mlsanityData to 7 Oct 2026

Be Careful with Rotation: A Uniform Backdoor Pattern for 3D Shape

arXiv 2211.16192 ↗First posted 28 Nov 2022cs.CV cs.CRcited by 4

Not on the map. Why?

Abstract (arXiv)

For saving cost, many deep neural networks (DNNs) are trained on third-party datasets downloaded from internet, which enables attacker to implant backdoor into DNNs. In 2D domain, inherent structures of different image formats are similar. Hence, backdoor attack designed for one image format will suite for others. However, when it comes to 3D world, there is a huge disparity among different 3D data structures. As a result, backdoor pattern designed for one certain 3D data structure will be disable for other data structures of the same 3D scene. Therefore, this paper designs a uniform backdoor pattern: NRBdoor (Noisy Rotation Backdoor) which is able to adapt for heterogeneous 3D data structures. Specifically, we start from the unit rotation and then search for the optimal pattern by noise generation and selection process. The proposed NRBdoor is natural and imperceptible, since rotation is a common operation which usually contains noise due to both the miss match between a pair of points and the sensor calibration error for real-world 3D scene. Extensive experiments on 3D mesh and point cloud show that the proposed NRBdoor achieves state-of-the-art performance, with negligible shape variation.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 37 of its 60 references that are in the map. First posted on arXiv 28 Nov 2022.

Builds on

37 of 60 references.

References · 37

  1. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  2. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  3. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
  4. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  5. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  6. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  7. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,220
  8. 2020PCT: Point cloud transformercited by 730
  9. 2021Point-BERT: Pre-training 3D Point Cloud Transformers with Masked Point Modelingcited by 647
  10. 2020Backdoor Learning: A Surveycited by 412
  11. 2020Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networkscited by 341
  12. 2020Invisible Backdoor Attack with Sample-Specific Triggerscited by 338
25 more
  1. 2020DiffusionNet: Discretization Agnostic Learning on Surfacescited by 163
  2. 2020Clean-Label Backdoor Attacks on Video Recognition Modelscited by 162
  3. 2018SentiNet: Detecting Localized Universal Attacks Against Deep Learning Systemscited by 148
  4. 2020Backdoor Attacks Against Deep Learning Systems in the Physical Worldcited by 129
  5. 2018PointCloud Saliency Mapscited by 112
  6. 2022Shadows can be Dangerous: Stealthy and Effective Physical-world Adversarial Attack by Natural Phenomenoncited by 101
  7. 2018MeshNet: Mesh Neural Network for 3D Shape Representationcited by 95
  8. 2019AdvPC: Transferable Adversarial Perturbations on 3D Point Cloudscited by 84
  9. 2021Backdoor Attack in the Physical Worldcited by 82
  10. 2019End-to-End Learnable Geometric Vision by Backpropagating PnP Optimizationcited by 65
  11. 2019Adversarial Attack and Defense on Point Setscited by 63
  12. 2021Subdivision-Based Mesh Convolution Networkscited by 60
  13. 2021A Backdoor Attack against 3D Point Cloud Classifierscited by 50
  14. 2019NeuronInspect: Detecting Backdoors in Neural Networks via Output Explanationscited by 49
  15. 2020Universal Adversarial Attack on Attention and the Resulting Dataset DAmageNetcited by 45
  16. 2021PointBA: Towards Backdoor Attacks in 3D Point Cloudcited by 34
  17. 2020MeshWalker: Deep Mesh Understanding by Random Walkscited by 32
  18. 2021Deep 3D-to-2D Watermarking: Embedding Messages in 3D Meshes and Extracting Them from 2D Renderingscited by 23
  19. 2019Man-in-the-Middle Attacks against Machine Learning Classifiers via Malicious Generative Modelscited by 19
  20. 2019Taking Care of The Discretization Problem: A Comprehensive Study of the Discretization Problem and A Black-Box Adversarial Attack in Discrete Integer Domaincited by 14
  21. 2020PointINet: Point Cloud Frame Interpolation Networkcited by 12
  22. 2021Detecting Backdoor Attacks Against Point Cloud Classifierscited by 9
  23. 2019Adversarial Examples Versus Cloud-based Detectors: A Black-box Empirical Studycited by 8
  24. 2021Poisoning MorphNet for Clean-Label Backdoor Attack to Point Cloudscited by 7
  25. 2022Neighboring Backdoor Attacks on Graph Convolutional Networkcited by 7

Built on by

Cited by year

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

4 papers.

  1. 2024PointNCBW: Towards Dataset Ownership Verification for Point Clouds via Negative Clean-label Backdoor Watermarkcited by 19
  2. 2025GaussTrap: Stealthy Poisoning Attacks on 3D Gaussian Splatting for Targeted Scene Confusioncited by 6
  3. 2024iBA: Backdoor Attack on 3D Point Cloud via Reconstructing Itselfcited by 4
  4. 2024Stealthy and Robust Backdoor Attack against 3D Point Clouds through Additional Point Featurescited by 0