On Adversarial Robustness of Point Cloud Semantic Segmentation
This addresses a critical safety gap for applications like autonomous driving, but it is an incremental study as it systematically analyzes existing models without proposing new defenses.
The paper tackles the problem of adversarial robustness in 3D point cloud semantic segmentation (PCSS), finding that all tested models are vulnerable to attacks, with attacking point color being particularly effective.
Recent research efforts on 3D point cloud semantic segmentation (PCSS) have achieved outstanding performance by adopting neural networks. However, the robustness of these complex models have not been systematically analyzed. Given that PCSS has been applied in many safety-critical applications like autonomous driving, it is important to fill this knowledge gap, especially, how these models are affected under adversarial samples. As such, we present a comparative study of PCSS robustness. First, we formally define the attacker's objective under performance degradation and object hiding. Then, we develop new attack by whether to bound the norm. We evaluate different attack options on two datasets and three PCSS models. We found all the models are vulnerable and attacking point color is more effective. With this study, we call the attention of the research community to develop new approaches to harden PCSS models.