Computer Engineering Commons^™

Learning Query And Image Similarities With Ranking Canonical Correlation Analysis, Ting Yao, Tao Mei, Chong-Wah Ngo

Research Collection School Of Computing and Information Systems

One of the fundamental problems in image search is to learn the ranking functions, i.e., similarity between the query and image. The research on this topic has evolved through two paradigms: feature-based vector model and image ranker learning. The former relies on the image surrounding texts, while the latter learns a ranker based on human labeled query-image pairs. Each of the paradigms has its own limitation. The vector model is sensitive to the quality of text descriptions, and the learning paradigm is difficult to be scaled up as human labeling is always too expensive to obtain. We demonstrate in this …

Vireo-Tno @ Trecvid 2015: Multimedia Event Detection, Hao Zhang, Yi-Jie Lu, Maaike De Boer, Frank Ter Haar, Zhaofan Qiu, Klamer Schutte, Wessel Kraaij, Chong-Wah Ngo

Research Collection School Of Computing and Information Systems

This paper presents an overview and comparative analysis of our systems designed for the TRECVID 2015 [1] multimedia event detection (MED) task. We submitted 17 runs, of which 5 each for the zeroexample, 10-example and 100-example subtasks for the Pre-Specified (PS) event detection and 2 runs for the 10-example subtask for the Ad-Hoc (AH) event detection. We did not participate in the Interactive Run. This year we focus on three different parts of the MED task: 1) extending the size of our concept bank and combining it with improved dense trajectories; 2) exploring strategies for semantic query generation (SQG); and …

Deep Multimodal Learning For Affective Analysis And Retrieval, Lei Pang, Shiai Zhu, Chong-Wah Ngo

Research Collection School Of Computing and Information Systems

Social media has been a convenient platform for voicing opinions through posting messages, ranging from tweeting a short text to uploading a media file, or any combination of messages. Understanding the perceived emotions inherently underlying these user-generated contents (UGC) could bring light to emerging applications such as advertising and media analytics. Existing research efforts on affective computation are mostly dedicated to single media, either text captions or visual content. Few attempts for combined analysis of multiple media are made, despite that emotion can be viewed as an expression of multimodal experience. In this paper, we explore the learning of highly …

Lesinn: Detecting Anomalies By Identifying Least Similar Nearest Neighbours, Guansong Pang, Kai Ming Ting, David Albrecht

Research Collection School Of Computing and Information Systems

We introduce the concept of Least Similar Nearest Neighbours (LeSiNN) and use LeSiNN to detect anomalies directly. Although there is an existing method which is a special case of LeSiNN, this paper is the first to clearly articulate the underlying concept, as far as we know. LeSiNN is the first ensemble method which works well with models trained using samples of one instance. LeSiNN has linear time complexity with respect to data size and the number of dimensions, and it is one of the few anomaly detectors which can apply directly to both numeric and categorical data sets. Our extensive …

Direct Or Indirect Match? Selecting Right Concepts For Zero-Example Case, Yi-Jie Lu, Maaike De Boer, Hao Zhang, Klamer Schutte, Wessel Kraaij, Chong-Wah Ngo

Research Collection School Of Computing and Information Systems

No abstract provided.

Big Data Proteogenomics And High Performance Computing: Challenges And Opportunities, Fahad Saeed

Parallel Computing and Data Science Lab Technical Reports

Proteogenomics is an emerging field of systems biology research at the intersection of proteomics and genomics. Two high-throughput technologies, Mass Spectrometry (MS) for proteomics and Next Generation Sequencing (NGS) machines for genomics are required to conduct proteogenomics studies. Independently both MS and NGS technologies are inflicted with data deluge which creates problems of storage, transfer, analysis and visualization. Integrating these big data sets (NGS+MS) for proteogenomics studies compounds all of the associated computational problems. Existing sequential algorithms for these proteogenomics datasets analysis are inadequate for big data and high performance computing (HPC) solutions are almost non-existent. The purpose of this …

Practical Guidance For Integrating Data Management Into Long-Term Ecological Monitoring Projects, Robert D. Sutter, Susan Wainscott, John R. Boetsch, Craig Palmer, David J. Rugg

Library Faculty Publications

Long-term monitoring and research projects are essential to understand ecological change and the effectiveness of management activities. An inherent characteristic of long-term projects is the need for consistent data collection over time, requiring rigorous attention to data management and quality assurance. Recent papers have provided broad recommendations for data management; however, practitioners need more detailed guidance and examples. We present general yet detailed guidance for the development of comprehensive, concise, and effective data management for monitoring projects. The guidance is presented as a graded approach, matching the scale of data management to the needs of the organization and the complexity …

From Physical Security To Cybersecurity, Arunesh Sinha, Thanh H. Nguyen, Debarun Kar, Matthew Brown, Milind Tambe, Albert Xin Jiang

Research Collection School Of Computing and Information Systems

Security is a critical concern around the world. In many domains from cybersecurity to sustainability, limited security resources prevent complete security coverage at all times. Instead, these limited resources must be scheduled (or allocated or deployed), while simultaneously taking into account the importance of different targets, the responses of the adversaries to the security posture, and the potential uncertainties in adversary payoffs and observations, etc. Computational game theory can help generate such security schedules. Indeed, casting the problem as a Stackelberg game, we have developed new algorithms that are now deployed over multiple years in multiple applications for scheduling of …

Big Data: Big Value And Big Concerns, Singapore Management University

Perspectives@SMU

Digital information can serve lots of purposes, but timeliness, relevance and privacy issues abound

A Modular Approach For Key-Frame Selection In Wide Area Surveillance Video Analysis, Almabrok Essa, Paheding Sidike, Vijayan K. Asari

Electrical and Computer Engineering Faculty Publications

This paper presents an efficient preprocessing algorithm for big data analysis. Our proposed key-frame selection method utilizes the statistical differences among subsequent frames to automatically select only the frames that contain the desired contextual information and discard the rest of the insignificant frames.

We anticipate that such key frame selection technique will have significant impact on wide area surveillance applications such as automatic object detection and recognition in aerial imagery. Three real-world datasets are used for evaluation and testing and the observed results are encouraging.

Improving Automatic Name-Face Association Using Celebrity Images On The Web, Zhineng Chen, Bailan Feng, Chong-Wah Ngo, Caiyan Jia, Xiangsheng Huang

Research Collection School Of Computing and Information Systems

This paper investigates the task of automatically associating faces appearing in images (or videos) with their names. Our novelty lies in the use of celebrity Web images to facilitate the task. Specifically, we first propose a method named Image Matching (IM), which uses the faces in images returned from name queries over an image search engine as the gallery set of the names, and a probe face is classified as one of the names, or none of them, according to their matching scores and compatibility characterized by a proposed Assigning-Thresholding (AT) pipeline. Noting IM could provide guidance for association for …

Presentation On Evaluating The Creation And Preservation Challenges Of Photogrammetry-Based 3d Models, Michael J. Bennett

UConn Library Presentations

No abstract provided.

Evaluating The Creation And Preservation Challenges Of Photogrammetry-Based 3d Models, Michael J. Bennett

Published Works

Though the roots of photogrammetry can be traced back to photography’s earliest days, only recent advances in both digital technology and software applications have put the possibilities of 3D modeling from 2D source images in the hands of the greater cultural heritage community. The possibilities of such 3D digital rendering are many. With these possibilities come unique digital preservation challenges. This study explores basic close-range photogrammetry as applied to sample archival objects. Additionally, the latest BagIt and ZIP-based bundling formats along with repository-based solutions are also surveyed as potential 3D data management and archiving aggregators for resulting 3D models.

Mining Patterns Of Unsatisfiable Constraints To Detect Infeasible Paths, Sun Ding, Hee Beng Kuan Tan, Lwin Khin Shar

Research Collection School Of Computing and Information Systems

Detection of infeasible paths is required in many areas including test coverage analysis, test case generation, security vulnerability analysis, etc. Existing approaches typically use static analysis coupled with symbolic evaluation, heuristics, or path-pattern analysis. This paper is related to these approaches but with a different objective. It is to analyze code of real systems to build patterns of unsatisfiable constraints in infeasible paths. The resulting patterns can be used to detect infeasible paths without the use of constraint solver and evaluation of function calls involved, thus improving scalability. The patterns can be built gradually. Evaluation of the proposed approach shows …

Cloud Computing, Contractibility, And Network Architecture, Christopher S. Yoo

All Faculty Scholarship

The emergence of the cloud is heightening the demands on the network in terms of bandwidth, ubiquity, reliability, latency, and route control. Unfortunately, the current architecture was not designed to offer full support for all of these services or to permit money to flow through it. Instead of modifying or adding specific services, the architecture could redesigned to make Internet services contractible by making the relevant information associated with these services both observable and verifiable. Indeed, several on-going research programs are exploring such strategies, including the NSF’s NEBULA, eXpressive Internet Architecture (XIA), ChoiceNet, and the IEEE’s Intercloud projects.

Teaching Cybersecurity Using The Cloud, Khaled Salah, Mohammad Hammoud, Sherali Zeadally

Information Science Faculty Publications

Cloud computing platforms can be highly attractive to conduct course assignments and empower students with valuable and indispensable hands-on experience. In particular, the cloud can offer teaching staff and students (whether local or remote) on-demand, elastic, dedicated, isolated, (virtually) unlimited, and easily configurable virtual machines. As such, employing cloud-based laboratories can have clear advantages over using classical ones, which impose major hindrances against fulfilling pedagogical objectives and do not scale well when the number of students and distant university campuses grows up. We show how the cloud paradigm can be leveraged to teach a cybersecurity course. Specifically, we share our …

Cumulonimbus Computing Concerns: Information Security In Public, Private, And Hybrid Cloud Computing, Daniel Adams

Senior Honors Theses

Companies of all sizes operating in all markets are moving toward cloud computing for greater flexibility, efficiency, and cost savings. The decision of how to adopt the cloud is a question of major security concern due to the fact that control is relinquished over certain portions of the IT ecosystem. This thesis presents the position that the main security decision in moving to cloud computing is choosing which type of cloud to employ for each portion of the network – the hybrid cloud approach. Vulnerabilities that exist on a public cloud will be explored, and recommendations on decision factors will …

Optimizing Cloud Use Under Interval Uncertainty, Vladik Kreinovich, Esthela Gallardo

Departmental Technical Reports (CS)

One of the main advantages of cloud computing is that it helps the users to save money: instead of buying a lot of computers to cover all their computations, the user can rent the computation time on the cloud to cover the rare peak spikes of computer need. From this viewpoint, it is important to find the optimal division between in-house and in-the-cloud computations. In this paper, we solve this optimization problem, both in the idealized case when we know the complete information about the costs and the user's need, and in a more realistic situation, when we only know …

Click-Boosting Multi-Modality Graph-Based Reranking For Image Search, Xiaopeng Yang, Yongdong Zhang, Ting Yao, Chong-Wah Ngo, Tao Mei

Research Collection School Of Computing and Information Systems

Image reranking is an effective way for improving the retrieval performance of keyword-based image search engines. A fundamental issue underlying the success of existing image reranking approaches is the ability in identifying potentially useful recurrent patterns from the initial search results. Ideally, these patterns can be leveraged to upgrade the ranks of visually similar images, which are also likely to be relevant. The challenge, nevertheless, originates from the fact that keyword-based queries are used to be ambiguous, resulting in difficulty in predicting the search intention. Mining useful patterns without understanding query is risky, and may lead to incorrect judgment in …

Csr: Small: Collaborative Research: Sane: Semantic-Aware Namespace In Exascale File Systems, Yifeng Zhu

University of Maine Office of Research Administration: Grant Reports

Explosive growth in volume and complexity of data exacerbates the key challenge facing the management of massive data in a way that fundamentally improves the ease and efficacy of their usage. Exascale storage systems in general rely on hierarchically structured namespace that leads to severe performance bottlenecks and makes it hard to support real-time queries on multi-dimensional attributes. Thus, existing storage systems, characterized by the hierarchical directory tree structure, are not scalable in light of the explosive growth in both the volume and the complexity of data. As a result, directory-tree based hierarchical namespace has become restrictive, difficult to use, …

Sheep Updates 2015 - Merredin, Bruce Mullan, Kate Pritchett, Kimbal Curtis, Chris Wilcox, Lynne Bradshaw, Geoff Lindon, Katherine Davies, Joe Young, Stephen Lee, Dawson Bradford, Khama Kelman, Lucy Anderton, Jaq Pearson, Jackie Jarvis, Ben Patrick

Sheep Updates

This session covers fourteen papers from different authors:

1. The Sheep Industry Business Innovation project, Bruce Mullan, Sheep Industry Development Director, Department of Agriculture and Food, Western Australia

2. Western Australian sheep stocktake, Kate Pritchett and Kimbal Curtis, Research Officers, Department of Agriculture and Food, Western Australia

3. Wool demand and supply - short term volatility, long term opportunities, Chris Wilcox, Principal of Poimena Analysis

4. Myths, Facts and the role of animal welfare in farming, Lynne Bradshaw, president, RSPCA WA

5. Latest research and development on breech strike prevention, Geoff Lindon, Manager Productivity and Animal Welfare, AWI

6. …

Evolution And Usage Of The Portal Data Archive: 10-Year Retrospective, Kristin A. Tufte, Robert Bertini, Morgan Harvey

Civil and Environmental Engineering Faculty Publications and Presentations

The Portal transportation data archive (http://portal.its.pdx.edu/) was begun in June 2004 in collaboration with the Oregon Department of Transportation, with a single data source: freeway loop detector data. In 10 years, Portal has grown to contain approximately 3 TB of transportation-related data from a wide variety of systems and sources, including freeway data, arterial signal data, travel times from Bluetooth detection systems, transit data, and bicycle count data. Over its 10-year existence, Portal has expanded both in the type of data that it receives and in the geographic regions from which it gets data. This paper discusses the …