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
1.47k stars 422 forks source link

Post-processing in this test case is so slow #36

Closed Gingersna closed 4 years ago

Gingersna commented 4 years ago

https://github.com/DevashishPrasad/CascadeTabNet/blob/7147b41edb6a7b75bfcb498921b2c7b678f391a6/Table%20Structure%20Recognition/main.py#L43-L53 test image:https://raw.githubusercontent.com/cndplab-founder/ICDAR2019_cTDaR/master/test/TRACKB2/cTDaR_t10080.jpg

kshitijkapadni commented 4 years ago

Post-processing for borderless tables depends on the cell mask given by the model. If the model doesn't give accurate results we are trying to make it accurate in post-processing. This may be the case which is making it slow.

Also, our research was focused on getting accuracy and we haven't focused on the time complexity.