↑
logo

Evaluating OCR on the Community Memory Corpus

Posted by ingve |2 hours ago |1 comments

treetalker 24 minutes ago

Is there a website of record to stay abreast of the state of the art in OCR? It seems like it changes frequently.

Last I heard, the native MacOS models were super good for both text and handwriting on consumer hardware and in moderately sized batches.