Full list on Google Scholar.
Learning font-style space using style-guided discriminator for few-shot font generation
Expert Systems With Applications 2024
A structured font-style latent space learned via a style-guided discriminator enables reliable few-shot font generation across diverse scripts.
Exploiting mixing regularization for truly unsupervised font synthesis
Pattern Recognition Letters 2023
Mixing regularization lets a GAN disentangle content from style without any label supervision, enabling truly unsupervised multi-domain font synthesis.
Real-time high quality font generation with Conditional Font GAN
Expert Systems With Applications 2022
A conditional GAN that generates high-quality fonts at practical runtime speeds, closing the gap between automated synthesis and human designer output.
FontNet: Closing the gap to font designer performance in font synthesis
AI for Content Creation Workshop, CVPR 2022
FontNet introduces triplet-loss style-space learning to push font synthesis quality toward professional designer standards.
Unpaired font family synthesis using conditional generative adversarial networks
Knowledge-Based Systems 2021
An unpaired conditional GAN that synthesizes full font families from partial examples, enabling typeface variation without paired training data.
SKFont: skeleton-driven Korean font generator with conditional deep adversarial networks
International Journal on Document Analysis and Recognition 2021
Using extracted glyph skeletons as structural guidance, SKFont generates Korean fonts conditioned on stroke-level features rather than pixel targets.
SkelGAN: A Font Image Skeletonization Method
Journal of Information Processing Systems 2021
SkelGAN uses an end-to-end GAN to extract clean stroke skeletons from character images, providing compact structural representations for downstream font synthesis.