江苏大学学报（自然科学版）2026，Vol.47Issue(1)：39-47,63,10.DOI:10.3969/j.issn.1671-7775.2026.01.006
基于线性注意和类别关联特征学习的在线动作检测
Online action detection based on linear attention and category association feature learning
摘要
Abstract
To fully and reasonably utilize the contextual features of actions,category-related features and predicted future features for rapid action detection in online action detection,the online action detection method based on linear attention and category association feature learning was proposed.The Transformer architecture was improved by employing lightweight linear self-attention based on the Hadamard product to reduce computational cost for video contextual feature learning.The action features from training samples were clustered to associate the video sequence context with action category features for achieving the effective learning of category-associated feature representations.By integrating contextual features,category-associated features and predicted future features,the action discrimination at corresponding moments was enhanced.The performance experiments were conducted on typical datasets to realize the hyperparameter selection analysis,and the working accuracy and operational efficiency of different methods were compared.The ablation experiments and visualization analysis were provided.The results show that on the Thumos14(TSN-Anet),Thumos14(TSN-Kinetics)and HDD datasets,the mAP values of the proposed method are respectively improved by 0.2,0.5 and 0.2 percentage point compared with the Colar method,which indicates that the new method outperforms the currently advanced Colar method.关键词
在线动作检测/深度学习/注意力机制/编码/上下文特征/Transformer/类别关联特征学习Key words
online action detection/deep learning/attention mechanism/encoding/contextual features/Transformer/category association feature learning分类
信息技术与安全科学引用本文复制引用
詹永照,孙慧敏,夏惠芬,任晓鹏..基于线性注意和类别关联特征学习的在线动作检测[J].江苏大学学报（自然科学版）,2026,47(1):39-47,63,10.基金项目
国家自然科学基金资助项目(61672268) （61672268）
