How does PaddleOCR performance compare to Tesseract?

[PPGitub]

How does PaddleOCR performance compare to Tesseract?

[MissPenguin]

Haven't compared with Tesseract recently. Here are some comparative evaluations made by external users for reference.

https://converter.app/blog/paddleocr-engine-example-and-benchmark
https://learnopencv.com/optical-character-recognition-using-paddleocr/

[Evezerest]

the OCR model used in the learnopencv tutorial is the Chinese-English one. Generally, for pure English scenarios, it is recommended to use the English model ('lang='en')
reference: https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.6/doc/doc_en/whl_en.md#21-use-by-code