The varying trajectory of Doppler frequency under changing speed motion conditionsare investigated in Highspeed Railway(HSR) scenarios.Based on the geometrical physical parameters,instantaneous Doppler trajectories an...The varying trajectory of Doppler frequency under changing speed motion conditionsare investigated in Highspeed Railway(HSR) scenarios.Based on the geometrical physical parameters,instantaneous Doppler trajectories and expression forms of the change rate arededuced,including acceleration and deceleration cases.These modified models provide more accurate and realisticapproximations in modeling rapidly fading channels.展开更多
目的无监督行人重识别可缓解有监督方法中数据集标注成本高的问题,其中无监督跨域自适应是最常见的行人重识别方案。现有UDA(unsupervised domain adaptive)行人重识别方法在聚类过程中容易引入伪标签噪声,存在对相似人群区分能力差等...目的无监督行人重识别可缓解有监督方法中数据集标注成本高的问题,其中无监督跨域自适应是最常见的行人重识别方案。现有UDA(unsupervised domain adaptive)行人重识别方法在聚类过程中容易引入伪标签噪声,存在对相似人群区分能力差等问题。方法针对上述问题,基于特征具有类内收敛性、类内连续性与类间外散性的特点,提出了一种基于近邻优化的跨域无监督行人重识别方法,首先采用有监督方法得到源域预训练模型,然后在目标域进行无监督训练。为增强模型对高相似度行人的辨识能力,设计了邻域对抗损失函数,任意样本与其他样本构成样本对,使类别确定性最强的一组样本对与不确定性最强的一组样本对之间进行对抗。为使类内样本特征朝着同一方向收敛,设计了特征连续性损失函数,将特征距离曲线进行中心归一化处理,在维持特征曲线固有差异的同时,拉近样本k邻近特征距离。结果消融实验结果表明损失函数各部分的有效性,对比实验结果表明,提出方法性能较已有方法更具优势,在Market-1501(1501 identities dataset from market)和DukeMTMC-reID(multi-target multi-camera person re-identification dataset from Duke University)数据集上的Rank-1和平均精度均值(mean average precision,mAP)指标分别达到了92.8%、84.1%和83.9%、71.1%。结论提出方法设计了邻域对抗损失与邻域连续性损失函数,增强了模型对相似人群的辨识能力,从而有效提升了行人重识别的性能。展开更多
基金supported in part by the NSFC Project under Grants No.61371070Fundamental Research Funds for the Central Universities under grant2015JBM011+2 种基金the Beijing Nova Programme(No.xx2016023)the Open Research Fund of National Mobile Communications Research Laboratory,Southeast University,under Grant No.2012D07Beijing Natural Science Foundation project under grant No.4142041
文摘The varying trajectory of Doppler frequency under changing speed motion conditionsare investigated in Highspeed Railway(HSR) scenarios.Based on the geometrical physical parameters,instantaneous Doppler trajectories and expression forms of the change rate arededuced,including acceleration and deceleration cases.These modified models provide more accurate and realisticapproximations in modeling rapidly fading channels.
文摘目的无监督行人重识别可缓解有监督方法中数据集标注成本高的问题,其中无监督跨域自适应是最常见的行人重识别方案。现有UDA(unsupervised domain adaptive)行人重识别方法在聚类过程中容易引入伪标签噪声,存在对相似人群区分能力差等问题。方法针对上述问题,基于特征具有类内收敛性、类内连续性与类间外散性的特点,提出了一种基于近邻优化的跨域无监督行人重识别方法,首先采用有监督方法得到源域预训练模型,然后在目标域进行无监督训练。为增强模型对高相似度行人的辨识能力,设计了邻域对抗损失函数,任意样本与其他样本构成样本对,使类别确定性最强的一组样本对与不确定性最强的一组样本对之间进行对抗。为使类内样本特征朝着同一方向收敛,设计了特征连续性损失函数,将特征距离曲线进行中心归一化处理,在维持特征曲线固有差异的同时,拉近样本k邻近特征距离。结果消融实验结果表明损失函数各部分的有效性,对比实验结果表明,提出方法性能较已有方法更具优势,在Market-1501(1501 identities dataset from market)和DukeMTMC-reID(multi-target multi-camera person re-identification dataset from Duke University)数据集上的Rank-1和平均精度均值(mean average precision,mAP)指标分别达到了92.8%、84.1%和83.9%、71.1%。结论提出方法设计了邻域对抗损失与邻域连续性损失函数,增强了模型对相似人群的辨识能力,从而有效提升了行人重识别的性能。