mlsanityData to 7 Oct 2026

The art of defense: letting networks fool the attacker

arXiv 2104.02963 ↗First posted 7 Apr 2021cs.CV cs.LGcited by 6

Not on the map. Why?

Abstract (arXiv)

Robust environment perception is critical for autonomous cars, and adversarial defenses are the most effective and widely studied ways to improve the robustness of environment perception. However, all of previous defense methods decrease the natural accuracy, and the nature of the DNNs itself has been overlooked. To this end, in this paper, we propose a novel adversarial defense for 3D point cloud classifier that makes full use of the nature of the DNNs. Due to the disorder of point cloud, all point cloud classifiers have the property of permutation invariant to the input point cloud. Based on this nature, we design invariant transformations defense (IT-Defense). We show that, even after accounting for obfuscated gradients, our IT-Defense is a resilient defense against state-of-the-art (SOTA) 3D attacks. Moreover, IT-Defense do not hurt clean accuracy compared to previous SOTA 3D defenses. Our code is available at: footnotesizeurlhttps://github.com/cuge1995/IT-Defense.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 63 of its 95 references that are in the map. First posted on arXiv 7 Apr 2021.

Builds on

63 of 95 references.

References · 63

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2020Language Models are Few-Shot Learnerscited by 32,199
  4. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  5. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  6. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  7. 2013Intriguing properties of neural networkscited by 5,114
  8. 2015You Only Look Once: Unified, Real-Time Object Detectioncited by 4,829
  9. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  10. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  11. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,220
  12. 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,324
38 more
  1. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
  2. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,259
  3. 2016Adversarial Machine Learning at Scalecited by 934
  4. 2019PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detectioncited by 890
  5. 2017Ensemble Adversarial Training: Attacks and Defensescited by 833
  6. 2016Universal adversarial perturbationscited by 774
  7. 2020PCT: Point cloud transformercited by 730
  8. 2020Fast is better than free: Revisiting adversarial trainingcited by 723
  9. 2018PointConv: Deep Convolutional Networks on 3D Point Cloudscited by 656
  10. 2019Adversarial Training for Free!cited by 638
  11. 2017One pixel attack for fooling deep neural networkscited by 579
  12. 2020On Adaptive Attacks to Adversarial Example Defensescited by 547
  13. 2017Countering Adversarial Images using Input Transformationscited by 517
  14. 2019Making Convolutional Networks Shift-Invariant Againcited by 390
  15. 2017Mitigating Adversarial Effects Through Randomizationcited by 382
  16. 2018Defense-GAN: Protecting Classifiers Against Adversarial Attacks Using Generative Modelscited by 362
  17. 2018Characterizing Adversarial Subspaces Using Local Intrinsic Dimensionalitycited by 280
  18. 2017PixelDefend: Leveraging Generative Models to Understand and Defend against Adversarial Examplescited by 243
  19. 2019Improving Adversarial Robustness via Promoting Ensemble Diversitycited by 213
  20. 2020Understanding and Improving Fast Adversarial Trainingcited by 213
  21. 2021PAConv: Position Adaptive Convolution with Dynamic Kernel Assembling on Point Cloudscited by 211
  22. 2018PU-Net: Point Cloud Upsampling Networkcited by 208
  23. 2016Wider or Deeper: Revisiting the ResNet Model for Visual Recognitioncited by 184
  24. 2019You Only Propagate Once: Accelerating Adversarial Training via Maximal Principlecited by 181
  25. 2018Stochastic Activation Pruning for Robust Adversarial Defensecited by 171
  26. 2017Towards Robust Neural Networks via Random Self-ensemblecited by 158
  27. 2018Generating 3D Adversarial Point Cloudscited by 144
  28. 2018PointCloud Saliency Mapscited by 112
  29. 2016Towards the Science of Security and Privacy in Machine Learningcited by 89
  30. 2019AdvPC: Transferable Adversarial Perturbations on 3D Point Cloudscited by 84
  31. 2019The Odds are Odd: A Statistical Test for Detecting Adversarial Examplescited by 75
  32. 2021PointCutMix: Regularization Strategy for Point Cloud Classificationcited by 69
  33. 2018Adv-BNN: Improved Adversarial Defense through Robust Bayesian Neural Networkcited by 68
  34. 2019ME-Net: Towards Effective Adversarial Robustness with Matrix Estimationcited by 65
  35. 2019Adversarial Attack and Defense on Point Setscited by 63
  36. 2020Learning Meta Face Recognition in Unseen Domainscited by 61
  37. 2020LG-GAN: Label Guided Adversarial Network for Flexible Targeted Attack of Point Cloud-based Deep Networkscited by 55
  38. 2019Mixup Inference: Better Exploiting Mixup to Defend Adversarial Attackscited by 54

50 most cited of 63 All 63 (JSON)

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 1; 2024: 2; 2025: 2.

6 papers.

  1. 2022Robust Structured Declarative Classifiers for 3D Point Clouds: Defending Adversarial Attacks with Implicit Gradientscited by 17
  2. 2023SAAM: Stealthy Adversarial Attack on Monocular Depth Estimationcited by 12
  3. 2025UniShield: An Adaptive Multi-Agent Framework for Unified Forgery Image Detection and Localizationcited by 9
  4. 2024A Random Ensemble of Encrypted Vision Transformers for Adversarially Robust Defensecited by 3
  5. 2024Robust 3D Point Clouds Classification based on Declarative Defenderscited by 2
  6. 2025KNN-Defense: Defense against 3D Adversarial Point Clouds using Nearest-Neighbor Searchcited by 0