces a deep learning approach using CNNs to improve offline handwritten signature verification, effectively distinguishing genuine signatures from skilled forgeries across multiple datasets with significant accuracy gains.

## Contribution

The study presents a novel CNN-based feature learning method incorporating skilled forgeries, achieving state-of-the-art results in signature verification without dataset-specific tuning.

## Key findings

- Achieved 1.72% EER on GPDS-160, outperforming previous methods.
- Features generalize well across different datasets without fine-tuning.
- Significant performance improvement over existing systems in multiple datasets.

## Abstract

Verifying the identity of a person using handwritten signatures is challenging in the presence of skilled forgeries, where a forger has access to a person's signature and deliberately attempt to imitate it. In offline (static) signature verification, the dynamic information of the signature writing process is lost, and it is difficult to design good feature extractors that can distinguish genuine signatures and skilled forgeries. This reflects in a relatively poor performance, with verification errors around 7% in the best systems in the literature. To address both the difficulty of obtaining good features, as well as improve system performance, we propose learning the representations from signature images, in a Writer-Independent format, using Convolutional Neural Networks. In particular, we propose a novel formulation of the problem that includes knowledge of skilled forgeries from a subset of users in the feature learning process, that aims to capture visual cues that distinguish genuine signatures and forgeries regardless of the user. Extensive experiments were conducted on four datasets: GPDS, MCYT, CEDAR and Brazilian PUC-PR datasets. On GPDS-160, we obtained a large improvement in state-of-the-art performance, achieving 1.72% Equal Error Rate, compared to 6.97% in the literature. We also verified that the features generalize beyond the GPDS dataset, surpassing the state-of-the-art performance in the other datasets, without requiring the representation to be fine-tuned to each particular dataset.