8 datasets found

The comma2k19 dataset is used to evaluate the robustness of lane detection models under physical-world adversarial attacks in autonomous driving.

The UAE-RS dataset is a benchmark for black-box adversarial attacks in remote sensing. It contains adversarial examples generated using the Mixup-Attack and Mixcut-Attack methods.

The NIPS2017 dataset is a collection of images used for adversarial attacks.

Sparse adversarial attacks fool deep neural networks (DNNs) through minimal pixel perturbations, often regularized by the ℓ0 norm. Recent efforts have replaced this norm with a...

FaceCat: Enhancing Face Recognition Security with a Unified Generative Model Framework

Generating adversarial examples is the art of creating a noise that is added to an input signal of a classifying neural network, and thus changing the network’s classification,...

The dataset used in this paper is MNIST, SVHN, CIFAR10, CIFAR100, and ImageNet.

The authors used the ImageNet-Compatible and CIFAR-10 datasets for targeted attack experiments.