search

yizhiwang96 / deepvecfont

[SIGGRAPH Asia 2021] DeepVecFont: Synthesizing High-quality Vector Fonts via Dual-modality Learning
MIT License
182 stars 31 forks source link

Predict on incomplete glyph set #32

Open matheusrhoden opened 1 year ago

matheusrhoden commented 1 year ago

Hi!

Is it possible to run the model against an incomplete glyph set?

For example, drawing a few glyphs myself and then having DVF generate the rest of the alphabet.

Thanks!

yizhiwang96 commented 1 year ago

Yes, it is exactly the 'few-shot generation' feature demonstrated in our paper. The reference number was 4 in our paper, and theoretically it can be any number, e.g., the number of your drawn glyphs.

matheusrhoden commented 1 year ago

@yizhiwang96 Got it, thanks. Just to clarify, though, do I have to (re)train the model in order to use different reference glyphs? Apparently the modality fusion checkpoint shape has to match the number of reference glyphs