alimama-creative/Self-Supervised-Text-Erasing

Self-supervised Text Erasing Model (STE)

Paper: https://arxiv.org/abs/2204.12743
Project Page: https://github.com/alimama-creative/Self-supervised-Text-Erasing

Description

The checkpoints are trained from the posterErase dataset. There are two versions with different training mechanism.

Self-supervised Text Trasing (ste_best_net_G.pth): To use it, please download from this page, and put it under './checkpoints/erasenet/ste/best_net_G.pth'

Finetuning after STE (ft_best_net_G.pth): To use it, please download from this page, and put it under './checkpoints/erasenet/ste/best_net_G.pth'

Usage

First, download the github project and install the python package.

git clone https://github.com/alimama-creative/Self-supervised-Text-Erasing.git
pip install -r requirements.txt

Then, follow the command line provied in the github to run the inference code.

python test.py --dataset_mode items --dataroot ./examples/poster --model erasenet --name ft --which_epoch best  # inferece with the ste model on poster 

python test.py --dataset_mode items --dataroot ./examples/poster --model erasenet --name ste --which_epoch best # inferece with the finetuned model model on poster