Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
129 tokens/sec
GPT-4o
28 tokens/sec
Gemini 2.5 Pro Pro
42 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Physical 3D Adversarial Attacks against Monocular Depth Estimation in Autonomous Driving (2403.17301v2)

Published 26 Mar 2024 in cs.CV and cs.CR

Abstract: Deep learning-based monocular depth estimation (MDE), extensively applied in autonomous driving, is known to be vulnerable to adversarial attacks. Previous physical attacks against MDE models rely on 2D adversarial patches, so they only affect a small, localized region in the MDE map but fail under various viewpoints. To address these limitations, we propose 3D Depth Fool (3D$2$Fool), the first 3D texture-based adversarial attack against MDE models. 3D$2$Fool is specifically optimized to generate 3D adversarial textures agnostic to model types of vehicles and to have improved robustness in bad weather conditions, such as rain and fog. Experimental results validate the superior performance of our 3D$2$Fool across various scenarios, including vehicles, MDE models, weather conditions, and viewpoints. Real-world experiments with printed 3D textures on physical vehicle models further demonstrate that our 3D$2$Fool can cause an MDE error of over 10 meters.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (44)
  1. Tesla ai day. https://youtu.be/j0z4FweCy4M?t=5295, a.
  2. Tesla use per-pixel depth estimation with self-supervised learning. https://youtu.be/hx7BXih7zx8?t=1334, b.
  3. Automold. https://github.com/UjjwalSaxena/Automold--Road-Augmentation-Library, 2022-12-20.
  4. Synthesizing robust adversarial examples. In Proceedings of the 35th International Conference on Machine Learning, (ICML), pages 284–293, 2018.
  5. Adversarial patch. arXiv preprint arXiv:1712.09665, 2017.
  6. Towards evaluating the robustness of neural networks. In 2017 IEEE Symposium on Security and Privacy (SP), pages 39–57, Los Alamitos, CA, USA, 2017. IEEE Computer Society.
  7. Adversarial attacks on monocular pose estimation. In 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 12500–12505, 2022.
  8. Physical attack on monocular depth estimation with optimal adversarial patches. In Computer Vision – ECCV 2022, pages 514–532, Cham, 2022. Springer Nature Switzerland.
  9. Gesture recognition using a depth camera for human robot collaboration on assembly line. Procedia Manufacturing, 3:518–525, 2015. 6th International Conference on Applied Human Factors and Ergonomics (AHFE 2015) and the Affiliated Conferences, AHFE 2015.
  10. Vision for mobile robot navigation: a survey. IEEE Transactions on Pattern Analysis and Machine Intelligence, 24(2):237–267, 2002.
  11. CARLA: An open urban driving simulator. In Proceedings of the 1st Annual Conference on Robot Learning, pages 1–16, 2017.
  12. Depth map prediction from a single image using a multi-scale deep network. In Neural Information Processing Systems, 2014.
  13. Robust physical-world attacks on deep learning visual classification. In 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 1625–1634, 2018.
  14. Digging into self-supervised monocular depth prediction. 2019.
  15. Explaining and harnessing adversarial examples. CoRR, abs/1412.6572, 2014.
  16. Aparate: Adaptive adversarial patch for cnn-based monocular depth estimation for autonomous navigation, 2023a.
  17. Saam: Stealthy adversarial attack on monoculor depth estimation. ArXiv, abs/2308.03108, 2023b.
  18. Neural 3d mesh renderer. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018.
  19. Adam: A method for stochastic optimization. In 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7-9, 2015, Conference Track Proceedings, 2015.
  20. Deeper depth prediction with fully convolutional residual networks. In 3D Vision (3DV), 2016 Fourth International Conference on, pages 239–248. IEEE, 2016.
  21. Learning depth from single monocular images using deep convolutional neural fields. IEEE T. Pattern Analysis and Machine Intelligence, 2015.
  22. Deep photo style transfer. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017.
  23. Slowtrack: Increasing the latency of camera-based perception in autonomous driving using adversarial examples. arXiv preprint arXiv:2312.09520, 2023a.
  24. Wip: Towards the practicality of the adversarial attack on object tracking in autonomous driving. In ISOC Symposium on Vehicle Security and Privacy (VehicleSec), 2023b.
  25. Trafficpredict: Trajectory prediction for heterogeneous traffic-agents. Proceedings of the AAAI Conference on Artificial Intelligence, 33:6120–6127, 2019.
  26. Understanding deep image representations by inverting them. In 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 5188–5196, Los Alamitos, CA, USA, 2015. IEEE Computer Society.
  27. CoMoGAN: continuous model-guided image-to-image translation. In CVPR, 2021.
  28. Monocular depth estimation using neural regression forest. In 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 5506–5514, 2016.
  29. Self-supervised Monocular Depth Estimation: Let’s Talk About The Weather. In The International Conference on Computer Vision (ICCV), 2023.
  30. Accessorize to a crime: Real and stealthy attacks on state-of-the-art face recognition. In Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security, page 1528–1540, New York, NY, USA, 2016. Association for Computing Machinery.
  31. Real-time stereo reconstruction in robotically assisted minimally invasive surgery. In Medical Image Computing and Computer-Assisted Intervention – MICCAI 2010, pages 275–282, Berlin, Heidelberg, 2010. Springer Berlin Heidelberg.
  32. Dta: Physical camouflage attacks using differentiable transformation network. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 15305–15314, 2022.
  33. Active: Towards highly transferable 3d physical camouflage for universal and robust vehicle evasion. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pages 4305–4314, 2023.
  34. Fooling automated surveillance cameras: Adversarial patches to attack person detection. In 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pages 49–55, Los Alamitos, CA, USA, 2019. IEEE Computer Society.
  35. Rain rendering for evaluating and improving robustness to bad weather. International Journal of Computer Vision, 2020.
  36. Fca: Learning a 3d full-coverage vehicle camouflage for multi-view physical adversarial attack. In Proceedings of the AAAI conference on artificial intelligence, pages 2414–2422, 2022.
  37. Dual attention suppression attack: Generate adversarial camouflage in physical world. In 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 8561–8570, Los Alamitos, CA, USA, 2021. IEEE Computer Society.
  38. Self-supervised monocular depth hints. In The International Conference on Computer Vision (ICCV), 2019.
  39. The Temporal Opportunist: Self-Supervised Multi-Frame Monocular Depth. In Computer Vision and Pattern Recognition (CVPR), 2021.
  40. Targeted adversarial perturbations for monocular depth prediction. In Advances in neural information processing systems, 2020.
  41. Physical adversarial attack on vehicle detector in the carla simulator. ArXiv, abs/2007.16118, 2020.
  42. Adversarial patch attacks on monocular depth estimation networks. IEEE Access, 8:179094–179104, 2020.
  43. Quantization aware attack: Enhancing transferable adversarial attacks by model quantization. IEEE Transactions on Information Forensics and Security, 19:3265–3278, 2024.
  44. Camou: Learning physical vehicle camouflages to adversarially attack detectors in the wild. In International Conference on Learning Representations, 2018.
Citations (8)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now