Research map / Adversarial robustness and security / Black-box and transfer attacks
Hard No-Box Adversarial Attack on Skeleton-Based Human Action Recognition with Skeleton-Motion-Informed Gradient
Zhengzhi Lu, He Wang, Ziyi Chang, Guoan Yang, Hubert P. H. Shum
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add second-order (acceleration) SMI gradient to I-FGSM (ablation) | no-box attack on NTU60 · NTU60 | perceptual deviation: 95.87 -> 65.60 | Improved |
| use second-order SMI instead of first-order SMI (ablation) | no-box attack on NTU60 · NTU60 | perceptual deviation: 84.63 -> 65.60 | Improved |
| add SMI gradient to SMART transfer attack (S2MI-FGSM) | SMART transferred black-box attack · NTU60, surrogate MS-G3D, victim STGCN | fooling rate (%): 1.71% -> 2.67% | Improved |
| add momentum (MI-FGSM) on skeleton data (ablation) | I-FGSM · NTU60 transferred SMART attack | fooling rate (%): 2.10% -> 2.10% | No clear effect |
| add momentum (MI-FGSM) on skeleton data (ablation) | I-FGSM · NTU60 no-box attack | perceptual deviation: 95.87 -> 131.77 | Hurt |
| use S2MI-FGSM instead of S2I-FGSM (ablation) | SMART attack · NTU60, victim MS-G3D | perceptual deviation: 1.69 -> 3.02 | Hurt |
| use multi-stream 2s-AGCN victim instead of single-stream js-AGCN | victim model · NTU datasets | fooling rate | Hurt |
Closest papers by meaning
- BASAR:Black-Box Attack on Skeletal Action Recognition (CVPR 2021 · similarity 0.92)
- Adversarial Bone Length Attack on Action Recognition (AAAI 2022 · similarity 0.91)
- Understanding the Robustness of Skeleton-Based Action Recognition Under Adversarial Attack (CVPR 2021 · similarity 0.91)
- TASAR: Transfer-based Attack on Skeletal Action Recognition (ICLR 2025 · similarity 0.88)
- Neuron: Learning Context-Aware Evolving Representations for Zero-Shot Skeleton Action Recognition (CVPR 2025 · similarity 0.84)
- Skeleton-based Action Recognition with Non-linear Dependency Modeling and Hilbert-Schmidt Independence Criterion (AAAI 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
