Detection of Adversarial Attacks in Robotic Perception

Deep Neural Networks (DNNs) achieve strong performance in semantic segmentation for robotic perception but remain vulnerable to adversarial attacks, threatening safety-critical applications. While robustness has been studied for image classification, semantic segmentation in robotic contexts requires specialized architectures and detection strategies.

Source: arXiv:2603.28594v1 - http://arxiv.org/abs/2603.28594v1 PDF: https://arxiv.org/pdf/2603.28594v1 Original Link: http://arxiv.org/abs/2603.28594v1