CTRNet
CTRNet copied to clipboard
This repository is the implementation of "Don't Forget Me: Accurate Background Recovery for Text Removal via Modeling Local-Global Context".
There is a mask image needed when fetch batch in dataset.py. Is mask generated by original image minus gt image?
awsome work~ Could you provide the trained model on examination paper images? thk~
Does this model also require a handwritten text detection model to generate mask?
Hello! Are the pretrained weights provided trained on SCUT-EnsText or SCUT-Syn datasets?