mlsanityData to 14 Oct 2026

Robust Structured Declarative Classifiers for 3D Point Clouds: Defending Adversarial Attacks with Implicit Gradients

arXiv 2203.15245 ↗First posted 29 Mar 2022cs.CVcited by 17

Not on the map. Why?

Abstract (arXiv)

Deep neural networks for 3D point cloud classification, such as PointNet, have been demonstrated to be vulnerable to adversarial attacks. Current adversarial defenders often learn to denoise the (attacked) point clouds by reconstruction, and then feed them to the classifiers as input. In contrast to the literature, we propose a family of robust structured declarative classifiers for point cloud classification, where the internal constrained optimization mechanism can effectively defend adversarial attacks through implicit gradients. Such classifiers can be formulated using a bilevel optimization framework. We further propose an effective and efficient instantiation of our approach, namely, Lattice Point Classifier (LPC), based on structured sparse coding in the permutohedral lattice and 2D convolutional neural networks (CNNs) that is end-to-end trainable. We demonstrate state-of-the-art robust point cloud classification performance on ModelNet40 and ScanNet under seven different attackers. For instance, we achieve 89.51% and 83.16% test accuracy on each dataset under the recent JGBA attacker that outperforms DUP-Net and IF-Defense with PointNet by ~70%. Demo code is available at https://zhang-vislab.github.io.

Key facts. Cited by 17 papers in the map (data through 14 Oct 2026). Builds on 49 of its 75 references that are in the map. First posted on arXiv 29 Mar 2022.

Builds on

49 of 75 references.

References · 49

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  4. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  5. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,442
  6. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  7. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,401
  8. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,117
  9. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,223
  10. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,268
  11. 2017Ensemble Adversarial Training: Attacks and Defensescited by 840
  12. 2018PointCNN: Convolution On X-Transformed Pointscited by 764
37 more
  1. 2020Convolutional Occupancy Networkscited by 761
  2. 2017OptNet: Differentiable Optimization as a Layer in Neural Networkscited by 599
  3. 2015Multi-view Convolutional Neural Networks for 3D Shape Recognitioncited by 565
  4. 2019Deep Learning for 3D Point Clouds: A Surveycited by 558
  5. 2019Meta-Learning with Implicit Gradientscited by 525
  6. 2019Differentiable Convex Optimization Layerscited by 465
  7. 2019Meta-Learning with Differentiable Convex Optimizationcited by 461
  8. 2016OctNet: Learning Deep 3D Representations at High Resolutionscited by 342
  9. 2020Point-GNN: Graph Neural Network for 3D Object Detection in a Point Cloudcited by 335
  10. 2019Simple Black-box Adversarial Attackscited by 251
  11. 2016Volumetric and Multi-View CNNs for Object Classification on 3D Datacited by 240
  12. 2020PointASNL: Robust Point Clouds Processing using Nonlocal Neural Networks with Adaptive Samplingcited by 239
  13. 2018SPLATNet: Sparse Lattice Networks for Point Cloud Processingcited by 158
  14. 2018Generating 3D Adversarial Point Cloudscited by 144
  15. 2020Physically Realizable Adversarial Examples for LiDAR Object Detectioncited by 137
  16. 2018PointCloud Saliency Mapscited by 112
  17. 2019HPLFlowNet: Hierarchical Permutohedral Lattice FlowNet for Scene Flow Estimation on Large-scale Point Cloudscited by 109
  18. 2019PU-GCN: Point Cloud Upsampling using Graph Convolutional Networkscited by 101
  19. 2021Robust Point Cloud Registration Framework Based on Deep Graph Matchingcited by 97
  20. 2019Extending Adversarial Attacks and Defenses to Deep 3D Point Cloud Classifierscited by 92
  21. 2019AdvPC: Transferable Adversarial Perturbations on 3D Point Cloudscited by 84
  22. 2019Deep Declarative Networks: A New Hopecited by 78
  23. 2021PointCutMix: Regularization Strategy for Point Cloud Classificationcited by 69
  24. 2019Adversarial Attack and Defense on Point Setscited by 63
  25. 2016A survey of sparse representation: algorithms and applicationscited by 58
  26. 2019Robustness of 3D Deep Learning in an Adversarial Settingcited by 55
  27. 2020LG-GAN: Label Guided Adversarial Network for Flexible Targeted Attack of Point Cloud-based Deep Networkscited by 55
  28. 2021A Backdoor Attack against 3D Point Cloud Classifierscited by 53
  29. 2021PointGuard: Provably Robust 3D Point Cloud Classificationcited by 49
  30. 2020On Isometry Robustness of Deep 3D Point Cloud Models under Adversarial Attackscited by 46
  31. 2020IF-Defense: 3D Adversarial Point Cloud Defense via Implicit Function based Restorationcited by 46
  32. 2019Point Cloud Oversegmentation with Graph-Structured Deep Metric Learningcited by 42
  33. 2020Minimal Adversarial Examples for Deep Learning on 3D Point Cloudscited by 41
  34. 2020Learning to Segment 3D Point Clouds in 2D Image Spacecited by 24
  35. 2021LatticeNet: Fast Spatio-Temporal Point Cloud Segmentation Using Permutohedral Latticescited by 11
  36. 2021The art of defense: letting networks fool the attackercited by 6
  37. 2020Stereo Frustums: A Siamese Pipeline for 3D Object Detectioncited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 4; 2023: 3; 2024: 5; 2025: 3; 2026: 2.

17 papers.

Sort:

  1. 2022Point Cloud Attacks in Graph Spectral Domain: When 3D Geometry Meets Graph Signal Processingcited by 16
  2. 2022PointDP: Diffusion-driven Purification against Adversarial Attacks on 3D Point Cloud Recognitioncited by 16
  3. 2022Learning Sub-Pixel Disparity Distribution for Light Field Depth Estimationcited by 12
  4. 20233DHacker: Spectrum-based Decision Boundary Generation for Hard-label 3D Point Cloud Attackcited by 9
  5. 2022Ada3Diff: Defending against 3D Adversarial Point Clouds via Adaptive Diffusioncited by 8
  6. 2023PointCert: Point Cloud Classification with Deterministic Certified Robustness Guaranteescited by 6
  7. 2023Adversarial Attacks and Defenses on 3D Point Cloud Classification: A Surveycited by 5
  8. 2024Towards Robust 3D Pose Transfer with Adversarial Learningcited by 4
  9. 2025PF3Det: A Prompted Foundation Feature Assisted Visual LiDAR 3D Detectorcited by 4
  10. 2024PNeRV: A Polynomial Neural Representation for Videoscited by 2
  11. 2024Robust 3D Point Clouds Classification based on Declarative Defenderscited by 2
  12. 2026Privacy-Concealing Cooperative Perception for BEV Scene Segmentationcited by 2
5 more
  1. 2024Improving Adversarial Robustness for 3D Point Cloud Recognition at Test-Time through Purified Self-Trainingcited by 0
  2. 2024Hard-Label Black-Box Attacks on 3D Point Cloudscited by 0
  3. 2025Robust Unsupervised Domain Adaptation for 3D Point Cloud Segmentation Under Source Adversarial Attackscited by 0
  4. 2025Certified L2-Norm Robustness of 3D Point Cloud Recognition in the Frequency Domaincited by 0
  5. 2026PWAVEP: Purifying Imperceptible Adversarial Perturbations in 3D Point Clouds via Spectral Graph Waveletscited by 0