A style-embedding masked autoencoder and a semi-supervised training scheme let a handwritten-text diffusion model generate training images for a new, unlabeled handwriting dataset, improving downstream recognition error.
In: European Conf
1 Pith paper cite this work, alongside 27 external citations. Polarity classification is still indexing.
1
Pith paper citing it
27
external citations · OpenAlex
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Semi-Supervised Adaptation of Diffusion Models for Handwritten Text Generation
A style-embedding masked autoencoder and a semi-supervised training scheme let a handwritten-text diffusion model generate training images for a new, unlabeled handwriting dataset, improving downstream recognition error.