How to convert the predicted bounding boxes to actual predicted text file.

clovaai / CRAFT-pytorch

Official implementation of Character Region Awareness for Text Detection (CRAFT)

MIT License

3.12k stars 885 forks source link

How to convert the predicted bounding boxes to actual predicted text file. #97

Open SanatMoharir opened 4 years ago

SanatMoharir commented 4 years ago

Hi @YoungminBaek @ClovaAIAdmin, While testing the code on the image I've got the heatmap, bounding box.png and the locations of bounding boxes but I want the text file which consists of extracted text. Is there any way to do that?

Thanks in advance!!

RulinShao commented 4 years ago

same question

NguyenHongSon1103 commented 4 years ago

@SanatMoharir it can't because this CRAFT model is not for optical character recognition (OCR) but for text detection. If you want extract text, you can use some OCR model like https://github.com/clovaai/deep-text-recognition-benchmark with belong to @clovaai too or other model.