Manifold Constraints for Imperceptible Adversarial Attacks on Point Clouds
Proceedings of the AAAI Conference on Artificial Intelligence · 24.03.2024 · 38 (6) · 5127-5135
DOI: 10.1609/aaai.v38i6.28318 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 24.03.2024. Verfasst von Keke Tang und 8 weiteren Autor:innen. Die Arbeit wurde bisher 17-mal zitiert (Crossref).
Abstract (Originalsprache)
Adversarial attacks on 3D point clouds often exhibit unsatisfactory imperceptibility, which primarily stems from the disregard for manifold-aware distortion, i.e., distortion of the underlying 2-manifold surfaces. In this paper, we develop novel manifold constraints to reduce such distortion, aiming to enhance the imperceptibility of adversarial attacks on 3D point clouds. Specifically, we construct a bijective manifold mapping between point clouds and a simple parameter shape using an invertible auto-encoder. Consequently, manifold-aware distortion during attacks can be captured within the parameter space. By enforcing manifold constraints that preserve local properties of the parameter shape, manifold-aware distortion is effectively mitigated, ultimately leading to enhanced imperceptibility. Extensive experiments demonstrate that integrating manifold constraints into conventional adversarial attack solutions yields superior imperceptibility, outperforming the state-of-the-art methods.
Zitieren
APA
Tang, K., He, X., Peng, W., Wu, J., Shi, Y., Liu, D., Zhou, P., Wang, W., & Tian, Z. (2024). Manifold Constraints for Imperceptible Adversarial Attacks on Point Clouds. Proceedings of the AAAI Conference on Artificial Intelligence, 38(6), 5127-5135. https://doi.org/10.1609/aaai.v38i6.28318
Vancouver
Tang K, He X, Peng W, Wu J, Shi Y, Liu D, et al. Manifold Constraints for Imperceptible Adversarial Attacks on Point Clouds. Proceedings of the AAAI Conference on Artificial Intelligence. 2024;38(6):5127-5135. doi:10.1609/aaai.v38i6.28318
BibTeX
@article{Tang2024,
author = {Tang, Keke and He, Xu and Peng, Weilong and Wu, Jianpeng and Shi, Yawen and Liu, Daizong and Zhou, Pan and Wang, Wenping and Tian, Zhihong},
title = {Manifold Constraints for Imperceptible Adversarial Attacks on Point Clouds},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2024},
volume = {38},
number = {6},
pages = {5127--5135},
doi = {10.1609/aaai.v38i6.28318}
}
EUVIMED