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
1.46k stars 422 forks source link

Not accurate results #102

Open Ahmad-AlShalabi opened 3 years ago

Ahmad-AlShalabi commented 3 years ago

@DevashishPrasad Thanks for sharing your hard work in this project.

I followed the pipeline using Colab but I did not get the same results, and for Bordered tables, I did not get the table structure(cells), and in some cases incomplete table structure (missing columns).

Could you please tell us where is the missing point?

Thanks in advance!

this is the result for demo image:

image

intersection:

image

detected cells:

image

image

Another demo image:

image

image

image

image

VincentJousse commented 3 years ago

I also have the same problem, it’s strange that the Colab example doesn’t give the same result as the one presented on GitHub.

martian1231 commented 3 years ago

any update on this issue?

AvikantSrivastava commented 3 years ago

@DevashishPrasad Thanks for sharing your hard work in this project.

I followed the pipeline using Colab but I did not get the same results, and for Bordered tables, I did not get the table structure(cells), and in some cases incomplete table structure (missing columns).

Could you please tell us where is the missing point?

Thanks in advance!

this is the result for demo image:

image

intersection:

image

detected cells:

image

image

Another demo image:

image

image

image

image

How did you get the intersection output?

XueTuo commented 2 years ago

@DevashishPrasad Thanks for sharing your hard work in this project. I followed the pipeline using Colab but I did not get the same results, and for Bordered tables, I did not get the table structure(cells), and in some cases incomplete table structure (missing columns). Could you please tell us where is the missing point? Thanks in advance! this is the result for demo image: image intersection: image detected cells: image image Another demo image: image image image image

How did you get the intersection output?

first you can find borderFunc.py and activate line 54-55 ,while it show the intersection output