AdvSPADE: Realistic Unrestricted Attacks for Semantic Segmentation
arXiv:1910.02354
Abstract
Due to the inherent robustness of segmentation models, traditional norm-bounded attack methods show limited effect on such type of models. In this paper, we focus on generating unrestricted adversarial examples for semantic segmentation models. We demonstrate a simple and effective method to generate unrestricted adversarial examples using conditional generative adversarial networks (CGAN) without any hand-crafted metric. The naïve implementation of CGAN, however, yields inferior image quality and low attack success rate. Instead, we leverage the SPADE (Spatially-adaptive denormalization) structure with an additional loss item to generate effective adversarial attacks in a single step. We validate our approach on the popular Cityscapes and ADE20K datasets, and demonstrate that our synthetic adversarial examples are not only realistic, but also improve the attack success rate by up to 41.0\% compared with the state of the art adversarial attack methods including PGD.
References in corpus (23)
- Adam: A Method for Stochastic Optimization
- Distilling the Knowledge in a Neural Network
- Conditional Generative Adversarial Nets
- Rethinking Atrous Convolution for Semantic Image Segmentation
- Intriguing properties of neural networks
- GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
- Spectral Normalization for Generative Adversarial Networks
- Generalised Dice overlap as a deep learning loss function for highly unbalanced segmentations
- Conditional Image Synthesis With Auxiliary Classifier GANs
- Feature Squeezing: Detecting Adversarial Examples in Deep Neural Networks
- Transferability in Machine Learning: from Phenomena to Black-Box Attacks using Adversarial Samples
- Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examples
- Understanding Adversarial Training: Increasing Local Stability of Neural Nets through Robust Optimization
- Adversarial Machine Learning at Scale
- Detecting Adversarial Samples from Artifacts
- Exploring the Landscape of Spatial Robustness
- Houdini: Fooling Deep Structured Prediction Models
- Constructing Unrestricted Adversarial Examples with Generative Models
- Extending Defensive Distillation
- Learning Hierarchical Semantic Image Manipulation through Structured Representations
- Context-Aware Synthesis and Placement of Object Instances
- 3D-Aware Scene Manipulation via Inverse Graphics
- Adaptive Generation of Unrestricted Adversarial Inputs