湖南大学学报（自然科学版）2024，Vol.51Issue(12)：129-138,10.DOI:10.16339/j.cnki.hdxbzkb.2024290
基于特征交互模块增强RGB-骨骼动作识别鲁棒性研究
Study on Enhancing the Robustness of RGB-skeleton Action Recognition Based on the Feature Interaction Module
摘要
Abstract
Malicious attackers can easily deceive neural networks by adding human-imperceptible adversarial noise to natural samples,leading to misclassification.To enhance the model's robustness against such adversarial perturbations,previous research has predominantly concentrated on the robustness of single-modal tasks,with insufficient exploration of multimodal scenarios.Therefore,this paper aims to improve the robustness of multimodal RGB-skeleton action recognition and introduces a robust action recognition framework based on a Feature Interaction Module(FIM),which extracts global information from adversarial samples to learn inter-modal joint representations for calibrating multi-modal features.A corresponding loss function tailored to this framework is also developed.Experimental results demonstrate that against CW attack,our method achieves a RI of 25.14%and an average robust accuracy of 48.99%on the NTURGB+D dataset,outperforming the latest SimMin+ExFMem method by 8.55 and 23.79 percentage points,respectively.These findings confirm that our approach surpasses others in enhancing robustness and balancing accuracy rates.关键词
计算机视觉/多模态/RGB-骨骼动作识别/对抗训练Key words
computer vision/multimodal/RGB-skeleton action recognition/adversarial training分类
信息技术与安全科学引用本文复制引用
侯永宏,刘超,刘鑫,岳焕景,杨敬钰..基于特征交互模块增强RGB-骨骼动作识别鲁棒性研究[J].湖南大学学报（自然科学版）,2024,51(12):129-138,10.基金项目
国家自然科学基金资助项目(62072331),National Natural Science Foundation of China(62072331) （62072331）
国家自然科学基金资助项目(62231018),National Natural Science Foundation of China(62231018) （62231018）
国家自然科学基金资助项目(62171309),National Natural Science Foundation of China(62171309) （62171309）
