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
1.47k
stars
422
forks
source link
Table Structure recognition - Post processing questions #84
Hi @DevashishPrasad Amazing paper and thank you for providing source code.
I have trained my cell and table detection model using detectron2 and i used PubTabNet dataset. Cells and Table detection is working fine, now i'm trying to generate xml file from model prediction. I tried using your post processing source code, but its not able to generate xml correctly. Because in my dataset, input image is cropped image not full image, so my dim will be not matching with yours.
I started to change your source code, but i was able to understand how rows are getting generated. But can you explain how your merging the rows/columns.
I'm getting below output(i have commented column code).
Hi @DevashishPrasad Amazing paper and thank you for providing source code.
I have trained my cell and table detection model using detectron2 and i used PubTabNet dataset. Cells and Table detection is working fine, now i'm trying to generate xml file from model prediction. I tried using your post processing source code, but its not able to generate xml correctly. Because in my dataset, input image is cropped image not full image, so my dim will be not matching with yours.
I started to change your source code, but i was able to understand how rows are getting generated. But can you explain how your merging the rows/columns.
I'm getting below output(i have commented column code).
Thank you for your help.