Home

Awesome

Latent HSJA

arXiv

Unrestricted Black-box Adversarial Attack Using GAN with Limited Queries

<img src="./resources/main.jpg" width="90%">

Authors

Abstract

Adversarial examples are inputs intentionally generated for fooling a deep neural network. Recent studies have proposed unrestricted adversarial attacks that are not norm-constrained. However, the previous unrestricted attack methods still have limitations to fool real-world applications in a black-box setting. In this paper, we present a novel method for generating unrestricted adversarial examples using GAN where an attacker can only access the top-1 final decision of a classification model. Our method, Latent-HSJA, efficiently leverages the advantages of a decision-based attack in the latent space and successfully manipulates the latent vectors for fooling the classification model.

Demonstration

<img src="./resources/showcase.jpg" width="70%">

Source Codes

Datasets

1. Celeb-HQ Facial Identity Recognition Dataset

<pre> <b>Dataset/</b> <b>train/</b> identity 1/ identity 2/ ... <b>test/</b> identity 1/ identity 2/ ... </pre>

2. Celeb-HQ Face Gender Recognition Dataset

<pre> <b>Dataset/</b> <b>train/</b> male/ female/ <b>test/</b> male/ female/ </pre>

Classification Models to Attack

Identity recognitionGender recognition
MNasNet1.078.35% (code | download)98.38% (code | download)
DenseNet12186.42% (code | download)98.15% (code | download)
ResNet1887.82% (code | download)98.55% (code | download)
ResNet10187.98% (code | download)98.05% (code | download)

Generate Datasets for Experiments with Encoding Networks

SIMLPIPSConsistency acc.Correctly consistency acc.
Identity Recognition0.64650.150470.78%67.00%(code | dataset)
Gender Recognition0.63550.162598.30%96.80%(code | dataset)

Training the Original Classification Models

Citation

If this work can be useful for your research, please cite our paper:

<pre> @inproceedings{na2022unrestricted, title={Unrestricted Black-Box Adversarial Attack Using GAN with Limited Queries}, author={Na, Dongbin and Ji, Sangwoo and Kim, Jong}, booktitle={European Conference on Computer Vision}, pages={467--482}, year={2022}, organization={Springer} } </pre>