AdvPC: Transferable Adversarial Perturbations on 3D Point Clouds

Abdullah Hamdi, Sara Rojas, Ali Kassem Thabet, Bernard Ghanem

Research output: Chapter in Book/Report/Conference proceedingConference contribution

72 Scopus citations

Abstract

Deep neural networks are vulnerable to adversarial attacks, in which imperceptible perturbations to their input lead to erroneous network predictions. This phenomenon has been extensively studied in the image domain, and has only recently been extended to 3D point clouds. In this work, we present novel data-driven adversarial attacks against 3D point cloud networks. We aim to address the following problems in current 3D point cloud adversarial attacks: they do not transfer well between different networks, and they are easy to defend against via simple statistical methods. To this extent, we develop a new point cloud attack (dubbed AdvPC) that exploits the input data distribution by adding an adversarial loss, after Auto-Encoder reconstruction, to the objective it optimizes. AdvPC leads to perturbations that are resilient against current defenses, while remaining highly transferable compared to state-of-the-art attacks. We test AdvPC using four popular point cloud networks: PointNet, PointNet++ (MSG and SSG), and DGCNN. Our proposed attack increases the attack success rate by up to 40% for those transferred to unseen networks (transferability), while maintaining a high success rate on the attacked network. AdvPC also increases the ability to break defenses by up to 38% as compared to other baselines on the ModelNet40 dataset. The code is available at https://github.com/ajhamdi/AdvPC.
Original languageEnglish (US)
Title of host publicationComputer Vision – ECCV 2020
PublisherSpringer International Publishing
Pages241-257
Number of pages17
ISBN (Print)9783030586096
DOIs
StatePublished - Oct 6 2020

Fingerprint

Dive into the research topics of 'AdvPC: Transferable Adversarial Perturbations on 3D Point Clouds'. Together they form a unique fingerprint.

Cite this