Open gsemyong opened 2 months ago
There is no option for OCR - this would add considerable megabytes to the codebase, something we can't afford for web runtime. OCR depends on a big and heavy set of libraries and also needs per-language training data files that we need to provide it access to. Possibly at some point we might consider a plugin for OCR, but this would have to be a separate project.
Haven't found any reference of OCR in the mupdf.js docs, but see that tesseract is mupdf's optional dependency. Is there an option do OCR using mupdf.js?