FeatureLens: A Highly Generalizable and Interpretable Framework for Detecting Adversarial Examples Based on Image Features
By: Zhigang Yang , Yuan Liu , Jiawei Zhang and more
Potential Business Impact:
Finds fake images that trick computers.
Although the remarkable performance of deep neural networks (DNNs) in image classification, their vulnerability to adversarial attacks remains a critical challenge. Most existing detection methods rely on complex and poorly interpretable architectures, which compromise interpretability and generalization. To address this, we propose FeatureLens, a lightweight framework that acts as a lens to scrutinize anomalies in image features. Comprising an Image Feature Extractor (IFE) and shallow classifiers (e.g., SVM, MLP, or XGBoost) with model sizes ranging from 1,000 to 30,000 parameters, FeatureLens achieves high detection accuracy ranging from 97.8% to 99.75% in closed-set evaluation and 86.17% to 99.6% in generalization evaluation across FGSM, PGD, CW, and DAmageNet attacks, using only 51 dimensional features. By combining strong detection performance with excellent generalization, interpretability, and computational efficiency, FeatureLens offers a practical pathway toward transparent and effective adversarial defense.
Similar Papers
SIFT-Graph: Benchmarking Multimodal Defense Against Image Adversarial Attacks With Robust Feature Graph
CV and Pattern Recognition
Makes AI see images better, even when fooled.
SliceLens: Fine-Grained and Grounded Error Slice Discovery for Multi-Instance Vision Tasks
CV and Pattern Recognition
Finds computer vision mistakes in tricky pictures.
Are We Learning the Right Features? A Framework for Evaluating DL-Based Software Vulnerability Detection Solutions
Software Engineering
Makes computer code checkers more trustworthy.