ID 原文 译文
56388 首先,根据攻击者对系统有效性、信息完整性、身份认证、隐私和机密性保护的不同安全目标,将5G潜在的安全威胁进行了分类,分析了各个层次可能面临的威胁与攻击手段. The survey begins with the potential security vulnerabilities of 5G, which can be classified according to systemavailability, information integrity, authentication, privacy, and confidentiality protection. Then, threats andattack methods encountered at different levels are analyzed.
56389 其次,简述了相关标准中的5G安全架构,讨论了5G及其演进系统潜在的物理层、网络层和应用层安全技术.最后,本文指出了未来继续提升5G及B5G安全的潜在研究方向. Subsequently, the 5G security architecture accordingto relevant standards is briefly described. Based on the threats and security architecture, security technologies arereviewed based on three categories of 5G and beyond 5G (B5G), such as the physical, network, and applicationlayers. Finally, the survey concludes by presenting observations on potential future directions of 5G and B5Gsecurity.
56390 目前,大部分活体检测的方法将活体检测任务视作有监督的二分类问题,进而努力充分提取真实人脸和攻击人脸的特征,在单个数据集内部训练和测试可以达到很高的准确率,但是在交叉数据集之间训练和测试往往效果不佳. Most of presentation attack detection (PAD) methods consider the task as a supervised binaryclassification problem. Many of these methods struggle to grasp adequate spoofing cues and generalize poorly. In the test, the attacks are regarded as out-of-distributions samples that naturally exhibit a higher feature reconstruction error in the latent space than real samples in the dataset.
56391 因此本文提出了一种新颖的基于隐空间约束的深度对抗网络,它通过半监督学习的方式进行对抗训练,在此过程中模型不仅仅可以获得正常样本在隐空间中的分布,还可以通过一种惩罚的方式对隐空间中正常样本的特征进行约束,这将带来更加有效和鲁棒的活体检测效果.测试过程中,攻击人脸样本将被视作离群的样本,它们相对于正常样例在隐空间中的表达具有更高的重构差.实验表明提出的模型相较于前沿的半监督异常检测方法具备明显的优势,并且在活体检测跨数据集和单数据集内达到了可比的效果或者目前最好的效果. In this paper, we formulate the face anti-spoofing detection as an anomaly detection task to tackle the generalization issue. Experiments show that our modelis clearly superior over cutting-edge semi-supervised abnormal detectors and achieves state-of-the-art results onboth intra- and inter-database testing. In group activity recognition, the hierarchical framework is widely used to represent the relationshipsbetween individuals and their corresponding groups and has achieved promising performance. However, existingmethods simply employ the max/average pooling in this framework, overlooking the distinct contributions ofdifferent individuals to the group activity recognition.
56392 为此,本文提出一种针对个体行为特征聚合的注意力池化机制,并依此建立了新型群体行为识别模型,以自底向上的方式同时实现个体行为与群体行为分层识别. In this paper, we propose a new contextual pooling scheme,named attentive pooling, which enables weighted information transition from individual actions to group activity.
56393 未验证所提方法的有效性,本文依托广泛使用的The Volleyball Dataset数据集上开展了一系列实验验证.结果显示,本文所提出的模型取得了较好的分类准确率,分类性能优于当前先进模型. Experimental results on the benchmark dataset demonstrate that theproposed scheme is significantly superior over the baseline and comparable to state-of-the-art methods.
56394 在高铁日常运营中,地质灾害或设备故障等突发事件导致铁路行车区间中断,长时间的区间中断通常会导致大面积的列车晚点,给旅客出行带来极大的延误,如何在区间中断下对列车进行实时调整是高速铁路调度运行的一项重要课题. During the daily operation of high-speed railways, unexpected events such as geological disastersor equipment failure may lead to railway segment blockages. A long-duration segment blockage usually leadsto large-scale train delays, which cause significant inconvenience to passengers. Adjusting train schedules inreal time during segment blockages is an important issue for the dispatch and operation of high-speed railways.
56395 本文针对固定时间的区间中断,构建了列车运行速度调整和运行图调整的混合整数非线性规划(mixed integer nonlinear programming, MINLP)模型. For a segment blockage of a known time interval, a mixed integer nonlinear programming (MINLP) model fortrain speed adjustment and train diagram adjustment is constructed.
56396 在该模型中,除了常见的取消车次和延迟发车调度策略,还将降速运行策略考虑在内,在满足行车安全和车站通行能力的约束条件下,以3种调度策略的加权延误影响最小化为目标. In this model, in addition to the commonscheduling strategies involving train cancellation and departure delay, the deceleration strategy is considered. Under the constraints of traffic safety and station capacity, the objective is to minimize the weighted delays ofthe three strategies.
56397 该模型基于列车运动学模型计算列车降速运行带来的延误时间,并通过移动闭塞原理控制列车运行安全间隔. Additionally, the delay time owing to train speed reduction is calculated based on the trainkinematic model, and the safe interval of train operation is controlled using the moving block principle.