基于相对骨骼点特征和时序自适应感受野的动作识别方法

针对长时间动作识别难以充分利用时空域信息的问题，提出了基于相对骨骼点特征和时序自适应感受野的动作识别方法。首先，该方法在特征获取部分增加了相对骨骼点特征，以满足节点多样性和互补性要求，将其分别输入到空域图卷积网络，获得空间中相邻关节聚合的局部特征。然后，设计了一个时序自适应感受野网络，以获取在时域中关节变化的局部特征，并且增加了网络对不同持续时长动作的适应性。最后，经过决策级融合模块，计算类别概率，得到分类结果。仿真结果表明，基于NTU RGB+D和Kinetics-skeleton两大基准数据集，对比多种主流方法，均取得了更高的识别准确率，分别为96.2%与60.1%。该方法可以较好地提取不同动作的区别性时间特征，提高了动作时空特征的判别能力。

Action Recognition Based on Relative Skeleton Point Features and Temporal Adaptive Receptive Field

Aimingattheproblemthatitisdifficulttomakefulluseoftemporalandspatialdomain informationforlong-termactionrecognition,anactionrecognitionmethodbasedonrelativeskele- tonpointfeaturesandtemporaladaptivereceptivefieldisproposed.Firstly,thismethodaddsrel- ativeskeletonpointfeaturesinthefeatureacquisitionparttomeetthenodediversityandcomple- mentarityrequirements,andrespectivelyinputsthemintothespatialgraphconvolutionnetwork toobtainthelocalfeaturesoftheadjacentjointaggregationinspace.Then,atime-seriesadaptive receptivefieldnetworkisdesignedtoobtainthelocalfeaturesofjointchangesinthetimedomain,andtoincreasetheadaptabilityofthenetworktoactionsofdifferentdurations.Finally,through thedecision-levelfusionmodule,thecategoryprobabilityiscalculatedtoobtaintheclassification result. ThesimulationresultsshowthatbasedonthetwobenchmarkdatasetsofNTU RGB+D andKinetics-skeleton,comparedwithothermainstream methods,thismethodhasachievedhigher recognitionaccuracies,whichare96.2%and60.1%,respectively.Themethodcanbetterextract thediscriminativetemporalfeaturesofdifferentactions,andimprovethediscriminationabilityof spatiotemporalactionfeatures.