ID 原文 译文
20575 首先，运用深度卷积神经网络的全连接层将不同视角下的人体姿态映射到与视角无关的高维空间，以构建空间域下深度行为视频的人体姿态模型(HPM)； Firstly, the fully connectedlayer of depth Convolution Neural Network (CNN) is creatively used to map human posture in differentperspectives to high-dimensional space that is independent with perspective to achieve the Human PostureModeling (HPM) of deep-performance video in spatial domain.
20576 其次，考虑视频序列帧之间的时空相关性，在每个神经元激活的时间序列中分段应用时间等级池化(RP)函数，实现对视频时间子序列的编码； Secondly, considering temporal-spatial correlation between video sequence frames, the Rank Pooling (RP) function is applied to the series of each neuron activated time to encode the video time sub-sequence,
20577 然后，将傅里叶时间金字塔(FTP)算法作用于每一个池化后的时间序列，并加以连接产生最终的时空特征表示； and then the Fourier Time Pyramid (FTP) is used to each pooled time series to produce the final spatio-temporal feature representation.
20578 最后，在不同数据集上，基于不同方法进行了行为识别分类测试。 Finally, differentmethods of behavior recognition classification are tested on several datasets.
20579 实验结果表明，该文方法(HPM+RP+FTP)提高了不同视角下深度视频识别准确率，在UWA3DII数据集中，比现有最好方法高出18%。 Experimental results show that theproposed method improves the accuracy of depth video recognition in different perspectives. In the UWA3DIIdatasets, the proposed method is 18% higher than the most recent method.
20580 此外，该文方法具有较好的泛化性能，在MSRDaily Activity3D数据集上得到82.5%的准确率。 The proposed method (HPM+RP+FTP) has a good generalization performance, achieving a 82.5% accuracy on dataset of MSR DailyActivity3D.
20581 针对现有的虚拟网络重构算法对物理网络中产生的碎片资源考虑不够周到，导致其对在线虚拟网络映射算法的性能改善不够显著的问题， The existing virtual network reconfiguration algorithms do not consider the fragment resources generated in the physical network, which results in the improvement of the performance of the online virtual network embedding algorithms is not obvious.
20582 该文定义了一种网络资源碎片度度量方法，并提出一种碎片感知的安全虚拟网络重构算法。 To solve this problem, a definition of network resource fragmentation is given, and a Fragment-Aware Secure Virtual Network Reconfiguration (FA-SVNR) algorithm is proposed.
20583 该算法通过周期性考虑物理网络中节点的碎片度，选择出待迁移虚拟节点集合；通过综合考虑物理网络的碎片度减小量和虚拟网络的映射开销减少量，选择出最佳的虚拟节点迁移方案。 In the process of reconfiguration, the virtual node set to be migrated is selected by considering the fragmentation of nodes in the physical network periodically, and the best virtual node migration scheme is selected by considering the reduction of the fragmentation of the physical network and the reduction of the embedding cost of the virtual network.
20584 仿真结果表明，该算法的请求接受率和收益开销比均优于当前的重构算法，特别是在收益开销比方面的优势更加明显。 Simulation results show that the proposed algorithm has the higher acceptance ratio and revenue to cost ratio compared with the existing virtual network reconfiguration algorithm, especially in the metric of revenue to cost ratio.