3 ms·
If you have the time then go for MLKit or any other OCR API, tesseract is pathetic for non-scanned/in the wild images, then put your parsers atop of the OCR out
by villgax 6y ago
If you have the time then go for MLKit or any other OCR API, tesseract is pathetic for non-scanned/in the wild images, then put your parsers atop of the OCR output.
If time is of the essence simply use AWS Textract & be done with its free tier.