A White-Box False Positive Adversarial Attack Method on Contrastive Loss Based Offline Handwritten Signature Verification Models

TL;DR

This paper introduces a novel white-box adversarial attack method for offline handwritten signature verification models based on contrastive loss, utilizing style transfer and new loss functions to improve attack success while maintaining minimal perturbations.

Contribution

The paper presents a new false positive attack approach with two innovative loss functions, achieving state-of-the-art results in white-box attacks on signature verification models.

Findings

State-of-the-art attack success rate achieved

Effective style transfer in handwriting styles demonstrated

Minimal perturbations maintained in generated adversarial samples

Abstract

In this paper, we tackle the challenge of white-box false positive adversarial attacks on contrastive loss based offline handwritten signature verification models. We propose a novel attack method that treats the attack as a style transfer between closely related but distinct writing styles. To guide the generation of deceptive images, we introduce two new loss functions that enhance the attack success rate by perturbing the Euclidean distance between the embedding vectors of the original and synthesized samples, while ensuring minimal perturbations by reducing the difference between the generated image and the original image. Our method demonstrates state-of-the-art performance in white-box attacks on contrastive loss based offline handwritten signature verification models, as evidenced by our experiments. The key contributions of this paper include a novel false positive attack method, two new loss functions, effective style transfer in handwriting styles, and superior performance in white-box false positive attacks compared to other white-box attack methods.