Knowledge Commons of Institute of Automation,CAS
Towards Effective Adversarial Attack on Point Cloud for 3D Classification | |
Chengcheng Ma1,4; Weiliang Meng1,2,3,4; Baoyuan Wu5,6; Shibiao Xu1,4; Xiaopeng Zhang1,2,4 | |
2021-07 | |
会议名称 | IEEE International Conference on Multimedia and Expo (ICME) 2021 |
会议日期 | July 5-9, 2021 |
会议地点 | Virtual |
摘要 | In the domain of 3D point cloud classification, deep learning based classifiers have made significant progress, while they have been also proven to be vulnerable on the adversarial attack at the same time. Some recent works employ the attack methods that devised for image classification such as projected gradient descent (PGD) to attack the 3D classifiers, but their performances seem quite limited when faced with statistical operations including point cloud denoising and point cloud upsampling. In this paper, we propose ‘SmoothAttack’, a new attack that can craft adversarial point clouds robust to statistical operations. SmoothAttack can be easily applied in both global constraint and pointwise constraint. Besides, we analyze the directions of perturbations onto the point cloud during the iteration process, where SmoothAttack can somehow stabilize the direction and make full use of the adversarial budgets. Experiments validate that our ‘SmoothAttack’ can raise the attack success rates against statistical defenses up to 98% for untargeted attack and 91% for targeted attack on ModelNet40 database when fooling the classifiers PointNet and DGCNN. |
语种 | 英语 |
七大方向——子方向分类 | 模式识别基础 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/47428 |
专题 | 多模态人工智能系统全国重点实验室_三维可视计算 多模态人工智能系统全国重点实验室 |
通讯作者 | Shibiao Xu; Xiaopeng Zhang |
作者单位 | 1.NLPR, Institute of Automation, Chinese Academy of Sciences 2.Zhejiang Lab 3.The State Key Laboratory of Computer Science, Institute of Software, Chinese Academy of Sciences, 4.School of Artificial Intelligence, University of Chinese Academy of Sciences 5.School of Data Science, The Chinese University of Hong Kong, Shenzhen, China 6.Shenzhen Research Institute of Big Data |
第一作者单位 | 模式识别国家重点实验室 |
通讯作者单位 | 模式识别国家重点实验室 |
推荐引用方式 GB/T 7714 | Chengcheng Ma,Weiliang Meng,Baoyuan Wu,et al. Towards Effective Adversarial Attack on Point Cloud for 3D Classification[C],2021. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
Towards_Effective_Ad(1296KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论