Reproduce “Multi-Content GAN for Few-Shot Font Style Transfer” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Samaneh Azadi, Matthew Fisher, Vladimir Kim, Zhaowen Wang, Eli Shechtman, Trevor Darrell
MC-GAN generates missing decorated capital letters by separating glyph shape prediction from ornamentation, then training both stages together. GlyphNet learns shape relationships from 10,000 Latin fonts, while OrnaNet decorates predicted shapes using observed examples. In a study of 33 downloaded fonts, 11 participants preferred its sets over an adapted comparison method 80.0% of the time under that protocol.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.