针对人体姿态模型设计新攻击，用关键点相似度衡量破坏效果。
A Black-Box Adversarial Attack on Human Pose Estimation and Keypoint-Based Action Recognition Models
- 用关键点相似度（OKS）作为反馈信号，直接攻击姿态结构。
- 在Penn Action数据集上使姿态质量下降0.08至0.15，动作识别准确率降6.18%~13.86%。
- 适用于主流姿态估计算法，对部署安全有警示意义。
人体姿态估计与基于关键点的动作识别模型正广泛应用于视频理解流程，但其对抗攻击脆弱性仍研究不足。以往时序一致的黑盒攻击多依赖边界框重叠度量（如IoU），然而姿态估计输出为关键点配置而非包围框，导致框级相似度难以有效评估姿态退化。本文提出OKS Attack，一种基于决策的黑盒攻击方法，以物体关键点相似度（OKS）作为攻击反馈信号，直接针对人体姿态的空间结构。在Penn Action数据集上的实验表明，OKS Attack能持续降低各类姿态估计算法的性能，平均OKS下降0.0802至0.1494。下游跨数据集动作识别评估中，攻击使准确率下降6.18至13.86个百分点，显著优于匹配查询的随机噪声扰动。该攻击对自顶向下与单阶段姿态估计算法均有效。源代码将公开于https://github.com/KacperM33/OKS_attack。
原文摘要 · Abstract (English)
Human pose estimation and keypoint-based action recognition models are increasingly deployed as components of video understanding pipelines, yet their vulnerability to adversarial attacks remains insufficiently studied. Temporally coherent black-box attacks have been previously studied in visual object tracking, where the attack feedback can be defined using bounding-box overlap measures such as Intersection over Union (IoU). However, human pose estimation produces keypoint configurations rather than enclosing boxes, making box-level similarity poorly suited for measuring pose degradation. We propose OKS Attack, a decision-based black-box attack that uses Object Keypoint Similarity (OKS) as the attack feedback signal, directly targeting the spatial structure of human poses rather than their enclosing boxes. Experiments on the Penn Action dataset show that OKS Attack consistently reduces pose quality across evaluated pose estimators, with mean OKS decreases ranging from 0.0802 to 0.1494. In a downstream cross-dataset action-recognition evaluation, the attack reduces accuracy by 6.18 to 13.86 percentage points and outperforms query-matched random-noise perturbations. The attack is effective across both top-down and single-stage pose estimation models. The source code will be made publicly available at https://github.com/KacperM33/OKS_attack
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。