PDFLattice was inspired by the need for seamless, accurate Mandarin–English PDF translation that preserves document structure. The project leverages smart document parsing and AI to retain tables, images, and formatting in translated output. Through rapid prototyping, we learned the importance of layout-aware NLP and plan to extend support for more languages and complex layouts in future iterations.
Built With
- deepseekocr
- metanllb
- pypdf
- weasyprint
Log in or sign up for Devpost to join the conversation.