检索结果-内蒙古大学图书馆

Fusing R Features and Local Features with Context-Aware Kernels for Action Recognition

学校读者我要写书评

暂无评论

International Journal of computer Vision 2016年第2期118卷 151-171页

作者： Yuan, Chunfeng Wu, Baoxin Li, Xi Hu, Weiming Maybank, Stephen Wang, Fangshi National Laboratory of Pattern Recognition Institute of Automation CAS Beijing China College of Computer Science and Technology Zhejiang University Hangzhou China Department of Computer Science and Information Systems Birkbeck College London United Kingdom School of Software Engineering Beijing Jiaotong University Beijing China

The performance of action recognition in video sequences depends significantly on the representation of actions and the similarity measurement between the representations. In this paper, we combine two kinds of features extracted from the spatio-temporal interest points with context-aware kernels for action recognition. For the action representation, local cuboid features extracted around interest points are very popular using a Bag of Visual Words (BOVW) model. Such representations, however, ignore potentially valuable information about the global spatio-temporal distribution of interest points. We propose a new global feature to capture the detailed geometrical distribution of interest points. It is calculated by using the 3D R transform which is defined as an extended 3D discrete Radon transform, followed by the application of a two-directional two-dimensional principal component analysis. For the similarity measurement, we model a video set as an optimized probabilistic hypergraph and propose a context-aware kernel to measure high order relationships among videos. The context-aware kernel is more robust to the noise and outliers in the data than the traditional context-free kernel which just considers the pairwise relationships between videos. The hyperedges of the hypergraph are constructed based on a learnt Mahalanobis distance metric. Any disturbing information from other classes is excluded from each hyperedge. Finally, a multiple kernel learning algorithm is designed by integrating the l2 norm regularization into a linear SVM classifier to fuse the R feature and the BOVW representation for action recognition. Experimental results on several datasets demonstrate the effectiveness of the proposed approach for action recognition. © 2015, Springer science+Business Media New York.

关键词： Principal component analysis

Conflict-driven hybrid observer-based anomaly detection

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Wang, Zheng Harirchi, Farshad Anand, Dhananjay Tang, CheeYee Moyne, James Tilbury, Dawn Department of Mechanical Engineering University of Michigan Ann ArborMI United States Department of Electrical Engineering and Computer Science University of Michigan Ann ArborMI United States Software and Systems Divison National Institute of Standards and Technology GaithersburgMD United States

This paper presents an anomaly detection method using a hybrid observer - which consists of a discrete state observer and a continuous state observer. We focus our attention on anomalies caused by intelligent attacks, which may bypass existing anomaly detection methods because neither the event sequence nor the observed residuals appear to be anomalous. Based on the relation between the continuous and discrete variables, we define three conflict types and give the conditions under which the detection of the anomalies is guaranteed. We call this method conflict-driven anomaly detection. The effectiveness of this method is demonstrated mathematically and illustrated on a Train-Gate (TG) system. Copyright © 2017, The Authors. All rights reserved.

关键词： State estimation

Towards physiology-aware DASH: Bandwidth-compliant prioritized clinical multimedia communication in ambulances

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Hosseini, Mohammad Jiang, Yu Berlin, Richard R. Sha, Lui Song, Houbing UIUC Tsinghua University Carle Foundation Hospital West Virginia University Illinois United States School of Software Tsinghua University China Department of Computer Science UIUC Carle Foundation Hospital UrbanaIL United States Department of Electrical Computer Software and Systems Engineering Embry-Riddle Aeronautical University Daytona Beach United States

The ultimate objective of medical cyber-physical systems is to enhance the safety and effectiveness of patient care. To ensure safe and effective care during emergency patient transfer from rural areas to center tertiary hospitals, reliable and real-time communication is essential. Unfortunately, real-time monitoring of patients involves transmission of various clinical multimedia data including videos, medical images, and vital signs, which requires use of mobile network with high-fidelity communication bandwidth. However, the wireless networks along the roads in rural areas range from 4G to 2G to low speed satellite links, which poses a significant challenge to transmit critical patient information. In this paper, we present a bandwidth-compliant criticality-aware system for transmission of massive clinical multimedia data adaptive to varying bandwidths during patient transport. Model-based clinical automata are used to determine the criticality of clinical multimedia data. We borrow concepts from DASH, and propose physiology-aware adaptation techniques to transmit more critical clinical data with higher fidelity in response to changes in disease, clinical states, and bandwidth condition. In collaboration with Carle's ambulance service center, we develop a bandwidth profiler, and use it as proof of concept to support our experiments. Our preliminary evaluation results show that our solutions ensure that most critical patient's clinical data are communicated with higher fidelity. Copyright © 2017, The Authors. All rights reserved.

关键词： Bandwidth

Phonetic temporal neural model for language identification

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Tang, Zhiyuan Wang, Dong Chen, Yixiang Li, Lantian Abel, Andrew Chengdu Institute of Computer Applications Chinese Academy of Sciences and University of Chinese Academy of Sciences Beijing100049 China Center for Speech and Language Technologies Tsinghua University Beijing100084 China Tsinghua National Laboratory for Information Science and Technology and the Center for Speech and Language Technologies Tsinghua University Beijing100084 China Department of Computer Science and Software Engineering Xi'an Jiaotong-Liverpool University Suzhou215123 China

Deep neural models, particularly the LSTM-RNN model, have shown great potential for language identification (LID). However, the use of phonetic information has been largely overlooked by most existing neural LID methods, although this information has been used very successfully in conventional phonetic LID systems. We present a phonetic temporal neural model for LID, which is an LSTM-RNN LID system that accepts phonetic features produced by a phone-discriminative DNN as the input, rather than raw acoustic features. This new model is similar to traditional phonetic LID methods, but the phonetic knowledge here is much richer: it is at the frame level and involves compacted information of all phones. Our experiments conducted on the Babel database and the AP16-OLR database demonstrate that the temporal phonetic neural approach is very effective, and significantly outperforms existing acoustic neural models. It also outperforms the conventional i-vector approach on short utterances and in noisy conditions. Copyright © 2017, The Authors. All rights reserved.

关键词： Deep neural networks

Investigating factors that affect the human perception on god class detection: an analysis based on a family of four controlled experiments

学校读者我要写书评

暂无评论

Journal of software engineering Research and Development 2017年第1期5卷 1-39页

作者： José Amancio M. Santos João B. Rocha-Junior Manoel Gomes de Mendonça Technology Department State University of Feira de Santana Avenida Transnordestina s/n - Novo Horizonte Feira de Santan Bahia Brazil Exact Science Department State University of Feira de Santana Bahia Brazil Fraunhofer Project Center for Software & Systems Engineering Federal University of Bahia Bahia Brazil Department of Computer Science Federal University of Bahia Bahia Brazil

Evaluation of design problems in object oriented systems, which we call code smells, is mostly a human-based task. Several studies have investigated the impact of code smells in practice. Studies focusing on human identification of code smells have shown low agreement among developers. Unfortunately, those studies do not attempt to investigate the reasons behind this phenomenon. This paper aims to investigate factors affecting human perception of code smells. Specifically, it focuses on factors affecting god class detection, one of the most known code smells. The investigation encompassed a family of four controlled experiments, covering potential factors affecting human detection of code smells. The method is incremental. In other words, each experiment produces insights to the next one. This allows the investigators to control specific factors affecting the agreement on god class detection. The factors addressed in this study are: i) developer experience, ii) developer knowledge, iii) developer training, iv) tool support for design comprehension, and v) software size. Our findings show that tool support for design comprehension is the only factor that does not affect the human perception of god class. The other factors impact this perception in some way. The area still needs more investigation and discussion on what we call the code smell conceptualization problem, to ensure similar criteria and thresholds on human-based code smell detection.

关键词：

Rating protocol design for extortion and cooperation in the crowdsourcing contest dilemma

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Lu, Jianfeng Xin, Yun Zhang, Zhao Tang, Shaojie Yan, Songyuan Tang, Changbing Department of Computer Science and Engineering Zhejiang Normal University Jinhua Zhejiang China State Key Laboratory for Novel Software Technology Nanjing University Nanjing Jiangsu China Department of Information Systems University of Texas at Dallas RichardsonTX75080 United States

Crowdsourcing has emerged as a paradigm for leveraging human intelligence and activity to solve a wide range of tasks. However, strategic workers will find enticement in their self-interest to free-ride and attack in a crowdsourcing contest dilemma game. Hence, incentive mechanisms are of great importance to overcome the inefficiency of the socially undesirable equilibrium. Existing incentive mechanisms are not effective in providing incentives for cooperation in crowdsourcing competitions due to the following features: heterogeneous workers compete against each other in a crowdsourcing platform with imperfect monitoring. In this paper, we take these features into consideration, and develop a novel game-theoretic design of rating protocols, which integrates binary rating labels with differential pricing to maximize the requester's utility, by extorting selfish workers and enforcing cooperation among them. By quantifying necessary and sufficient conditions for the sustainable social norm, we formulate the problem of maximizing the revenue of the requester among all sustainable rating protocols, provide design guidelines for optimal rating protocols, and design a low-complexity algorithm to select optimal design parameters which are related to differential punishments and pricing schemes. Simulation results demonstrate how intrinsic parameters impact on design parameters, as well as the performance gain of the proposed rating protocol. Copyright © 2017, The Authors. All rights reserved.

关键词： Crowdsourcing

A Teacher-Student framework for zero-resource neural machine translation

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Cheny, Yun Liuz, Yang Cheng, Yong Li, Victor O.K. Department of Electrical and Electronic Engineering University of Hong Kong State Key Laboratory of Intelligent Technology Systems Tsinghua National Laboratory for Information Science and Technology Department of Computer Science and Technology Tsinghua University Beijing Jiangsu Collaborative Innovation Center for Language Competence Jiangsu Institute for Interdisciplinary Information Sciences Tsinghua University Beijing China

While end-to-end neural machine translation (NMT) has made remarkable progress recently, it still suffers from the data scarcity problem for low-resource language pairs and domains. In this paper, we propose a method for zero-resource NMT by assuming that parallel sentences have close probabilities of generating a sentence in a third language. Based on this assumption, our method is able to train a source-to-target NMT model ("student") without parallel corpora available, guided by an existing pivot-to-target NMT model ("teacher") on a source-pivot parallel corpus. Experimental results show that the proposed method significantly improves over a baseline pivot-based model by +3.0 BLEU points across various language pairs. Copyright © 2017, The Authors. All rights reserved.

关键词： Neural machine translation

SmartVM: A Multi-Layer Microservice-Based Platform for Deploying SaaS

学校读者我要写书评

暂无评论

SmartVM: A Multi-Layer Microservice-Based Platform for Deplo...

IEEE International Symposium on Parallel and Distributed Processing with Applications and IEEE International Conference on Ubiquitous Computing and Communications (ISPA/IUCC)

作者： Xi Zheng Jiaojiao Jiang Yuqun Zhang Yao Deng Min Fu Tianlei Zheng Xiao Liu Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen China School of Software and Electrical Engineering Swinburne University of Technology Melbourne Australia School of Information Technology Deakin University Melbourne Australia Data61 CSIRO Sydney Australia School of Computing and Information Systems The University of Melbourne Melbourne Australia

ISBN: (纸本)9781538637913

With the advent of software-as-a-Service (SaaS), SaaS developers are facing many challenges associated with the multi-tenancy and the dramatically increased number of users. In order to achieve resource-optimized, on-demand dynamic scaling across multiple tenants, and reduce costs, in this paper, a new platform, named SmartVM, is created to enable SaaS developer to create, customize, and deploy SaaS solutions in a multi-tier microservice-based manner. We develop an e-commerce SaaS prototype to evaluate effectiveness and efficiency of SmartVM. The results show that the SmartVM deployments outperforms the conventional monolithic and microservice deployments in smart monitoring, cost reduction, and resource optimization.

关键词： Containers Monitoring Business software as a service Australia Scalability Measurement

Supplementary material for the information sciences paper: An experimental study of hyper-heuristic selection and acceptance mechanism for combinatorial t-way test suite generation

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Zamli, Kamal Z. Din, Fakhrud Kendall, Graham Ahmed, Bestoun S. IBM Centre of Excellence Faculty of Computer Systems and Software Engineering Universiti Malaysia Pahang Lebuhraya Tun Razak Kuantan Pahang Darul Makmur26300 Malaysia School of Computer Science University of Nottingham Malaysia Campus Jalan Broga Semenyih Selangor Darul Ehsan43500 Malaysia Department of Computer Science Faculty of Electrical Engineering Czech Technical University Karlovo n'am. 13 Pracha 2121 Czech Republic