Handwritten Signature Generation Using Denoising Diffusion Probabilistic Models with Auxiliary Classification Processes

Automatic signature verification has been widely studied for authentication purposes in real life, but limited data availability still poses a significant challenge. To address this issue, we propose a method with a denoising diffusion probabilistic model (DDPM) to generate artificial signatures tha...

ver descrição completa

Detalhes bibliográficos

Principais autores:	Dong-Jin Hong, Won-Du Chang, Eui-Young Cha
Formato:	Artigo
Idioma:	Inglês
Publicado em:	MDPI AG 2024-11-01
coleção:	Applied Sciences
Assuntos:	handwritten signature recognition image generation denoising diffusion probabilistic model auxiliary classification
Acesso em linha:	https://www.mdpi.com/2076-3417/14/22/10233
Tags:	Adicionar Tag Sem tags, seja o primeiro a adicionar uma tag!