search

lcy0604 / CTRNet

This repository is the implementation of "Don't Forget Me: Accurate Background Recovery for Text Removal via Modeling Local-Global Context".
MIT License
77 stars 7 forks source link

SCUT-syn dataset doesn't include bounding box annotation. How do you generate soft mask? #2

Open DerekRay opened 1 year ago

lcy0604 commented 1 year ago

you can use a scene text detection model, or directly use (img - lbl) to generate the mask

Derek Ray @.***> 于2022年9月13日周二 20:39写道：

— Reply to this email directly, view it on GitHub https://github.com/lcy0604/CTRNet/issues/2, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJ7RD3UE33NZ4RLKNEV3OVTV6BYZHANCNFSM6AAAAAAQLM3FTE . You are receiving this because you are subscribed to this thread.Message ID: @.***>