Journal Article10.1109/tmm.2023.3304896
Robust Geometry-Dependent Attack for 3D Point Clouds
9
TL;DR: A novel Geometry-Dependent Attack (GDA), which aims to generate more robust adversarial point clouds with lower perturbation costs by capturing and preserving the geometry-guided topology information.
read more
Abstract: Deep learning models for point clouds have shown to be vulnerable to adversarial attacks, which have received increasing attention in various safety-critical applications such as autonomous driving, robotics, and surveillance. Since existing 3D attack methods either modify the local points or perform global point-wise perturbations over the point cloud, they fail to capture the dependency between neighboring points for preserving the geometrical context and topological smoothness of the original 3D object. In this article, we propose a novel Geometry-Dependent Attack (GDA), which aims to generate more robust adversarial point clouds with lower perturbation costs by capturing and preserving the geometry-guided topology information. Specifically, we first analyze the geometric information of each benign point cloud following the graph signal processing and disentangle it into low-frequency (flat) and high-frequency (contour) components. Then, considering the varying characteristics of smoothness and sharpness after disentanglement, we design two collaborative patch-aware and point-aware attacks to perturb these two components separately to misclassify the 3D object. We test the proposed GDA attack using five popular point cloud networks (PointNet, PointNet++, DGCNN, PointTransformer, and PointMLP) on both ModelNet40 and ShapNetPart datasets. Experimental results show that our GDA attack achieves 100% success rates with the lowest perturbation cost. It also demonstrates the increased capability to defeat several existing defense models over other competing attacks.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Point Cloud Attacks in Graph Spectral Domain: When 3D Geometry Meets Graph Signal Processing
Daizong Liu,Wei Hu,Xin Li +2 more
TL;DR: Point cloud attacks in the graph spectral domain perturb the spectral coefficients of point cloud data to manipulate its geometric structure.
11
Cross-Task Knowledge Transfer for Semi-supervised Joint 3D Grounding and Captioning
Yang Liu,Daizong Liu,Zongming Guo,Wei Hu +3 more
- 26 Oct 2024
TL;DR: This paper introduces a semi-supervised approach for 3D visual grounding and captioning, leveraging knowledge transfer from a correlated task to overcome the scarcity of annotated data, enabling accurate object localization in 3D scenes with limited text-object labels.
2
Frequency-Aware GAN for Imperceptible Transfer Attack on 3D Point Clouds
Xiaowen Cai,Yunbo Tao,Daizong Liu,Pan Zhou,Xiaoye Qu,Jianfeng Dong,Keke Tang,Lichao Sun +7 more
- 26 Oct 2024
TL;DR: This paper proposes a frequency-aware GAN for black-box transfer attacks on 3D point clouds, achieving high attack success rates and imperceptibility by leveraging Graph Fourier Transform and dual learning schemes to preserve geometric structures and improve transferability.
2
PointMLFF: Robust Point Cloud Analysis Based on Multi-Level Feature Fusion
Miao Yin,Lei Tan,Ping Wang,Jinshuo Zhang,Xiuyang Zhao +4 more
- 30 Jun 2024
TL;DR: PointMLFF proposes a robust 3D point cloud analysis model that captures local geometric features and abnormal changes through multi-level feature fusion, achieving high classification accuracy (88.9%) on challenging datasets and outperforming advanced methods in various downstream tasks.
Imperceptible Transfer Attack on Large Vision-Language Models
Xiaowen Cai,Daizong Liu,Runwei Guan,Pan Zhou +3 more
- 06 Apr 2025
TL;DR: This paper proposes Imperceptible Transfer Attack (ITA), a novel method to generate prompt/model-agnostic adversarial samples for Large Vision-Language Models, enhancing transferability and imperceptibility through visual transformations and gradient approximation.
References
•Posted Content
Explaining and Harnessing Adversarial Examples
TL;DR: The authors argue that the primary cause of neural networks' vulnerability to adversarial perturbation is their linear nature, which is supported by new quantitative results while giving the first explanation of the most intriguing fact about adversarial examples: their generalization across architectures and training sets.
15.9K
PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
R. Qi Charles,Hao Su,Mo Kaichun,Leonidas J. Guibas +3 more
- 21 Jul 2017
TL;DR: This paper designs a novel type of neural network that directly consumes point clouds, which well respects the permutation invariance of points in the input and provides a unified architecture for applications ranging from object classification, part segmentation, to scene semantic parsing.
A tutorial on spectral clustering
TL;DR: In this article, the authors present the most common spectral clustering algorithms, and derive those algorithms from scratch by several different approaches, and discuss the advantages and disadvantages of these algorithms.
Towards Evaluating the Robustness of Neural Networks
Nicholas Carlini,David Wagner +1 more
- 22 May 2017
TL;DR: In this paper, the authors demonstrate that defensive distillation does not significantly increase the robustness of neural networks by introducing three new attack algorithms that are successful on both distilled and undistilled neural networks with 100% probability.
•Posted Content
Towards Deep Learning Models Resistant to Adversarial Attacks
TL;DR: This work studies the adversarial robustness of neural networks through the lens of robust optimization, and suggests the notion of security against a first-order adversary as a natural and broad security guarantee.