Due to the booming of autonomous driving, in which LiDAR plays a critical role in the task of environment perception, its reliability issues have drawn much attention recently. LiDARs usually utilize deep neural models for 3D point cloud perception, which have been demonstrated to be vulnerable to imperceptible adversarial examples. However, prior work usually manipulates point clouds in the digital world without considering the physical working principle of the actual LiDAR. As a result, the generated adversarial point clouds may be realizable and effective in simulation but cannot be perceived by physical LiDARs. In this work, we introduce the physical principle of LiDARs and propose a new method for generating 3D adversarial point clouds in accord with it that can achieve two types of spoofing attacks: object hiding and object creating. We also evaluate the effectiveness of the proposed method with two 3D object detectors on the KITTI vision benchmark.
Generating 3D Adversarial Point Clouds under the Principle of LiDARs
Bo Yang (Zhejiang University), Yushi Cheng (Tsinghua University), Zizhi Jin (Zhejiang University), Xiaoyu Ji (Zhejiang University) and Wenyuan Xu (Zhejiang University)
View More Papers
Yulong Cao (University of Michigan), Yanan Guo (University of Pittsburgh), Takami Sato (UC Irvine), Qi Alfred Chen (UC Irvine), Z....Read More
Brian Kondracki (Stony Brook University), Babak Amin Azad (Stony Brook University), Najmeh Miramirkhani (Stony Brook University), Nick Nikiforakis (Stony Brook...Read More
Gonzalo De La Torre Parra (University of the Incarnate Word, TX, USA), Luis Selvera (Secure AI and Autonomy Lab, The...Read More
Xueluan Gong (Wuhan University), Yanjiao Chen (Zhejiang University), Jianshuo Dong (Wuhan University), Qian Wang (Wuhan University)Read More