Beyond Human Forgeries: An Investigation into Detecting Diffusion-Generated Handwriting

Carriere G, Nikolaidou K, Kordon FJ, Mayr M, Seuret M, Christlein V (2023)

Publication Language: English

Publication Type: Conference contribution

Publication year: 2023

Publisher: Springer

Series: Lecture Notes in Computer Science

City/Town: Cham

Book Volume: 14193

Pages Range: 5-19

Conference Proceedings Title: Document Analysis and Recognition – ICDAR 2023 Workshops

Event location: San José, CA

ISBN: 9783031414978

DOI: 10.1007/978-3-031-41498-5_1

Abstract

Methods for detecting forged handwriting are usually based on the assumption that the forged handwriting is produced by humans. Authentic-looking handwriting, however, can also be produced synthetically. Diffusion-based generative models have recently gained popularity as they produce striking natural images and are also able to realistically mimic a person’s handwriting. It is, therefore, reasonable to assume that these models will be used to forge handwriting in the near future, adding a new layer to handwriting forgery detection. We show for the first time that the identification of synthetic handwritten data is possible by a small Convolutional Neural Network (ResNet18) reaching accuracies of 90%. We further investigate the existence of distinct discriminative features in synthetic handwriting data produced by latent diffusion models that could be exploited to build stronger detection methods. Our experiments indicate that the strongest discriminative features do not come from generation artifacts, letter shapes, or the generative model’s architecture, but instead originate from real-world artifacts in genuine handwriting that are not reproduced by generative methods.

Authors with CRIS profile

Involved external institutions

Luleå University of Technology Sweden (SE)

How to cite

APA:

Carriere, G., Nikolaidou, K., Kordon, F.J., Mayr, M., Seuret, M., & Christlein, V. (2023). Beyond Human Forgeries: An Investigation into Detecting Diffusion-Generated Handwriting. In Mickael Coustaty, Alicia Fornés (Eds.), Document Analysis and Recognition – ICDAR 2023 Workshops (pp. 5-19). San José, CA, US: Cham: Springer.

MLA:

Carriere, Guillaume, et al. "Beyond Human Forgeries: An Investigation into Detecting Diffusion-Generated Handwriting." Proceedings of the ICDAR 2023: Document Analysis and Recognition, San José, CA Ed. Mickael Coustaty, Alicia Fornés, Cham: Springer, 2023. 5-19.

BibTeX: Download