PulseAugur
EN
LIVE 12:11:45

New Attack Method Preserves Motion Quality in Skeleton Action Recognition

By PulseAugur Editorial · [4 sources] ·

Researchers have developed a new method for creating imperceptible adversarial attacks on skeleton-based human action recognition systems. This approach aims to preserve the naturalness of motion data, unlike previous methods that introduced noticeable perturbations. The new technique uses a distribution-based adversarial attack and a novel metric for evaluating motion quality, which is claimed to be more aligned with human perception. Experiments show this method is effective in both successfully attacking action recognizers and maintaining the quality of the altered motion data, raising concerns about the robustness of current systems. AI

IMPACT Highlights potential vulnerabilities in AI-driven action recognition systems, necessitating improved robustness and security measures.

RANK_REASON The cluster contains academic papers detailing novel research in adversarial attacks on skeleton-based human action recognition.

Read on Hugging Face Daily Papers →

AI-generated summary · Google Gemini · from 4 sources. How we write summaries →

New Attack Method Preserves Motion Quality in Skeleton Action Recognition

COVERAGE [4]

  1. Hugging Face Daily Papers TIER_1 English(EN) ·

    Quality-Preserving Imperceptible Adversarial Attack on Skeleton-based Human Action Recognition

    Adversarial attacks on skeletal human action recognition have received significant attention. However, existing methods typically introduce noise-like perturbations that degrade motion quality post-attack, and thereby are inherently perceptible with recent advancements in S-HAR s…

  2. arXiv cs.CV TIER_1 English(EN) · Ziyi Chang, Kanglei Zhou, Xiaohui Liang, Hubert P. H. Shum ·

    Quality-Preserving Imperceptible Adversarial Attack on Skeleton-based Human Action Recognition

    arXiv:2606.13022v1 Announce Type: new Abstract: Adversarial attacks on skeletal human action recognition have received significant attention. However, existing methods typically introduce noise-like perturbations that degrade motion quality post-attack, and thereby are inherently…

  3. arXiv cs.CV TIER_1 English(EN) · Hubert P. H. Shum ·

    Quality-Preserving Imperceptible Adversarial Attack on Skeleton-based Human Action Recognition

    Adversarial attacks on skeletal human action recognition have received significant attention. However, existing methods typically introduce noise-like perturbations that degrade motion quality post-attack, and thereby are inherently perceptible with recent advancements in S-HAR s…

  4. arXiv cs.CV TIER_1 English(EN) · Shengkai Sun, Zhiyong Cheng, Zefan Zhang, Jianfeng Dong, Zhihui Li, Meng Wang ·

    Exploring Adaptive Masked Reconstruction for Self-Supervised Skeleton-Based Action Recognition

    arXiv:2606.11450v1 Announce Type: new Abstract: Recently, masked skeleton reconstruction models have emerged as strong action representation learners, driving significant progress in self-supervised skeleton-based action recognition. However, existing state-of-the-art methods mus…

RELATED ENTITIES

RELATED TOPICS