search

DevashishPrasad / CascadeTabNet

This repository contains the code and implementation details of the CascadeTabNet paper "CascadeTabNet: An approach for end to end table detection and structure recognition from image-based documents"
MIT License
1.5k stars 428 forks source link

detect results is smoething wrong #40

Open cqray1990 opened 4 years ago

cqray1990 commented 4 years ago

when i test an image in Examples files named cTDaR_t10120.jpg, it just detect one table,and cells picture you show is as follows:

and i use epoch_36 to test, besides the code run to " table: [ 323 208 1135 557] [Table status] : Processing table with lines" it can not exit,stop here all the time cells

nuaalyh commented 4 years ago

I encountered the same problem, have you solved it？

nuaalyh commented 4 years ago

I have solved my problem.My code is running on the server which does not support visual code: cv2.imshow( ). I commented the relevant code and then it worked.

DevashishPrasad commented 4 years ago

@kshitijkapadni if you can have a look at this issue.

kshitijkapadni commented 4 years ago

That is just the Printing issue and as @nuaalyh said it was solved by commenting out a few lines of visualization. Basically the coordinates are according to the whole page size and the image is cropped later only for the table and the boxes are being plotted on that image. As we require global coordinates i.e table and cell coordinates for the table with respect to the whole page, the XML generated is correct.

kkissmart commented 4 years ago

Hi I am not even able to get this,

Here is what I get,

https://github.com/DevashishPrasad/CascadeTabNet/issues/66

can anyone help me with this? Thanks so much