AEGIS: A Semantic GAN and Evidential Learning Frameworkfor Robust Adversarial Detection in Vision Sensors

Researchers introduced AEGIS, a semantic-aware adversarial detection framework for vision sensors, combining a SemantiGAN module and evidential deep learning to filter adversarial inputs and provide calibrated uncertainty estimates. On Tiny ImageNet, AEGIS achieved 92.1% AUROC, 90.2% AUPRC, and 90.7% accuracy, outperforming softmax-based detectors across six attack categories.

arXiv:2606.28416v1 Announce Type: new Abstract: Deep neural networks DNNs have shown outstanding performance in visual recognition tasks within vision sensor networks; however, they are still vulnerable to adversarial manipulations and imperceptible perturbations that can lead to erroneous predictions. To address that, this paper presents AEGIS, a semantic aware and uncertainty guided adversarial detection framework designed for robust image classification in vision sensors pipelines. At its core, a SemantiGAN module functions as a multi class semantic discriminator, identifying and filtering visually inconsistent adversarial inputs before they propagate further in the pipeline. For inputs that pass this stage, a stochastic augmentation process generates test time variations, from which handcrafted instability metrics FlipScore, Prediction Inconsistency, Layerwise Cosine Similarity early and mid layers , and Entropy are computed. These features are aggregated into a compact five dimensional vector and processed by an Evidential Deep Learning EDL classifier, which models output evidence using a Dirichlet distribution to yield both class predictions and calibrated uncertainty estimates. Evaluations on the Tiny ImageNet dataset across six categories clean, FGSM, PGD, patch based, functional, and geometric attacks demonstrate the effectiveness of AEGIS. The proposed framework achieves an AUROC of 92.1\%, an AUPRC of 90.2\%, and an accuracy of 90.7\%, outperforming conventional softmax-based detectors in terms of detection performance, robustness, interpretability, and uncertainty calibration.