mlsanityData to 7 Oct 2026

Adversarial Training of Self-supervised Monocular Depth Estimation against Physical-World Attacks

arXiv 2301.13487 ↗First posted 31 Jan 2023cs.CV cs.AIcited by 17

Not on the map. Why?

Abstract (arXiv)

Monocular Depth Estimation (MDE) is a critical component in applications such as autonomous driving. There are various attacks against MDE networks. These attacks, especially the physical ones, pose a great threat to the security of such systems. Traditional adversarial training method requires ground-truth labels hence cannot be directly applied to self-supervised MDE that does not have ground-truth depth. Some self-supervised model hardening techniques (e.g., contrastive learning) ignore the domain knowledge of MDE and can hardly achieve optimal performance. In this work, we propose a novel adversarial training method for self-supervised MDE models based on view synthesis without using ground-truth depth. We improve adversarial robustness against physical-world attacks using L0-norm-bounded perturbation in training. We compare our method with supervised learning based and contrastive learning based methods that are tailored for MDE. Results on two representative MDE networks show that we achieve better robustness against various adversarial attacks with nearly no benign performance degradation.

Key facts. Cited by 17 papers in the map (data through 7 Oct 2026). Builds on 59 of its 69 references that are in the map. First posted on arXiv 31 Jan 2023.

Builds on

59 of 69 references.

References · 59

  1. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  2. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,482
  3. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  4. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  5. 2013Intriguing properties of neural networkscited by 5,114
  6. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  7. 2020Exploring Simple Siamese Representation Learningcited by 2,853
  8. 2018Unsupervised Feature Learning via Non-Parametric Instance-level Discriminationcited by 1,838
  9. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
  10. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
  11. 2019Contrastive Multiview Codingcited by 1,297
  12. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,259
38 more
  1. 2017Ensemble Adversarial Training: Attacks and Defensescited by 833
  2. 2016Unsupervised Monocular Depth Estimation with Left-Right Consistencycited by 819
  3. 2019Self-Supervised Learning of Pretext-Invariant Representationscited by 806
  4. 2019Square Attack: a query-efficient black-box adversarial attack via random searchcited by 682
  5. 2017Adversarial Patchcited by 514
  6. 2019Unlabeled Data Improves Adversarial Robustnesscited by 415
  7. 2016Unsupervised CNN for Single View Depth Estimation: Geometry to the Rescuecited by 368
  8. 2018GeoNet: Unsupervised Learning of Dense Depth, Optical Flow and Camera Posecited by 341
  9. 2019Unsupervised Scale-consistent Depth and Ego-motion Learning from Monocular Videocited by 244
  10. 2019Unsupervised Embedding Learning via Invariant and Spreading Instance Featurecited by 231
  11. 2020D3VO: Deep Depth, Deep Pose and Deep Uncertainty for Monocular Visual Odometrycited by 206
  12. 2021The Temporal Opportunist: Self-Supervised Multi-Frame Monocular Depthcited by 190
  13. 2018Adversarial Learning for Semi-Supervised Semantic Segmentationcited by 173
  14. 2019Are Labels Required for Improving Adversarial Robustness?cited by 162
  15. 2019Making an Invisibility Cloak: Real World Adversarial Attacks on Object Detectorscited by 155
  16. 2017Learning Depth from Monocular Videos using Direct Methodscited by 154
  17. 2019Adversarial Sensor Attack on LiDAR-based Perception in Autonomous Drivingcited by 151
  18. 2018DF-Net: Unsupervised Joint Learning of Depth and Flow using Cross-Task Consistencycited by 148
  19. 2020Adversarial Self-Supervised Contrastive Learningcited by 138
  20. 2020Physically Realizable Adversarial Examples for LiDAR Object Detectioncited by 137
  21. 2019Self-Supervised Monocular Depth Hintscited by 127
  22. 2017On the Robustness of Semantic Segmentation Models to Adversarial Attackscited by 126
  23. 2021Invisible for both Camera and LiDAR: Security of Multi-Sensor Fusion based Perception in Autonomous Driving Under Physical-World Attackscited by 101
  24. 2020Contrastive Learning with Adversarial Examplescited by 91
  25. 2021SG-Net: Spatial Granularity Network for One-Stage Video Instance Segmentationcited by 87
  26. 2021Adversarial Laser Beam: Effective Physical-World Attack to DNNs in a Blinkcited by 75
  27. 2020MonoRec: Semi-Supervised Dense Reconstruction in Dynamic Environments from a Single Moving Cameracited by 59
  28. 2022Physical Attack on Monocular Depth Estimation with Optimal Adversarial Patchescited by 57
  29. 2019Towards Adversarially Robust Object Detectioncited by 55
  30. 2019Recurrent Neural Network for (Un-)supervised Learning of Monocular VideoVisual Odometry and Depthcited by 50
  31. 2021Excavating the Potential Capacity of Self-Supervised Monocular Depth Estimationcited by 48
  32. 2021Exploring Adversarial Robustness of Multi-Sensor Perception Systems in Self Drivingcited by 45
  33. 2021Robust and Accurate Object Detection via Adversarial Learningcited by 42
  34. 2020Pseudo RGB-D for Self-Improving Monocular SLAM and Depth Predictioncited by 41
  35. 2020Targeted Adversarial Perturbations for Monocular Depth Predictioncited by 40
  36. 2020Learning Monocular Visual Odometry via Self-Supervised Long-Term Modelingcited by 39
  37. 2020Dynamic Divide-and-Conquer Adversarial Training for Robust Semantic Segmentationcited by 38
  38. 2021Adversarial Training Helps Transfer Learning via Better Representationscited by 36

50 most cited of 59 All 59 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 3; 2024: 8; 2025: 4; 2026: 2.

17 papers.

Sort:

By area:

  1. 2023CLUSTSEG: Clustering for Universal Segmentationcited by 32
  2. 2024Mitigating the Curse of Dimensionality for Certified Robustness via Dual Randomized Smoothingcited by 15
  3. 2023Fusion is Not Enough: Single Modal Attacks on Fusion Models for 3D Object Detectioncited by 11
  4. 2023Adversarial Examples in the Physical World: A Surveycited by 10
  5. 2025Learning-based 3D Reconstruction in Autonomous Driving: A Comprehensive Surveycited by 9
  6. 2025PosterGen: Aesthetic-Aware Multi-Modal Paper-to-Poster Generation via Multi-Agent LLMscited by 8
  7. 2024Self-supervised Adversarial Training of Monocular Depth Estimation against Physical-World Attackscited by 5
  8. 2024Benchmarking Robust Self-Supervised Learning Across Diverse Downstream Taskscited by 5
  9. 2024Prototypical Transformer as Unified Motion Learnerscited by 4
  10. 2024Improving Domain Generalization in Self-supervised Monocular Depth Estimation via Stabilized Adversarial Trainingcited by 4
  11. 2024BadPart: Unified Black-box Adversarial Patch Attacks against Pixel-wise Regression Taskscited by 3
  12. 2024Engineering Trustworthy AI: A Developer Guide for Empirical Risk Minimizationcited by 3
5 more
  1. 2026Certified vs. Empirical Adversarial Robust-ness via Hybrid Convolutions with Attention Stochasticitycited by 1
  2. 2024Physical Adversarial Attack on Monocular Depth Estimation via Shape-Varying Patchescited by 0
  3. 2025DepthVanish: Optimizing Adversarial Interval Structures for Stereo-Depth-Invisible Patchescited by 0
  4. 2025Pixel-Optimization-Free Patch Attack on Stereo Depth Estimationcited by 0
  5. 2026Adversarial Calibration Attack on Autonomous Vehiclescited by 0