计算机应用研究2023，Vol.40Issue(12)：3848-3853,6.DOI:10.19734/j.issn.1001-3695.2023.05.0216
特征采样运动信息增强的动作识别方法
Action recognition method with feature sampling and motion information enhancement
摘要
Abstract
Based on deep models,video action recognition typically involves sampling the input video and then extracting fea-tures from the obtained video frames to classify actions.Therefore,the video frame sampling method directly affects the effec-tiveness of action recognition.Aiming to sample key and effective features while enhanced the motion information in videos,this paper proposed a LGMeNet based on a feature-level sampling strategy.Firstly,it used a feature-level sampling module to uni-formly select frames with the same motion information from the input data.Secondly,it employed a local motion feature extrac-tion module to compute short-term motion features using a similarity function.Finally,it utilized a LSTM network in the global motion feature extraction module to calculate multi-scale long-term motion features.Experimental evaluations show that LGMeNet achieves accuracies of 97.7％and 56.9％on the UCF101 and Something-SomethingV1 datasets,respectively.The results of this study demonstrate the effectiveness of LGMeNet in enhancing action recognition and highlight its significance for further advancements in related research areas.关键词
深度学习/动作识别/视频采样/时间建模Key words
deep learning/action recognition/video sampling/temporal modeling分类
信息技术与安全科学引用本文复制引用
罗会兰,包中生..特征采样运动信息增强的动作识别方法[J].计算机应用研究,2023,40(12):3848-3853,6.基金项目
国家自然科学基金资助项目(61862031) （61862031）
江西省主要学科技术带头人领军人才计划资助项目(20213BCJ22004) （20213BCJ22004）
江西省学位与研究生教育教学改革研究重点项目(JXYJG-2020-120) （JXYJG-2020-120）
