Contextual Attribution Maps-Guided Transferable Adversarial Attack for 3D Object Detection

Mumuxin Cai; Xupeng Wang; Ferdous Sohel; Hang Lei

doi:10.3390/rs16234409

Back

Contextual Attribution Maps-Guided Transferable Adversarial Attack for 3D Object Detection

Journal article

Open access

Peer reviewed

Contextual Attribution Maps-Guided Transferable Adversarial Attack for 3D Object Detection

Mumuxin Cai, Xupeng Wang, Ferdous Sohel and Hang Lei

Remote sensing (Basel, Switzerland), Vol.16(23), 4409

2024

DOI: https://doi.org/10.3390/rs16234409

Files and links (1)

pdf

Published1.58 MBDownload View

CC BY V4.0, Open Access

Abstract

Environmental Sciences

Environmental Sciences & Ecology

Geology

Geosciences, Multidisciplinary

Imaging Science & Photographic Technology

Life Sciences & Biomedicine

Physical Sciences

Remote Sensing

Science & Technology

Technology

The study of LiDAR-based 3D object detection and its robustness under adversarial attacks has achieved great progress. However, existing adversarial attack methods mainly focus on the targeted object, which destroys the integrity of the object and makes the attack easy to perceive. In this work, we propose a novel adversarial attack against deep 3D object detection models named the contextual attribution maps-guided attack (CAMGA). Based on the combinations of subregions in the context area and their impact on the prediction results, contextual attribution maps can be generated. An attribution map exposes the influence of individual subregions in the context area on the detection results and narrows down the scope of the adversarial attack. Subsequently, perturbations are generated under the guidance of a dual loss, which is proposed to suppress the detection results and maintain visual imperception simultaneously. The experimental results proved that the CAMGA method achieved an attack success rate of over 68% on three large-scale datasets and 83% on the KITTI dataset. Meanwhile, the CAMGA has a transfer attack success rate of at least 50% against all four victim detectors, as they all overly rely on contextual information.

Details

Title: Contextual Attribution Maps-Guided Transferable Adversarial Attack for 3D Object Detection
Authors/Creators: Mumuxin Cai - University of Electronic Science and Technology of China
Xupeng Wang - University of Electronic Science and Technology of China
Ferdous Sohel - Murdoch University, Centre for Crop and Food Innovation
Hang Lei - China Post (China)
Publication Details: Remote sensing (Basel, Switzerland), Vol.16(23), 4409
Publisher: MDPI
Number of pages: 20
Grant note: 2022ZDZX0005; 24ZDZX0001; 24ZDZX0002 / Sichuan Provincial Research Plan Project National Natural Science Foundation of China; National Natural Science Foundation of China (NSFC)
Identifiers: 991005726581007891
Murdoch Affiliation: School of Information Technology
Language: English
Resource Type: Journal article

Metrics

2 File views/ downloads

8 Record Views

3 Times Cited - Web of Science

InCites Highlights

These are selected metrics from InCites Benchmarking & Analytics tool, related to this output

Collaboration types: Domestic collaboration; International collaboration
Citation topics: 4 Electrical Engineering, Electronics & Computer Science; 4.17 Computer Vision & Graphics; 4.17.2798 Stereo Depth Estimation
Web Of Science research areas: Environmental Sciences; Geosciences, Multidisciplinary; Imaging Science & Photographic Technology; Remote Sensing
ESI research areas: Geosciences