SynSig2Vec: Learning Representations from Synthetic Dynamic Signatures for Real-world Verification

TL;DR

SynSig2Vec introduces a novel approach to dynamic signature verification by synthesizing signatures with varying distortions and training a ranking model, achieving state-of-the-art results without using skilled forgeries.

Contribution

The paper presents a neuromotor inspired synthesis method and a ranking-based learning framework that does not require skilled forgeries for training, outperforming existing methods.

Findings

Surpasses state-of-the-art on two benchmarks

Does not require skilled forgeries for training

Effective synthesis of signatures with different distortions

Abstract

An open research problem in automatic signature verification is the skilled forgery attacks. However, the skilled forgeries are very difficult to acquire for representation learning. To tackle this issue, this paper proposes to learn dynamic signature representations through ranking synthesized signatures. First, a neuromotor inspired signature synthesis method is proposed to synthesize signatures with different distortion levels for any template signature. Then, given the templates, we construct a lightweight one-dimensional convolutional network to learn to rank the synthesized samples, and directly optimize the average precision of the ranking to exploit relative and fine-grained signature similarities. Finally, after training, fixed-length representations can be extracted from dynamic signatures of variable lengths for verification. One highlight of our method is that it requires neither skilled nor random forgeries for training, yet it surpasses the state-of-the-art by a large margin on two public benchmarks.