Skip to main content
SpringerLink
Log in

PAGN: perturbation adaption generation network for point cloud adversarial defense

  • Regular Paper
  • Published:
Multimedia Systems Aims and scope Submit manuscript

Abstract

With the development of point cloud processing technology, the point cloud has been applied to many applications such as autonomous driving, scene reconstruction, and object detection. These applications are safety-critical applications and very important to our life. However, the research of 3D model adversarial attacks got rare attention in recent years. It is important to study how to improve the adversarial robustness of the 3D data processing model based on adversarial training. In this paper, we propose a novel perturbation adaption generation network for point cloud adversarial defense, which can adaptively generate adversarial point cloud samples according to the point cloud to improve the robustness of the deep learning model for the 3D model classification. Specifically, we apply the content similarity measure method to avoid the over perturbation to guarantee that the distortion of the adversarial sample is within an acceptable range. Finally, extensive experiments based on different attack methods on ModelNet40 demonstrate the effectiveness of our approach.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2

Similar content being viewed by others

References

  1. Achlioptas, P., Diamanti, O., Mitliagkas, I., Guibas, L.J.: Learning representations and generative models for 3d point clouds. In: ICML 2018, Stockholmsmässan, Stockholm, Sweden, July 10–15, 2018, pp. 40–49 (2018)

  2. Alexiou, E., Ebrahimi, T.: Point cloud quality assessment metric based on angular similarity. In: ICME 2018, pp. 1–6 (2018)

  3. Dong, X., Chen, D., Zhou, H., Hua, G., Zhang, W., Yu, N.: Self-robust 3D point recognition via gather-vector guidance. In: CVPR 2020, Seattle, WA, USA, June 13–19, 2020, pp. 11513–11521. IEEE (2020)

  4. Du, Q.: 3D point cloud registration denoising method for human motion image using deep learning algorithm. Multim. Syst. 26(1), 75–82 (2020)

    Article  Google Scholar 

  5. Gao, Y., Dai, Q.: View-based 3D object retrieval: challenges and approaches. IEEE Multim. 21(3), 52–57 (2014)

    Article  Google Scholar 

  6. Klokov, R., Boyer, E., Verbeek, J.: Discrete point flow networks for efficient point cloud generation. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J. (eds.) ECCV 2020—16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXIII, Lecture Notes in Computer Science, vol. 12368, pp. 694–710. Springer (2020)

  7. Klokov, R., Lempitsky, V.S.: Escape from cells: deep KD-networks for the recognition of 3D point cloud models. In: ICCV 2017, Venice, Italy, October 22–29, 2017, pp. 863–872 (2017)

  8. Kong, L., Dai, R.: Object-detection-based video compression for wireless surveillance systems. IEEE Multim. 24(2), 76–85 (2017)

    Article  Google Scholar 

  9. Kumawat, S., Raman, S.: LP-3DCNN: unveiling local phase in 3D convolutional neural networks. In: CVPR 2019, Long Beach, CA, USA, June 16–20, 2019, pp. 4903–4912 (2019)

  10. Li, C., Zaheer, M., Zhang, Y., Póczos, B., Salakhutdinov, R.: Point cloud GAN. In: ICLR 2019 Workshop, New Orleans, Louisiana, United States, May 6, 2019 (2019)

  11. Liu, D., Yu, R., Su, H.: Extending adversarial attacks and defenses to deep 3D point cloud classifiers. In: ICIP 2019, Taipei, Taiwan, September 22–25, 2019, pp. 2279–2283. IEEE (2019)

  12. Liu, X., Li, Y., Wu, C., Hsieh, C.: ADV-BNN: improved adversarial defense through robust Bayesian neural network. In: ICLR 2019, New Orleans, LA, USA, May 6–9, 2019. OpenReview.net (2019)

  13. Liu, Y., Fan, B., Xiang, S., Pan, C.: Relation-shape convolutional neural network for point cloud analysis. In: CVPR 2019, Long Beach, CA, USA, June 16–20, 2019, pp. 8895–8904 (2019)

  14. Liu, Z., Wang, Z., Ma, C., Zhang, C., Mitani, J., Fukui, Y.: Shape alignment and shape orientation analysis-based 3D shape retrieval system. Multim. Syst. 16(4–5), 319–333 (2010)

    Article  Google Scholar 

  15. Nie, W., Wang, K., Liang, Q., He, R.: Panorama based on multi-channel-attention CNN for 3D model recognition. Multim. Syst. 25(6), 655–662 (2019)

    Article  Google Scholar 

  16. Perez, E., Strub, F., de Vries, H., Dumoulin, V., Courville, A.C.: Film: visual reasoning with a general conditioning layer. In: S.A. McIlraith, K.Q. Weinberger (eds.) (AAAI-18), New Orleans, Louisiana, USA, February 2–7, 2018, pp. 3942–3951. AAAI Press (2018)

  17. Qi, C.R., Su, H., Mo, K., Guibas, L.J.: Pointnet: deep learning on point sets for 3D classification and segmentation. In: CVPR 2017, pp. 652–660 (2017)

  18. Qi, C.R., Yi, L., Su, H., Guibas, L.J.: Pointnet++: deep hierarchical feature learning on point sets in a metric space. In: Advances in neural information processing systems, pp. 5099–5108 (2017)

  19. Sun, Y., Wang, Y., Liu, Z., Siegel, J.E., Sarma, S.E.: Pointgrow: autoregressively learned point cloud generation with self-attention. In: WACV 2020, Snowmass Village, CO, USA, March 1–5, 2020, pp. 61–70. IEEE (2020)

  20. Szegedy, C., Zaremba, W., Sutskever, I., Bruna, J., Erhan, D., Goodfellow, I.J., Fergus, R.: Intriguing properties of neural networks. In: ICLR 2014, Banff, AB, Canada, April 14-16, 2014, Conference Track Proceedings (2014)

  21. Tramèr, F., Kurakin, A., Papernot, N., Goodfellow, I.J., Boneh, D., McDaniel, P.D.: Ensemble adversarial training: Attacks and defenses. In: ICLR 2018, Vancouver, BC, Canada, April 30–May 3, 2018, Conference Track Proceedings. OpenReview.net (2018)

  22. Wang, L., He, L., Mishra, A., Li, C.: Active contours driven by local Gaussian distribution fitting energy. Signal Process. 89(12), 2435–2447 (2009)

    Article  Google Scholar 

  23. Wang, Y., Sun, Y., Liu, Z., Sarma, S.E., Bronstein, M.M., Solomon, J.M.: Dynamic graph CNN for learning on point clouds. ACM Trans. Graph 38(5), 146:1–146:12 (2019)

  24. Wu, Z., Song, S., Khosla, A., Yu, F., Zhang, L., Tang, X., Xiao, J.: 3D shapenets: a deep representation for volumetric shapes. In: CVPR 2015, Boston, MA, USA, June 7–12, 2015, pp. 1912–1920 (2015)

  25. Xiang, C., Qi, C.R., Li, B.: Generating 3D adversarial point clouds. In: CVPR 2019, Long Beach, CA, USA, June 16–20, 2019, pp. 9136–9144 (2019)

  26. Yang, G., Huang, X., Hao, Z., Liu, M., Belongie, S.J., Hariharan, B.: Pointflow: 3D point cloud generation with continuous normalizing flows. In: ICCV 2019, Seoul, Korea (South), October 27–November 2, 2019, pp. 4540–4549. IEEE (2019)

  27. Zhao, Y., Wu, Y., Chen, C., Lim, A.: On isometry robustness of deep 3D point cloud models under adversarial attacks. In: CVPR 2020, Seattle, WA, USA, June 13–19, 2020, pp. 1198–1207. IEEE (2020)

  28. Zheng, T., Chen, C., Yuan, J., Li, B., Ren, K.: Pointcloud saliency maps. In: ICCV 2019, Seoul, Korea (South), October 27–November 2, 2019, pp. 1598–1606 (2019)

  29. Zhou, H., Chen, D., Liao, J., Chen, K., Dong, X., Liu, K., Zhang, W., Hua, G., Yu, N.: Lg-gan: label guided adversarial network for flexible targeted attack of point cloud based deep networks. In: CVPR 2020, pp. 10353–10362 (2020)

  30. Zhou, H., Chen, K., Zhang, W., Fang, H., Zhou, W., Yu, N.: Dup-net: Denoiser and upsampler network for 3d adversarial point clouds defense. In: ICCV 2019, Seoul, Korea (South), October 27–November 2, 2019, pp. 1961–1970. IEEE (2019)

Download references

Acknowledgements

This work was supported in part by the National Key Research and Development Program of China (2020YFB1711704), the National Natural Science Foundation of China (61772359, 61572356, 61872267, 61502477, 61471263), the grant of Tianjin New Generation Artificial Intelligence Major Program (18ZXZNGX00150, 19ZXZNGX00110), the Open Project Program of the State Key Lab of CAD & CG, Zhejiang University (Grant No. A2005, A2012) and the Tianjin Science Foundation for Young Scientists (19JCQNJC00500).

Author information

Authors and Affiliations

  1. The School of Microelectronics, Tianjin University, Tianjin, 300072, China

    Qi Liang & Qiang Li

  2. The School of Electrical and Information Engineering, Tianjin University, Tianjin, 300072, China

    Weizhi Nie & An-An Liu

Authors
  1. Qi Liang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Qiang Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Weizhi Nie
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. An-An Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Weizhi Nie.

Additional information

Communicated by B.-K. Bao.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Liang, Q., Li, Q., Nie, W. et al. PAGN: perturbation adaption generation network for point cloud adversarial defense. Multimedia Systems 28, 851–859 (2022). https://doi.org/10.1007/s00530-022-00887-w

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00530-022-00887-w

Keywords

Search

Navigation