Saved in:
Bibliographic Details
Main Authors: Ahitoliev, Andrii, Berezin, Pavlo
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.27487
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913165394051072
author Ahitoliev, Andrii
Berezin, Pavlo
author_facet Ahitoliev, Andrii
Berezin, Pavlo
contents Handwritten text generation (HTG) conditioned on writer style has been widely studied for Latin scripts, but remains underexplored for low-resource and non-Latin writing systems, leaving open how well existing models generalise beyond the Latin domain. Cyrillic, particularly Ukrainian, lacks both large-scale writer-labeled datasets and empirical evidence of such generalisation. To address this gap, we construct a Ukrainian handwritten word dataset of 126,177 images from 308 writers using connected-component segmentation, quality filtering, and targeted oversampling of underrepresented Ukrainian characters. We retrain DiffusionPen, a MobileNetV2 triplet-loss style encoder with a CANINE-conditioned latent diffusion U-Net, on this dataset without architectural modification, testing direct transfer from Latin to Cyrillic. We evaluate cross-domain style transfer in three settings: cross-lingual transfer from IAM English samples, zero-shot transfer to an early 20th-century Ukrainian manuscript, and few-shot imitation of contemporary writers. The model produces legible, style-consistent word images, indicating that few-shot latent diffusion models generalize beyond the Latin-script domain. We release the dataset, trained models, and evaluation protocol as a reproducible benchmark for writer-aware Cyrillic HTG, providing a foundation for extending stylized HTG to other underrepresented writing systems.
format Preprint
id arxiv_https___arxiv_org_abs_2605_27487
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Diffusion-Based Ukrainian Handwritten Text Generation with Cross-Domain Style Transfer
Ahitoliev, Andrii
Berezin, Pavlo
Computer Vision and Pattern Recognition
Artificial Intelligence
I.4.9
Handwritten text generation (HTG) conditioned on writer style has been widely studied for Latin scripts, but remains underexplored for low-resource and non-Latin writing systems, leaving open how well existing models generalise beyond the Latin domain. Cyrillic, particularly Ukrainian, lacks both large-scale writer-labeled datasets and empirical evidence of such generalisation. To address this gap, we construct a Ukrainian handwritten word dataset of 126,177 images from 308 writers using connected-component segmentation, quality filtering, and targeted oversampling of underrepresented Ukrainian characters. We retrain DiffusionPen, a MobileNetV2 triplet-loss style encoder with a CANINE-conditioned latent diffusion U-Net, on this dataset without architectural modification, testing direct transfer from Latin to Cyrillic. We evaluate cross-domain style transfer in three settings: cross-lingual transfer from IAM English samples, zero-shot transfer to an early 20th-century Ukrainian manuscript, and few-shot imitation of contemporary writers. The model produces legible, style-consistent word images, indicating that few-shot latent diffusion models generalize beyond the Latin-script domain. We release the dataset, trained models, and evaluation protocol as a reproducible benchmark for writer-aware Cyrillic HTG, providing a foundation for extending stylized HTG to other underrepresented writing systems.
title Diffusion-Based Ukrainian Handwritten Text Generation with Cross-Domain Style Transfer
topic Computer Vision and Pattern Recognition
Artificial Intelligence
I.4.9
url https://arxiv.org/abs/2605.27487