基于上下文聚类的云虚拟机异常检测与识别策略北大核心CSCD

根据虚拟机的运行特点,文章提出一种基于上下文聚类的虚拟机异常检测策略。该策略采用一种新的聚类初始中心点选取策略将具有相似上下文运行环境的虚拟机实例进行聚集,对影响空间的局部异常因子算法进行增量式改进,针对每一个上下文类簇构建了上下文异常检测模型。对实时采集的虚拟机实例按照其包含的上下文信息将其匹配到相应上下文异常检测模型中,相应的上下文异常检测模型能够对新采集的虚拟机实例进行增量式异常检测。多个数值实验证明文章提出的异常检测模型和识别算法是有效且高效的。     

融合全局与空间多尺度上下文信息的车辆重识别

目的 车辆重识别指判断不同摄像设备拍摄的车辆图像是否属于同一辆车的检索问题。现有车辆重识别算法使用车辆的全局特征或额外的标注信息,忽略了对多尺度上下文信息的有效抽取。对此,本文提出了一种融合全局与空间多尺度上下文信息的车辆重识别模型。方法 首先,设计一个全局上下文特征选择模块,提取车辆的细粒度判别信息,并且进一步设计了一个多尺度空间上下文特征选择模块,利用多尺度下采样的方式,从全局上下文特征选择模块输出的判别特征中获得其对应的多尺度特征。然后,选择性地集成来自多级特征的空间上下文信息,生成车辆图像的前景特征响应图,以此提升模型对于车辆空间位置特征的感知能力。最后,模型组合了标签平滑的交叉熵损失函数和三元组损失函数,以提升模型对强判别车辆特征的整体学习能力。结果 在VeRi-776(vehicle re-idendification-776)数据集上,与模型PNVR(part-regularized near-duplicate vehicle re-identification)相比,本文模型的mAP(mean average precision)和rank-1 (cumulative...     

基于单词-章节关联的科技论文摘要

为科技论文生成自动摘要,这能够帮助作者更快撰写摘要,是自动文摘的研究内容之一.相比于常见的新闻文档,科技论文具有文档结构性强、逻辑关系明确等特点.目前,主流的编码-解码的生成式文摘模型主要考虑文档的序列化信息,很少深入探究文档的篇章结构信息.为此,文中针对科技论文的特点,提出了一种基于"单词-章节-文档"层次结构的自动摘要模型,利用单词与章节的关联作用增强文本结构的层次性和层级之间的交互性,从而筛选出科技论文的关键信息.除此之外,该模型还扩充了一个上下文门控单元,旨在更新优化上下文向量,从而能更全面地捕获上下文信息.实验结果表明,提出的模型可有效提高生成文摘在ROUGE评测方法上的各项指标性能.     

基于单词-章节关联的科技论文摘要

为科技论文生成自动摘要,这能够帮助作者更快撰写摘要,是自动文摘的研究内容之一.相比于常见的新闻文档,科技论文具有文档结构性强、逻辑关系明确等特点.目前,主流的编码-解码的生成式文摘模型主要考虑文档的序列化信息,很少深入探究文档的篇章结构信息.为此,文中针对科技论文的特点,提出了一种基于"单词-章节-文档"层次结构的自动摘要模型,利用单词与章节的关联作用增强文本结构的层次性和层级之间的交互性,从而筛选出科技论文的关键信息.除此之外,该模型还扩充了一个上下文门控单元,旨在更新优化上下文向量,从而能更全面地捕获上下文信息.实验结果表明,提出的模型可有效提高生成文摘在ROUGE评测方法上的各项指标性能.     

基于BERT和GCN的引文推荐模型

采用一个组织良好的数据集和基于深度学习的模型,实现根据上下文获得论文的引文推荐。模型包括一个文档编码器和一个上下文编码器,使用图卷积网络层(GCN)和预训练模型BERT[1]的双向编码器表示。通过修改相关的PeerRead数据集,建立一个PeerReadPlus新数据集,它包含引用文献的上下文语句和论文元数据。结果表明,在采用BERT进行上下文编码,改进上下文侧的表示学习后,该模型的性能获得了显著的提升,平均精度均值(MAP)和召回率(Recall@K)均提高了28%以上。     

基于ECPA神经网络的情绪原因识别方法

情绪原因识别是文本情绪分析领域中的一个前沿研究方向。传统情绪原因识别方法需要进行规则制定、抽取特征,而该文从情绪原因的语言特点出发,结合Bi-LSTM模型和注意力机制,提出一种基于情绪上下文位置注意力神经网络的情绪原因识别方法(ECPA)。该方法考虑了情绪词和情绪类别中的情绪信息,学习了Bi-LSTM模型建模后的上下文语义信息,引入了基于位置信息的注意力机制模型,进而构建情绪原因识别模型。实验结果证明,该方法在情绪原因识别任务中的有效性,并取得了目前最优的性能,同时对情绪归因方法具有一定的指导作用。     

上下文场景识别模型的稀疏贝叶斯判别学习方法

在机器人场景识别问题中,将连续场景的相关性通过基于隐马尔可夫模型的上下文模型进行描述.采用不同于传统的使用生成模型方法学习上下文场景识别模型的方式,首先引入稀疏贝叶斯学习机对上下文模型中图像特征的后验概率进行建模,然后通过贝叶斯原理将稀疏贝叶斯模型与隐马尔可夫模型结合,提出一种能够实现上下文场景识别模型的判别学习方法.在真实场景数据库上的实验结果表明,由该方法得到的上下文场景识别系统具有很好的场景识别能力和泛化特性.     

基于卷积神经网络的中文景点识别研究

命名实体识别是自然语言处理任务的重要环节。近年来,基于深度学习的通用命名实体识别模型取得显著效果。而在旅游领域,中文旅游景点实体识别主要依赖于特征工程的方法。提出一种基于CNN-BiLSTM-CRF的网络模型,该模型不使用任何人工特征,通过神经网络充分对文本的局部信息特征进行抽象化抽取和表示,并学习和利用文本的上下文信息,实现对景点实体的识别。实验结果显示,该方法能够有效识别中文旅游景点实体,并在实验中取得[F1]值93.9%的效果。     

基于上下文语义信息的铁路扣件状态检测

针对传统"视觉词包(BOW)模型"识别铁路扣件状态时仅利用扣件图像的特征域,忽略其空间域中上下文语义信息的缺点,提出了一种基于上下文语义信息的扣件检测模型.在传统"视觉词包模型"的基础上,引入吉布斯随机场模型对图像中像素的空间相关性进行建模,将图像块在特征域的相似性与空间域的上下文语义约束关系结合,更准确地定义视觉单词;利用潜在狄利克雷分布(LDA)学习扣件图像的主题分布;采用支持向量机(SVM)对扣件进行分类识别.对4类扣件图像的分类实验证明:模型能够有效提高扣件分类精度.     

基于注意力机制的多模态幽默识别

近年来,幽默识别逐渐成为自然语言处理领域的热点研究之一。已有的研究多聚焦于文本上的幽默识别,在多模态数据上开展此任务的研究相对较少,现有方法在学习模态间交互信息上存在不足。该文提出了基于注意力机制的模态融合模型,首先对单模态上下文进行独立编码,得到单一模态的特征向量;然后将注意力机制作用于两种模态的特征序列,使用层级注意力结构捕获多模态信息在段落上下文中的关联与交互。该文在UR-FUNNY公开数据集上进行了实验,相比之前最优结果在精确率上提升了1.37%。实验表明,该文提出的模型能很好地对多模态上下文进行建模,引入多模态交互信息和段落上下文信息可提高幽默识别的性能。     

Context based object categorization: A critical survey

The goal of object categorization is to locate and identify instances of an object category within an image. Recognizing an object in an image is difficult when images include occlusion, poor quality, noise or background clutter, and this task becomes even more challenging when many objects are present in the same scene. Several models for object categorization use appearance and context information from objects to improve recognition accuracy. Appearance information, based on visual cues, can successfully identify object classes up to a certain extent. Context information, based on the interaction among objects in the scene or global scene statistics, can help successfully disambiguate appearance inputs in recognition tasks. In this work we address the problem of incorporating different types of contextual information for robust object categorization in computer vision. We review different ways of using contextual information in the field of object categorization, considering the most common levels of extraction of context and the different levels of contextual interactions. We also examine common machine learning models that integrate context information into object recognition frameworks and discuss scalability, optimizations and possible future approaches.     

Probabilistic Models of Appearance for 3-D Object Recognition

We describe how to model the appearance of a 3-D object using multiple views, learn such a model from training images, and use the model for object recognition. The model uses probability distributions to describe the range of possible variation in the object's appearance. These distributions are organized on two levels. Large variations are handled by partitioning training images into clusters corresponding to distinctly different views of the object. Within each cluster, smaller variations are represented by distributions characterizing uncertainty in the presence, position, and measurements of various discrete features of appearance. Many types of features are used, ranging in abstraction from edge segments to perceptual groupings and regions. A matching procedure uses the feature uncertainty information to guide the search for a match between model and image. Hypothesized feature pairings are used to estimate a viewpoint transformation taking account of feature uncertainty. These methods have been implemented in an object recognition system, OLIVER. Experiments show that OLIVER is capable of learning to recognize complex objects in cluttered images, while acquiring models that represent those objects using relatively few views.     

基于视频深度学习的时空双流人物动作识别模型

深度学习在人物动作识别方面已取得较好的成效,但当前仍然需要充分利用视频中人物的外形信息和运动信息。为利用视频中的空间信息和时间信息来识别人物行为动作,提出一种时空双流视频人物动作识别模型。该模型首先利用两个卷积神经网络分别抽取视频动作片段空间和时间特征,接着融合这两个卷积神经网络并提取中层时空特征,最后将提取的中层特征输入到3D卷积神经网络来完成视频中人物动作的识别。在数据集UCF101和HMDB51上,进行视频人物动作识别实验。实验结果表明,所提出的基于时空双流的3D卷积神经网络模型能够有效地识别视频人物动作。     

Human-object integrated assembly intention recognition for context-aware human-robot collaborative assembly

Human-robot collaborative (HRC) assembly combines the advantages of robot's operation consistency with human's cognitive ability and adaptivity, which provides an efficient and flexible way for complex assembly tasks. In the process of HRC assembly, the robot needs to understand the operator's intention accurately to assist the collaborative assembly tasks. At present, operator intention recognition considering context information such as assembly objects in a complex environment remains challenging. In this paper, we propose a human-object integrated approach for context-aware assembly intention recognition in the HRC, which integrates the recognition of assembly actions and assembly parts to improve the accuracy of the operator's intention recognition. Specifically, considering the real-time requirements of HRC assembly, spatial-temporal graph convolutional networks (ST-GCN) model based on skeleton features is utilized to recognize the assembly action to reduce unnecessary redundant information. Considering the disorder and occlusion of assembly parts, an improved YOLOX model is proposed to improve the focusing capability of network structure on the assembly parts that are difficult to recognize. Afterwards, taking decelerator assembly tasks as an example, a rule-based reasoning method that contains the recognition information of assembly actions and assembly parts is designed to recognize the current assembly intention. Finally, the feasibility and effectiveness of the proposed approach for recognizing human intentions are verified. The integration of assembly action recognition and assembly part recognition can facilitate the accurate operator's intention recognition in the complex and flexible HRC assembly environment.     

Observing Human-Object Interactions: Using Spatial and Functional Compatibility for Recognition

Interpretation of images and videos containing humans interacting with different objects is a daunting task. It involves understanding scene/event, analyzing human movements, recognizing manipulable objects, and observing the effect of the human movement on those objects. While each of these perceptual tasks can be conducted independently, recognition rate improves when interactions between them are considered. Motivated by psychological studies of human perception, we present a Bayesian approach which integrates various perceptual tasks involved in understanding human-object interactions. Previous approaches to object and action recognition rely on static shape/appearance feature matching and motion analysis, respectively. Our approach goes beyond these traditional approaches and applies spatial and functional constraints on each of the perceptual elements for coherent semantic interpretation. Such constraints allow us to recognize objects and actions when the appearances are not discriminative enough. We also demonstrate the use of such constraints in recognition of actions from static images without using any motion information.     

基于时空增强式交互网络的组合动作识别

组合动作识别是计算机视觉领域一个新的挑战,它旨在识别未见过的动作与物体的组合。传统的动作识别模型往往会在物体外观与动作类别之间建立联系,引入错误的偏置,在面对未见过的动作与物体的组合时性能急剧恶化。现有解决方法是忽视外观信息,以物体的坐标和身份等信息作为输入,建立以物体为中心的模型。受此启发,提出了时空增强式交互模型。首先在基础网络的不同深度提取并聚合多级别物体特征;然后构建物体分支,使用时空增强模块和物体交互模块分别对物体特征进行增强以及建模物体的移动和交互模式;最终将该分支的输出与基础网络的输出融合用于动作分类,使模型兼顾外观信息和物体交互信息。在多个数据集上的广泛实验证明了所提模型的有效性。     

一种上下文信息融合的安全帽识别算法

为了预防人员防护缺失导致的生产事故,着力探究复杂施工场景下人员安全帽佩戴情况的智能化识别。在一阶段目标检测算法的基础上,针对安全帽识别问题中的小目标和安全帽纹理信息缺失的问题,提出提取并融合上下文信息,以增强模型的表征学习能力。首先,为解决特征鉴别力不足的问题,提出局部上下文感知模块和全局上下文融合模块。局部上下文感知模块能够融合人体头部信息和安全帽信息获取具有鉴别力的特征表示;全局上下文融合模块将高层的语义信息与浅层特征融合,提升浅层特征的抽象能力。其次,为了解决小目标识别问题,提出使用多个不同的目标检测模块分别识别不同大小的目标。在构建的复杂施工场景下的安全帽识别数据集上的实验结果表明：提出的2个模块将mAP提高了11.46个百分点,安全帽识别的平均精度提高了10.55个百分点。本文提出的方法具有速度快、精度高的特点,为智慧工地提供了有效的技术解决方案。     

The illumination-invariant recognition of 3D objects using localcolor invariants

Traditional approaches to three dimensional object recognition exploit the relationship between three dimensional object geometry and two dimensional image geometry. The capability of object recognition systems can be improved by also incorporating information about the color of object surfaces. Using physical models for image formation, the authors derive invariants of local color pixel distributions that are independent of viewpoint and the configuration, intensity, and spectral content of the scene illumination. These invariants capture information about the distribution of spectral reflectance which is intrinsic to a surface and thereby provide substantial discriminatory power for identifying a wide range of surfaces including many textured surfaces. These invariants can be computed efficiently from color image regions without requiring any form of segmentation. The authors have implemented an object recognition system that indexes into a database of models using the invariants and that uses associated geometric information for hypothesis verification and pose estimation. The approach to recognition is based on the computation of local invariants and is therefore relatively insensitive to occlusion. The authors present several examples demonstrating the system's ability to recognize model objects in cluttered scenes independent of object configuration and scene illumination. The discriminatory power of the invariants has been demonstrated by the system's ability to process a large set of regions over complex scenes without generating false hypotheses     

Using a structured world model in flexible recognition of two dimensional patterns

An attempt for a world model driven recognition process of two dimensional visual patterns is presented. Patterns are organized in a world model according to their shapes. The world model is in a form of a lattice-like structure. The nodes represent either properties or relations of considered objects. The lattice structure of the world model allows us to recognize objects by partial information, because of the lattice property, e.g. the greatest lower bound. Examples of a computer recognition of two dimensional patterns are shown.