ShapeAdv: Generating Shape-Aware Adversarial 3D Point Clouds

2020/05/24 by Kibok Lee, Zhuoyuan Chen, Lee, Kibok +7 · 4 citations
Computer Science · Engineering · Materials Science · #Adversarial Robustness in Machine Learning #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #FOS: Electrical engineering #High-Velocity Impact and Material Behavior #Image and Video Processing (eess.IV) #Machine Learning (cs.LG) #cs.CV #cs.LG #eess.IV #electronic engineering #information engineering

paper · pdf · doi:10.48550/arxiv.2005.11626

3D Point Clouds, Adversarial Learning

arxiv created 2020/05/24 · openalex publication_date 2020/05/24 · arxiv updated 2020/05/26 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We introduce ShapeAdv, a novel framework to study shape-aware adversarial perturbations that reflect the underlying shape variations (e.g., geometric deformations and structural differences) in the 3D point cloud space. We develop shape-aware adversarial 3D point cloud attacks by leveraging the learned latent space of a point cloud auto-encoder where the adversarial noise is applied in the latent space. Specifically, we propose three different variants including an exemplar-based one by guiding the shape deformation with auxiliary data, such that the generated point cloud resembles the shape morphing between objects in the same category. Different from prior works, the resulting adversarial 3D point clouds reflect the shape variations in the 3D point cloud space while still being close to the original one. In addition, experimental evaluations on the ModelNet40 benchmark demonstrate that our adversaries are more difficult to defend with existing point cloud defense methods and exhibit a higher attack transferability across classifiers. Our shape-aware adversarial attacks are orthogonal to existing point cloud based attacks and shed light on the vulnerability of 3D deep neural networks.

Citations

• ShapeNet: An Information-Rich 3D Model Repository
• Improving Network Robustness against Adversarial Attacks with Compact Convolution
• Dex-Net 2.0: Deep Learning to Plan Robust Grasps with Synthetic Point Clouds and Analytic Grasp Metrics
• PixelDefend: Leveraging Generative Models to Understand and Defend against Adversarial Examples
• PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space

Cited by

• OBJVanish: Physically Realizable Text-to-3D Adv. Generation of LiDAR-Invisible Objects
• Explainability-Aware One Point Attack for Point Cloud Neural Networks
• Cage-Based Deformation for Transferable and Undefendable Point Cloud Attack
• KNN-Defense: Defense against 3D Adversarial Point Clouds using Nearest-Neighbor Search

Related

• Adversarial shape perturbations on 3D point clouds
• Geometric Adversarial Attacks and Defenses on 3D Point Clouds
• Generating 3D Adversarial Point Clouds
• Transferable 3D Adversarial Shape Completion using Diffusion Models
• Hide in Thicket: Generating Imperceptible and Rational Adversarial Perturbations on 3D Point Clouds