Segue: Side-information Guided Generative Unlearnable Examples for Facial Privacy Protection in Real World

This project is a protective noise generation algorithm, which mainly solves the problem of face information being learned by unauthorized neural networks.

run

./run.sh

run.sh 分三步运行

第一步：训练噪声

python noise_train.py $method $model $log $log_path

第二步：生成不可学习数据集

python mknoisedata.py $method $model $log $log_path

第三步：代理模型在不可学习数据集上训练

python train_model.py $method $model $log $log_path