PointCert: Point Cloud Classification with Deterministic Certified Robustness Guarantees. (arXiv:2303.01959v1 [cs.CR])

Point cloud classification is an essential component in many
security-critical applications such as autonomous driving and augmented
reality. However, point cloud classifiers are vulnerable to adversarially
perturbed point clouds. Existing certified defenses against adversarial point
clouds suffer from a key limitation: their certified robustness guarantees are
probabilistic, i.e., they produce an incorrect certified robustness guarantee
with some probability. In this work, we propose a general framework, namely
PointCert, that can transform an arbitrary point cloud classifier to be
certifiably robust …

adversarial applications augmented reality autonomous autonomous driving certified classification cloud clouds critical driving framework general guarantee key point robustness security vulnerable work