OCR-VQGAN, a discrete image encoder (tokenizer and detokenizer) for figure images in Paper2Fig100k dataset. Implementation of OCR Perceptual loss for clear text-within-image generation. Fork from VQGAN in CompVis/taming-transformers

GitHub repo scaffold for a “script-writing bot” that can (a) mimic archaic styles and (b) transliterate into historical/occult alphabets (e.g., Theban / “Alphabet of Honorius,” “Celestial” alphabet) while also supporting genuine ancient languages (Latin, Koine/Attic Greek, Coptic) when possible.

📝 Rewrite texts in the style of ancient works using historical scripts and languages, with style adaptation and transliteration tools for unique expressions.