mlsanityData to 7 Oct 2026

Physical Passive Patch Adversarial Attacks on Visual Odometry Systems

arXiv 2207.05729 ↗First posted 11 Jul 2022cs.CV cs.LGcited by 6

Not on the map. Why?

Abstract (arXiv)

Deep neural networks are known to be susceptible to adversarial perturbations -- small perturbations that alter the output of the network and exist under strict norm limitations. While such perturbations are usually discussed as tailored to a specific input, a universal perturbation can be constructed to alter the model's output on a set of inputs. Universal perturbations present a more realistic case of adversarial attacks, as awareness of the model's exact input is not required. In addition, the universal attack setting raises the subject of generalization to unseen data, where given a set of inputs, the universal perturbations aim to alter the model's output on out-of-sample data. In this work, we study physical passive patch adversarial attacks on visual odometry-based autonomous navigation systems. A visual odometry system aims to infer the relative camera motion between two corresponding viewpoints, and is frequently used by vision-based autonomous navigation systems to estimate their state. For such navigation systems, a patch adversarial perturbation poses a severe security issue, as it can be used to mislead a system onto some collision course. To the best of our knowledge, we show for the first time that the error margin of a visual odometry model can be significantly increased by deploying patch adversarial attacks in the scene. We provide evaluation on synthetic closed-loop drone navigation data and demonstrate that a comparable vulnerability exists in real data. A reference implementation of the proposed method and the reported experiments is provided at https://github.com/patchadversarialattacks/patchadversarialattacks.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 28 of its 40 references that are in the map. First posted on arXiv 11 Jul 2022.

Builds on

28 of 40 references.

References · 28

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  3. 2013Intriguing properties of neural networkscited by 5,114
  4. 2015ORB-SLAM: a Versatile and Accurate Monocular SLAM Systemcited by 983
  5. 2016Universal adversarial perturbationscited by 774
  6. 2019Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailmentcited by 663
  7. 2017Adversarial Patchcited by 514
  8. 2018Black-box Generation of Adversarial Text Sequences to Evade Deep Learning Classifierscited by 341
  9. 2018Audio Adversarial Examples: Targeted Attacks on Speech-to-Textcited by 268
  10. 2019Unsupervised Scale-consistent Depth and Ego-motion Learning from Monocular Videocited by 244
  11. 2019Adversarial Policies: Attacking Deep Reinforcement Learningcited by 209
  12. 2020D3VO: Deep Depth, Deep Pose and Deep Uncertainty for Monocular Visual Odometrycited by 206
16 more
  1. 2018Generating 3D Adversarial Point Cloudscited by 144
  2. 2020TartanVO: A Generalizable Learning-based VOcited by 107
  3. 2017Universal Adversarial Perturbations Against Semantic Image Segmentationcited by 92
  4. 2020An Analysis of Adversarial Attacks and Defenses on Autonomous Driving Modelscited by 67
  5. 2021A Survey On Universal Adversarial Attackcited by 59
  6. 2020Unadversarial Examples: Designing Objects for Robust Visioncited by 45
  7. 2020Over-the-Air Adversarial Flickering Attacks against Video Recognition Networkscited by 35
  8. 2020Adversarial Patch Attacks on Monocular Depth Estimation Networkscited by 29
  9. 2020Adversarial Examples in Deep Learning for Multivariate Time Series Regressioncited by 26
  10. 2019SelfVIO: Self-Supervised Deep Monocular Visual-Inertial Odometry and Depth Estimationcited by 23
  11. 2018Adversarial Attacks, Regression, and Numerical Stability Regularizationcited by 14
  12. 2019Adversarial T-shirt! Evading Person Detectors in A Physical Worldcited by 13
  13. 2021A Simple and Strong Baseline for Universal Targeted Attacks on Siamese Visual Trackingcited by 9
  14. 2019Daedalus: Breaking Non-Maximum Suppression in Object Detection via Adversarial Examplescited by 8
  15. 2018High Dimensional Spaces, Deep Learning and Adversarial Examplescited by 5
  16. 2019Exploring the Robustness of NMT Systems to Nonsensical Inputscited by 5

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 3; 2024: 1; 2025: 1.

6 papers.

  1. 2022A Survey on Physical Adversarial Attack in Computer Visioncited by 24
  2. 2023Flying Adversarial Patches: Manipulating the Behavior of Deep Learning-based Autonomous Multirotorscited by 1
  3. 2023Kidnapping Deep Learning-based Multirotors using Optimized Flying Adversarial Patchescited by 1
  4. 2023Robust Adversarial Attacks Detection for Deep Learning based Relative Pose Estimation for Space Rendezvouscited by 0
  5. 2024Sparse patches adversarial attacks via extrapolating point-wise informationcited by 0
  6. 2025Black-box Adversarial Attacks on CNN-based SLAM Algorithmscited by 0