The Surya OCR toolkit is a suite of models for OCR in 90+ languages, line detection, layout analysis, table recognition, and reading order detection. Surya has 13k Github Stars, is used by hundreds of organizations, and benchmarks well against cloud services.
Here's what Surya can do.
Surya handles OCR in over 90 languages. See benchmarks.
Surya does state of the art line detection in any language. See benchmarks.
Surya does state of the art layout analysis on a range of documents. See benchmarks.
Surya does state of the art reading order detection on a range of documents. See benchmarks.
Along with tabled, surya does state of the art table detection and recognition on a range of documents. See benchmarks.
