Saved in:
Bibliographic Details
Main Authors: Guo, Ji, Zhou, Long, Wang, Zhijin, He, Jiaming, Song, Qiyang, Chen, Aiguo, Jiang, Wenbo
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.16154
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915297839022080
author Guo, Ji
Zhou, Long
Wang, Zhijin
He, Jiaming
Song, Qiyang
Chen, Aiguo
Jiang, Wenbo
author_facet Guo, Ji
Zhou, Long
Wang, Zhijin
He, Jiaming
Song, Qiyang
Chen, Aiguo
Jiang, Wenbo
contents In recent years, deep learning-based Monocular Depth Estimation (MDE) models have been widely applied in fields such as autonomous driving and robotics. However, their vulnerability to backdoor attacks remains unexplored. To fill the gap in this area, we conduct a comprehensive investigation of backdoor attacks against MDE models. Typically, existing backdoor attack methods can not be applied to MDE models. This is because the label used in MDE is in the form of a depth map. To address this, we propose BadDepth, the first backdoor attack targeting MDE models. BadDepth overcomes this limitation by selectively manipulating the target object's depth using an image segmentation model and restoring the surrounding areas via depth completion, thereby generating poisoned datasets for object-level backdoor attacks. To improve robustness in physical world scenarios, we further introduce digital-to-physical augmentation to adapt to the domain gap between the physical world and the digital domain. Extensive experiments on multiple models validate the effectiveness of BadDepth in both the digital domain and the physical world, without being affected by environmental factors.
format Preprint
id arxiv_https___arxiv_org_abs_2505_16154
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle BadDepth: Backdoor Attacks Against Monocular Depth Estimation in the Physical World
Guo, Ji
Zhou, Long
Wang, Zhijin
He, Jiaming
Song, Qiyang
Chen, Aiguo
Jiang, Wenbo
Computer Vision and Pattern Recognition
In recent years, deep learning-based Monocular Depth Estimation (MDE) models have been widely applied in fields such as autonomous driving and robotics. However, their vulnerability to backdoor attacks remains unexplored. To fill the gap in this area, we conduct a comprehensive investigation of backdoor attacks against MDE models. Typically, existing backdoor attack methods can not be applied to MDE models. This is because the label used in MDE is in the form of a depth map. To address this, we propose BadDepth, the first backdoor attack targeting MDE models. BadDepth overcomes this limitation by selectively manipulating the target object's depth using an image segmentation model and restoring the surrounding areas via depth completion, thereby generating poisoned datasets for object-level backdoor attacks. To improve robustness in physical world scenarios, we further introduce digital-to-physical augmentation to adapt to the domain gap between the physical world and the digital domain. Extensive experiments on multiple models validate the effectiveness of BadDepth in both the digital domain and the physical world, without being affected by environmental factors.
title BadDepth: Backdoor Attacks Against Monocular Depth Estimation in the Physical World
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.16154