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...

Full beskrivning

Bibliografiska uppgifter

Huvudupphov:	Dong-Jin Hong, Won-Du Chang, Eui-Young Cha
Materialtyp:	Artigo
Språk:	Inglês
Utgiven:	MDPI AG 2024-11-01
Serie:	Applied Sciences
Ämnen:	handwritten signature recognition image generation denoising diffusion probabilistic model auxiliary classification
Länkar:	https://www.mdpi.com/2076-3417/14/22/10233
Taggar:	Lägg till en tagg Inga taggar, Lägg till första taggen!