This code is for the paper Smooth Adversarial Examples
The original code is built based on the TensorFlow. The attack method are implemented in attacks_SAE.py and attacks_tf_SAE.py
The Laplacian graph is generated by MATALAB codes. These codes are under folder 'matlabGenerate', which show how to generate graphs for the smooth attack.
The codes for magnification are under the folder 'magnify'.
We are trying to translate the MATALAB codes for generating Laplacian graph into python codes, and make the framework more unify. These codes are under folder 'SAE'. It is still under debuging.
-
Install dependencies. CleverHans
-
Run the experiments on ImageNet.
cd wholeExperiments
python inc_scw.py
-----------------------------------
Smooth Adversarial Examples
Copyright (C) 2020 Hanwei Zhang - Inria Rennes Bretagne Atlantique
Smooth Adversarial Examples is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.
Smooth Adversarial Examples is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
To receive a copy of the GNU General Public License, see <http://www.gnu.org/licenses/>.
-------------------------------------