search

Ucas-HaoranWei / GOT-OCR2.0

Official code implementation of General OCR Theory: Towards OCR-2.0 via a Unified End-to-end Model
5.66k stars 469 forks source link

Could you provide the bounding boxes of the text found please #66

Open Whizboy-Arnold opened 1 month ago

Whizboy-Arnold commented 1 month ago

I would like to have a bounding box of text found in plain text mode @Ucas-HaoranWei

Dineshkumar-Anandan-ZS0367 commented 1 month ago

How to get the bounding box for the extracted data from the given image

marcelMaier commented 1 month ago

How to get the bounding box for the extracted data from the given image

not possible right now, this issue is requesting this feature.

Ucas-HaoranWei commented 1 month ago

How to get the bounding box for the extracted data from the given image

not possible right now, this issue is requesting this feature. Hi, this feature can be achieved easily via fine-tuning the model： The data： https://www.modelscope.cn/datasets/milter/700k-layout-data The results： 39889b46761db7da4d8c33cb87b7069

ande4485 commented 4 weeks ago

@Ucas-HaoranWei is it possible for you to share your fine tuned model ? Thank you.

ajmssc commented 1 week ago

@Ucas-HaoranWei could you share the code for fine-tuning with bounding boxes?