craft-text-detector
craft-text-detector copied to clipboard
Packaged, Pytorch-based, easy to use, cross-platform version of the CRAFT text detector
## 🐛 Bug some prediction_result["boxes"] can have 0 size this raises an opencv exception when try to crop images by export_detected_regions ## To Reproduce Steps to reproduce the behavior: 1....
It would be great if we could add padding to bounding box. For example, in this image,  tesseract incorrectly parses 7 as /. Padding would most likely help.
Hello! Thanks, very good work! Easy to use! There is a small question. For educational purposes, I do the recognition of car numbers. If I take data from boxes or...