Multi-view Correlation based Black-box Adversarial Attack for 3D Object Detection

对抗制 点云 计算机科学 激光雷达 人工智能 深度学习 对象(语法) 目标检测 计算机视觉 光学(聚焦) 分割 图像分割 遥感 地理 物理 光学
作者
Bingyu Liu,Yuhong Guo,Jianan Jiang,Jian Tang,Weihong Deng
出处
期刊:Knowledge Discovery and Data Mining 卷期号:: 1036-1044 被引量:2
标识
DOI:10.1145/3447548.3467432
摘要

Deep neural networks have made tremendous progress in 3D object detection, which is an important task especially in autonomous driving scenarios. Benefited from the breakthroughs in deep learning and sensor technologies, 3D object detection methods based on different sensors, such as camera and LiDAR, have developed rapidly. Meanwhile, more and more researches notice that the abundant information contained in the multi-view data can be used to obtain more accurate understanding of the 3D surrounding environment. Therefore, many sensor-fusion 3D object detection methods have been proposed. As safety is critical in autonomous driving and the deep neural networks are known to be vulnerable to adversarial examples with visually imperceptible perturbations, it is significant to investigate adversarial attacks for 3D object detection. Recent works have shown that both image-based and LiDAR-based networks can be attacked by the adversarial examples while the attacks to the sensor-fusion models, which tend to be more robust, haven't been studied. To this end, we propose a simple multi-view correlation based adversarial attack method for the camera-LiDAR fusion 3D object detection models and focus on the black-box attack setting which is more practical in real-world systems. Specifically, we first design a generative network to generate image adversarial examples based on an auxiliary image semantic segmentation network. Then, we develop a cross-view perturbation projection method by exploiting the camera-LiDAR correlations to map each image adversarial example to the space of the point cloud data to form the point cloud adversarial examples in the LiDAR view. Extensive experiments on the KITTI dataset demonstrate the effectiveness of the proposed method.

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
XSY完成签到 ,获得积分10
刚刚
刚刚
舒适鸣凤的应助被当你采纳,获得10
刚刚
淡定的伊完成签到 ,获得积分10
1秒前
隐形曼青的应助被三又一十八采纳,获得10
1秒前
2397完成签到,获得积分20
2秒前
今后的应助被淡定的板栗采纳,获得10
2秒前
huoo发布了新的文献求助10
3秒前
3秒前
做完实验打boss完成签到,获得积分10
3秒前
臭屁发布了新的文献求助10
3秒前
Nole的应助被精明丹妗采纳,获得10
3秒前
赵妍发布了新的文献求助10
3秒前
3秒前
4秒前
L912294993完成签到,获得积分10
4秒前
自然的小鸭子完成签到,获得积分10
5秒前
5秒前
zj发布了新的文献求助10
5秒前
单薄雪巧发布了新的文献求助10
5秒前
乐乐的应助被herococa采纳,获得10
5秒前
keyan完成签到,获得积分10
5秒前
大方的契发布了新的文献求助10
6秒前
6秒前
7秒前
7秒前
orixero的应助被Seorina采纳,获得10
7秒前
年轻山河完成签到 ,获得积分10
8秒前
传奇3的应助被十一采纳,获得10
8秒前
传奇3的应助被动人的臻采纳,获得10
8秒前
tomatoGG发布了新的文献求助10
8秒前
Kin发布了新的文献求助10
9秒前
李天王发布了新的文献求助10
10秒前
Akim的应助被初海歌采纳,获得10
11秒前
小叮当完成签到,获得积分10
11秒前
12秒前
3503264269发布了新的文献求助10
12秒前
领导范儿的应助被番茄炒蛋采纳,获得30
12秒前
13秒前
Felicia完成签到,获得积分10
13秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 2000
Handbook of Sport Coaching Research 1000
Orthognathic Surgery - 2 Volume Set 500
高分子基功能界面水蒸发材料与应用研究 500
外场增强的水汽发电研究 500
基于结构调控的功能化水诱导发电材料与器件研究 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8193335
求助须知:如何正确求助?哪些是违规求助? 9579489
关于积分的说明 21865844
捐赠科研通 7792843
什么是DOI,文献DOI怎么找? 3437210
关于科研通互助平台的介绍 2560687
邀请新用户注册赠送积分活动 2483996