[Recruitment]
Associate/Assistant Professor in Artificial Intelligence for Space-Enabled Technologies, Durham University

We are looking for applicants in Artificial Intelligence, Computer Vision, Edge Computing, Digital Twins, Human Computer Interaction, User Modelling, Robotics or Resilient Computing with potentials/achievements in informing space applications.

The post hoder will enjoy 1) a permanent (equivalent to US tenured) position at a top 100 university, 2) significantly reduced teaching, 3) a fully-funded PhD, 4) travel budget, 5) chance for a 2-year fully-funded Post-Doc.

Hard No-Box Adversarial Attack on Skeleton-Based Human Action Recognition with Skeleton-Motion-Informed Gradient

Zhengzhi Lu, He Wang, Ziyi Chang, Guoan Yang and Hubert P. H. Shum
Proceedings of the 2023 IEEE/CVF International Conference on Computer Vision (ICCV), 2023

Core A* Conference H5-Index: 228# Core A* Conference

Hard No-Box Adversarial Attack on Skeleton-Based Human Action Recognition with Skeleton-Motion-Informed Gradient
‡ According to Core Ranking 2023"

Abstract

Recently, methods for skeleton-based human activity recognition have been shown to be vulnerable to adversarial attacks. However, these attack methods require either the full knowledge of the victim (i.e. white-box attacks), access to training data (i.e. transfer-based attacks) or frequent model queries (i.e. black-box attacks). All their requirements are highly restrictive, raising the question of how detrimental the vulnerability is. In this paper, we show that the vulnerability indeed exists. To this end, we consider a new attack task: the attacker has no access to the victim model or the training data or labels, where we coin the term hard no-box attack. Specifically, we first learn a motion manifold where we define an adversarial loss to compute a new gradient for the attack, named skeleton-motion-informed (SMI) gradient. Our gradient contains information of the motion dynamics, which is different from existing gradient-based attack methods that compute the loss gradient assuming each dimension in the data is independent. The SMI gradient can augment many gradient-based attack methods, leading to a new family of no-box attack methods. Extensive evaluation and comparison show that our method imposes a real threat to existing classifiers. They also show that the SMI gradient improves the transferability and imperceptibility of adversarial samples in both no-box and transfer-based black-box settings.

Downloads

YouTube

Citations

BibTeX

@inproceedings{lu23hard,
 author={Lu, Zhengzhi and Wang, He and Chang, Ziyi and Yang, Guoan and Shum, Hubert P. H.},
 booktitle={Proceedings of the 2023 IEEE/CVF International Conference on Computer Vision},
 series={ICCV '23},
 title={Hard No-Box Adversarial Attack on Skeleton-Based Human Action Recognition with Skeleton-Motion-Informed Gradient},
 year={2023},
 month={10},
 pages={4574--4583},
 numpages={10},
 doi={10.1109/ICCV51070.2023.00424},
 publisher={IEEE/CVF},
 location={Paris, France},
}

RIS

TY  - CONF
AU  - Lu, Zhengzhi
AU  - Wang, He
AU  - Chang, Ziyi
AU  - Yang, Guoan
AU  - Shum, Hubert P. H.
T2  - Proceedings of the 2023 IEEE/CVF International Conference on Computer Vision
TI  - Hard No-Box Adversarial Attack on Skeleton-Based Human Action Recognition with Skeleton-Motion-Informed Gradient
PY  - 2023
Y1  - 10 2023
SP  - 4574
EP  - 4583
DO  - 10.1109/ICCV51070.2023.00424
PB  - IEEE/CVF
ER  - 

Plain Text

Zhengzhi Lu, He Wang, Ziyi Chang, Guoan Yang and Hubert P. H. Shum, "Hard No-Box Adversarial Attack on Skeleton-Based Human Action Recognition with Skeleton-Motion-Informed Gradient," in ICCV '23: Proceedings of the 2023 IEEE/CVF International Conference on Computer Vision, pp. 4574-4583, Paris, France, IEEE/CVF, Oct 2023.

Supporting Grants

Similar Research

Qianhui Men, Edmond S. L. Ho, Hubert P. H. Shum and Howard Leung, "Focalized Contrastive View-Invariant Learning for Self-Supervised Skeleton-Based Action Recognition", Neurocomputing, 2023
Ying Huang, Hubert P. H. Shum, Edmond S. L. Ho and Nauman Aslam, "High-Speed Multi-Person Pose Estimation with Deep Feature Transfer", Computer Vision and Image Understanding (CVIU), 2020
Jingtian Zhang, Hubert P. H. Shum, Jungong Han and Ling Shao, "Action Recognition from Arbitrary Views Using Transferable Dictionary Learning", IEEE Transactions on Image Processing (TIP), 2018
Jingtian Zhang, Lining Zhang, Hubert P. H. Shum and Ling Shao, "Arbitrary View Action Recognition via Transfer Dictionary Learning on Synthetic Training Data", Proceedings of the 2016 IEEE International Conference on Robotics and Automation (ICRA), 2016
Meng Li, Howard Leung and Hubert P. H. Shum, "Human Action Recognition via Skeletal and Depth Based Feature Fusion", Proceedings of the 2016 ACM International Conference on Motion in Games (MIG), 2016
Zheming Zuo, Daniel Organisciak, Hubert P. H. Shum and Longzhi Yang, "Saliency-Informed Spatio-Temporal Vector of Locally Aggregated Descriptors and Fisher Vectors for Visual Action Recognition", Proceedings of the 2018 British Machine Vision Conference Workshop on Image Analysis for Human Facial and Activity Recognition (IAHFAR), 2018

 

 

Last updated on 17 February 2024
RSS Feed