Open Access. Powered by Scholars. Published by Universities.®

Computer Sciences Commons™

Open Access. Powered by Scholars. Published by Universities.®

Singapore Management University

Discipline
Keyword
Publication Year
Publication
Publication Type
File Type

Articles 4711 - 4740 of 9025

Full-Text Articles in Computer Sciences

Fixation And Confusion: Investigating Eye-Tracking Participants' Exposure To Information In Personas, Joni Salminen, Bernard J. Jansen, Jisun An, Soon-Gyo Jung, Lene Nielsen, Haewoon Kwak Mar 2018

Fixation And Confusion: Investigating Eye-Tracking Participants' Exposure To Information In Personas, Joni Salminen, Bernard J. Jansen, Jisun An, Soon-Gyo Jung, Lene Nielsen, Haewoon Kwak

Research Collection School Of Computing and Information Systems

To more effectively convey relevant information to end users of persona profiles, we conducted a user study consisting of 29 participants engaging with three persona layout treatments. We were interested in confusion engendered by the treatments on the participants, and conducted a within-subjects study in the actual work environment, using eye-tracking and talk-aloud data collection. We coded the verbal data into classes of informativeness and confusion and correlated it with fixations and durations on the Areas of Interests recorded by the eye-tracking device. We used various analysis techniques, including Mann-Whitney, regression, and Levenshtein distance, to investigate how confused users differed …

Go to article

Automatic Persona Generation (Apg): A Rationale And Demonstration, Soon-Gyo Jung, Joni Salminen, Haewoon Kwak, Jisun An, Bernard J Jansen Mar 2018

Automatic Persona Generation (Apg): A Rationale And Demonstration, Soon-Gyo Jung, Joni Salminen, Haewoon Kwak, Jisun An, Bernard J Jansen

Research Collection School Of Computing and Information Systems

We present Automatic Persona Generation (APG), a methodology and system for quantitative persona generation using large amounts of online social media data. The system is operational, beta deployed with several client organizations in multiple industry verticals and ranging from small-to-medium sized enterprises to large multi-national corporations. Using a robust web framework and stable back-end database, APG is currently processing tens of millions of user interactions with thousands of online digital products on multiple social media platforms, such as Facebook and YouTube. APG identifies both distinct and impactful user segments and then creates persona descriptions by automatically adding pertinent features, such …

Go to article

W-Air: Enabling Personal Air Pollution Monitoring On Wearables, Balz Maag, Zimu Zhou, Lothar Thiele Mar 2018

W-Air: Enabling Personal Air Pollution Monitoring On Wearables, Balz Maag, Zimu Zhou, Lothar Thiele

Research Collection School Of Computing and Information Systems

Accurate, portable and personal air pollution sensing devices enable quantification of individual exposure to air pollution, personalized health advice and assistance applications. Wearables are promising (e.g., on wristbands, attached to belts or backpacks) to integrate commercial off-the-shelf gas sensors for personal air pollution sensing. Yet previous research lacks comprehensive investigations on the accuracies of air pollution sensing on wearables. In response, we proposed W-Air, an accurate personal multi-pollutant monitoring platform for wearables. We discovered that human emissions introduce non-linear interference when low-cost gas sensors are integrated into wearables, which is overlooked in existing studies. W-Air adopts a sensor-fusion calibration scheme …

Go to article

An Efficient And Privacy-Preserving Biometric Identification Scheme In Cloud Computing, Liehuang Zhu, Chuan Zhang, Chang Xu, Ximeng Liu, Cheng Huang Mar 2018

An Efficient And Privacy-Preserving Biometric Identification Scheme In Cloud Computing, Liehuang Zhu, Chuan Zhang, Chang Xu, Ximeng Liu, Cheng Huang

Research Collection School Of Computing and Information Systems

Biometric identification has become increasingly popular in recent years.With the development of cloud computing, database owners are motivated to outsource the large size of biometric data and identification tasks to the cloud to get rid of the expensive storage and computation costs, which, however, brings potential threats to users’ privacy. In this paper, we propose an efficient and privacy-preserving biometric identification outsourcing scheme. Specifically, the biometric: To execute a biometric identification, the database owner encrypts the query data and submits it to the cloud. The cloud performs identification operations over the encrypted database and returns the result to the database …

Go to article

A New Revocable And Re-Delegable Proxy Signature And Its Application, Shengmin Xu, Guomin Yang, Yi Mu Mar 2018

A New Revocable And Re-Delegable Proxy Signature And Its Application, Shengmin Xu, Guomin Yang, Yi Mu

Research Collection School Of Computing and Information Systems

With the popularity of cloud computing and mobile Apps, on-demand services such as on-line music or audio streaming and vehicle booking are widely available nowadays. In order to allow efficient delivery and management of the services, for large-scale on-demand systems, there is usually a hierarchy where the service provider can delegate its service to a top-tier (e.g., countrywide) proxy who can then further delegate the service to lower level (e.g., region-wide) proxies. Secure (re-)delegation and revocation are among the most crucial factors for such systems. In this paper, we investigate the practical solutions for achieving re-delegation and revocation utilizing proxy …

Go to article

Vmkdo: Verifiable Multi-Keyword Search Over Encrypted Cloud Data For Dynamic Data-Owner, Yibin Miao, Jianfeng Ma, Ximeng Liu, Zhiquan Liu, Limin Shen, Fushan Wei Mar 2018

Vmkdo: Verifiable Multi-Keyword Search Over Encrypted Cloud Data For Dynamic Data-Owner, Yibin Miao, Jianfeng Ma, Ximeng Liu, Zhiquan Liu, Limin Shen, Fushan Wei

Research Collection School Of Computing and Information Systems

The advantages of cloud computing encourage individuals and enterprises to outsource their local data storage and computation to cloud server, however, data security and privacy concerns seriously hinder the practicability of cloud storage. Although searchable encryption (SE) technique enables cloud server to provide fundamental encrypted data retrieval services for data-owners, equipping with a result verification mechanism is still of prime importance in practice as semi-trusted cloud server may return incorrect search results. Besides, single keyword search inevitably incurs many irrelevant results which result in waste of bandwidth and computation resources. In this paper, we are among the first to tackle …

Go to article

Attributed Social Network Embedding, Lizi Liao, Xiangnan He, Hanwang Zhang, Tat-Seng Chua Mar 2018

Attributed Social Network Embedding, Lizi Liao, Xiangnan He, Hanwang Zhang, Tat-Seng Chua

Research Collection School Of Computing and Information Systems

Embedding network data into a low-dimensional vector space has shown promising performance for many real-world applications, such as node classification and entity retrieval. However, most existing methods focused only on leveraging network structure. For social networks, besides the network structure, there also exists rich information about social actors, such as user profiles of friendship networks and textual content of citation networks. These rich attribute information of social actors reveal the homophily effect, exerting huge impacts on the formation of social networks. In this paper, we explore the rich evidence source of attributes in social networks to improve network embedding. We …

Go to article

Constant-Size Ciphertexts In Threshold Attribute-Based Encryption Without Dummy Attributes, Willy Susilo, Guomin Yang, Fuchun Guo, Qiong Huang Mar 2018

Constant-Size Ciphertexts In Threshold Attribute-Based Encryption Without Dummy Attributes, Willy Susilo, Guomin Yang, Fuchun Guo, Qiong Huang

Research Collection School Of Computing and Information Systems

Attribute-based encryption (ABE) is an augmentation of public key encryption that allows users to encrypt and decrypt messages based on users' attributes. In a (t, s) threshold ABE, users who can decrypt a ciphertext must hold at least t attributes among the s attributes specified by the encryptor. At PKC 2010, Herranz, Laguillaumie and Raft& proposed the first threshold ABE with constant-size ciphertexts. In order to ensure the encryptor can flexibly select the attribute set and a threshold value, they use dummy attributes to satisfy the decryption requirement. The advantage of their scheme is that any addition or removal of …

Go to article

A Model Of Competition Between Perpetual Software And Software As A Service, Zhiling Guo, Dan Ma Mar 2018

A Model Of Competition Between Perpetual Software And Software As A Service, Zhiling Guo, Dan Ma

Research Collection School Of Computing and Information Systems

Duplicate record, see https://ink.library.smu.edu.sg/sis_research/3744/. Software as a service (SaaS) has grown to be a significant segment of many software product markets. SaaS vendors, which charge customers based on use and continuously improve the quality of their products, have put competitive pressure on traditional perpetual software vendors, which charge a licensing fee and periodically upgrade the quality of their software. We develop an analytical model to study the competitive pricing strategies of an incumbent perpetual software vendor in the presence of a SaaS competitor. We find that, depending on both the SaaS quality improvement rate and the network effect, the …

Go to article

A Model Of Competition Between Perpetual Software And Software As A Service, Zhiling Guo, Dan Ma Mar 2018

A Model Of Competition Between Perpetual Software And Software As A Service, Zhiling Guo, Dan Ma

Research Collection School Of Computing and Information Systems

Software as a service (SaaS) has grown to be a significant segment of many software product markets. SaaS vendors, which charge customers based on use and continuously improve the quality of their products, have put competitive pressure on traditional perpetual software vendors, which charge a licensing fee and periodically upgrade the quality of their software. We develop an analytical model to study the competitive pricing strategies of an incumbent perpetual software vendor in the presence of a SaaS competitor. We find that, depending on both the SaaS quality improvement rate and the network effect, the perpetual software vendor adopts one …

Go to article

Multigeneration Product Diffusion In The Presence Of Strategic Consumers, Zhiling Guo, Jianqing Chen Mar 2018

Multigeneration Product Diffusion In The Presence Of Strategic Consumers, Zhiling Guo, Jianqing Chen

Research Collection School Of Computing and Information Systems

Frequent new product releases pose significant challenges for firms as they manage successive generations of product diffusion. We develop an analytical model to study the effect of different purchase options by strategic consumers on a firm's profit and the firm's strategies for the timing and pricing of its successive generations of product diffusion. We show that consumers' strategic behavior, although adversely affecting the sales of the first-generation product, positively influences the sales of the second-generation product through an initial “seeding” effect. The influence of strategic consumers on profit and sales depends largely on the discount-to-price ratio of the first generation …

Go to article

Do Your Friends Make You Buy This Brand?: Modeling Social Recommendation With Topics And Brands, Minh Duc Luu, Ee Peng Lim Mar 2018

Do Your Friends Make You Buy This Brand?: Modeling Social Recommendation With Topics And Brands, Minh Duc Luu, Ee Peng Lim

Research Collection School Of Computing and Information Systems

Consumer behavior and marketing research have shown that brand has significant influence on product reviews and product purchase decisions. However, there is very little work on incorporating brand related factors into product recommender systems. Meanwhile, the similarity in brand preference between a user and other socially connected users also affects her adoption decisions. To integrate seamlessly the individual and social brand related factors into the recommendation process, we propose a novel model called Social Brand–Item–Topic (SocBIT). As the original SocBIT model does not enforce non-negativity, which poses some difficulty in result interpretation, we also propose a non-negative version, called SocBIT(Formula …

Go to article

An Efficient And Expressive Ciphertext-Policy Attribute-Based Encryption Scheme With Partially Hidden Access Structures, Revisited, Hui Cui, Robert H. Deng, Junzuo Lai, Xun Yi, Surya Nepal Mar 2018

An Efficient And Expressive Ciphertext-Policy Attribute-Based Encryption Scheme With Partially Hidden Access Structures, Revisited, Hui Cui, Robert H. Deng, Junzuo Lai, Xun Yi, Surya Nepal

Research Collection School Of Computing and Information Systems

Ciphertext-policy attribute-based encryption (CP-ABE) has been regarded as one of the promising solutions to protect data security and privacy in cloud storage services. In a CP-ABE scheme, an access structure is included in the ciphertext, which, however, may leak sensitive information about the underlying plaintext and the privileged recipients in that anyone who sees the ciphertext is able to learn the attributes of the privileged recipients from the associated access structure. In order to address this issue, CP-ABE with partially hidden access structures was introduced where each attribute is divided into an attribute name and an attribute value and the …

Go to article

Obfuscation At-Source: Privacy In Context-Aware Mobile Crowd-Sourcing, Thivya Kandappu, Archan Misra, Shih-Fen Cheng, Randy Tandriansyah, Hoong Chuin Lau Mar 2018

Obfuscation At-Source: Privacy In Context-Aware Mobile Crowd-Sourcing, Thivya Kandappu, Archan Misra, Shih-Fen Cheng, Randy Tandriansyah, Hoong Chuin Lau

Research Collection School Of Computing and Information Systems

By effectively reaching out to and engaging larger population of mobile users, mobile crowd-sourcing has become a strategy to perform large amount of urban tasks. The recent empirical studies have shown that compared to the pull-based approach, which expects the users to browse through the list of tasks to perform, the push-based approach that actively recommends tasks can greatly improve the overall system performance. As the efficiency of the push-based approach is achieved by incorporating worker's mobility traces, privacy is naturally a concern. In this paper, we propose a novel, 2-stage and user-controlled obfuscation technique that provides a trade off-amenable …

Go to article

Scaling Human Activity Recognition Via Deep Learning-Based Domain Adaptation, Md Abdullah Hafiz Khan, Nirmalya Roy, Archan Misra Mar 2018

Scaling Human Activity Recognition Via Deep Learning-Based Domain Adaptation, Md Abdullah Hafiz Khan, Nirmalya Roy, Archan Misra

Research Collection School Of Computing and Information Systems

We investigate the problem of making human activityrecognition (AR) scalable–i.e., allowing AR classifiers trainedin one context to be readily adapted to a different contextualdomain. This is important because AR technologies can achievehigh accuracy if the classifiers are trained for a specific individualor device, but show significant degradation when the sameclassifier is applied context–e.g., to a different device located ata different on-body position. To allow such adaptation withoutrequiring the onerous step of collecting large volumes of labeledtraining data in the target domain, we proposed a transductivetransfer learning model that is specifically tuned to the propertiesof convolutional neural networks (CNNs). Our model, …

Go to article

Engagemon: Multi-Modal Engagement Sensing For Mobile Games, Sinh Huynh, Seungmin Kim, Jeonggil Ko, Rajesh Krishna Balan, Youngki Lee Mar 2018

Engagemon: Multi-Modal Engagement Sensing For Mobile Games, Sinh Huynh, Seungmin Kim, Jeonggil Ko, Rajesh Krishna Balan, Youngki Lee

Research Collection School Of Computing and Information Systems

Understanding the engagement levels players have with a game is a useful proxy for evaluating the game design and user experience. This is particularly important for mobile games as an alternative game is always just an easy download away. However, engagement is a subjective concept and usually requires fine-grained highly disruptive interviews or surveys to determine accurately. In this paper, we present EngageMon, a first-of-its-kind system that uses a combination of sensors from the smartphone (touch events), a wristband (photoplethysmography and electrodermal activity sensor readings), and an external depth camera (skeletal motion information) to accurately determine the engagement level of …

Go to article

An Lstm Model For Cloze-Style Machine Comprehension, Shuohang Wang, Jing Jiang Mar 2018

An Lstm Model For Cloze-Style Machine Comprehension, Shuohang Wang, Jing Jiang

Research Collection School Of Computing and Information Systems

Machine comprehension is concerned with teaching machines to answer reading comprehension questions. In this paper we adopt an LSTM-based model we designed earlier for textual entailment and propose two new models for cloze-style machine comprehension. In our first model, we treat the document as a premise and the question as a hypothesis, and use an LSTM with attention mechanisms to match the question with the document. This LSTM remembers the best answer token found in the document while processing the question. Furthermore, we observe some special properties of machine comprehension and propose a two-layer LSTM model. In this model, we …

Go to article

Sclib: A Practical And Lightweight Defense Against Component Hijacking In Android Applications, Daoyuan Wu, Yao Cheng, Debin Gao, Yingjiu Li, Robert H. Deng Mar 2018

Sclib: A Practical And Lightweight Defense Against Component Hijacking In Android Applications, Daoyuan Wu, Yao Cheng, Debin Gao, Yingjiu Li, Robert H. Deng

Research Collection School Of Computing and Information Systems

Cross-app collaboration via inter-component communication is a fundamental mechanism on Android. Although it brings the benefits such as functionality reuse and data sharing, a threat called component hijacking is also introduced. By hijacking a vulnerable component in victim apps, an attack app can escalate its privilege for operations originally prohibited. Many prior studies have been performed to understand and mitigate this issue, but no defense is being deployed in the wild, largely due to the deployment difficulties and performance concerns. In this paper we present SCLib, a secure component library that performs in-app mandatory access control on behalf of app …

Go to article

Mining Sandboxes: Are We There Yet?, Lingfeng Bao, Tien Duy B. Le, David Lo Mar 2018

Mining Sandboxes: Are We There Yet?, Lingfeng Bao, Tien Duy B. Le, David Lo

Research Collection School Of Computing and Information Systems

The popularity of Android platform on mobile devices has attracted much attention from many developers and researchers, as well as malware writers. Recently, Jamrozik et al. proposed a technique to secure Android applications referred to as mining sandboxes. They used an automated test case generation technique to explore the behavior of the app under test and then extracted a set of sensitive APIs that were called. Based on the extracted sensitive APIs, they built a sandbox that can block access to APIs not used during testing. However, they only evaluated the proposed technique with benign apps but not investigated whether …

Go to article

Urlnet: Learning A Url Representation With Deep Learning For Malicious Url Detection, Hung Le, Hong Quang Pham, Doyen Sahoo, Steven C. H. Hoi Mar 2018

Urlnet: Learning A Url Representation With Deep Learning For Malicious Url Detection, Hung Le, Hong Quang Pham, Doyen Sahoo, Steven C. H. Hoi

Research Collection School Of Computing and Information Systems

Malicious URLs host unsolicited content and are used to perpetrate cybercrimes. It is imperative to detect them in a timely manner. Traditionally, this is done through the usage of blacklists, which cannot be exhaustive, and cannot detect newly generated malicious URLs. To address this, recent years have witnessed several efforts to perform Malicious URL Detection using Machine Learning. The most popular and scalable approaches use lexical properties of the URL string by extracting Bag-of-words like features, followed by applying machine learning models such as SVMs. There are also other features designed by experts to improve the prediction performance of the …

Go to article

An Efficient And Expressive Ciphertext-Policy Attribute-Based Encryption Scheme With Partially Hidden Access Structures, Revisited, Hui Cui, Robert H. Deng, Junzuo Lai, Xun Yi, Surya Nepal Mar 2018

An Efficient And Expressive Ciphertext-Policy Attribute-Based Encryption Scheme With Partially Hidden Access Structures, Revisited, Hui Cui, Robert H. Deng, Junzuo Lai, Xun Yi, Surya Nepal

Research Collection School Of Computing and Information Systems

Ciphertext-policy attribute-based encryption (CP-ABE) has been regarded as one of the promising solutions to protect data security and privacy in cloud storage services. In a CP-ABE scheme, an access structure is included in the ciphertext, which, however, may leak sensitive information about the underlying plaintext and the privileged recipients in that anyone who sees the ciphertext is able to learn the attributes of the privileged recipients from the associated access structure. In order to address this issue, CP-ABE with partially hidden access structures was introduced where each attribute is divided into an attribute name and an attribute value and the …

Go to article

The Way You Move: The Effect Of A Robot Surrogate Movement In Remote Collaboration, Martin Feick, Lora Oehlberg, Anthony Tang, André Miede, Ehud Sharlin Mar 2018

The Way You Move: The Effect Of A Robot Surrogate Movement In Remote Collaboration, Martin Feick, Lora Oehlberg, Anthony Tang, André Miede, Ehud Sharlin

Research Collection School Of Computing and Information Systems

In this paper, we discuss the role of the movement trajectory and velocity enabled by our tele-robotic system (ReMa) for remote collaboration on physical tasks. Our system reproduces changes in object orientation and position at a remote location using a humanoid robotic arm. However, even minor kinematics differences between robot and human arm can result in awkward or exaggerated robot movements. As a result, user communication with the robotic system can become less efficient, less fluent and more time intensive.

Go to article

Education In The Age Of Artificial Intelligence: How Will Technology Shape Learning?, Keng Siau Mar 2018

Education In The Age Of Artificial Intelligence: How Will Technology Shape Learning?, Keng Siau

Research Collection School Of Computing and Information Systems

The age of Artificial Intelligence (AI) is here! Higher education needs to prepare students for a world in which AI plays an increasingly dominant role. What are the jobs that can be replaced easily? How would higher education be affected in the age of AI, robotics, machine learning, and automation? How can higher education excel and flourish in the age of AI?

Go to article

Building Trust In Artificial Intelligence, Machine Learning, And Robotics, Keng Siau, Weiyu Wang Mar 2018

Building Trust In Artificial Intelligence, Machine Learning, And Robotics, Keng Siau, Weiyu Wang

Research Collection School Of Computing and Information Systems

In this article, we look at trust in artificial intelligence, machine learning (ML), and robotics. We first review the concept of trust in AI and examine how trust in AI may be different from trust in other technologies. We then discuss the differences between interpersonal trust and trust in technology and suggest factors that are crucial in building initial trust and developing continuous trust in artificial intelligence.

Go to article

Hiddencode: Hidden Acoustic Signal Capture With Vibration Energy Harvesting, Guohao Lan, Dong Ma, Mahbub Hassan, Wen Hu Mar 2018

Hiddencode: Hidden Acoustic Signal Capture With Vibration Energy Harvesting, Guohao Lan, Dong Ma, Mahbub Hassan, Wen Hu

Research Collection School Of Computing and Information Systems

The feasibility of using vibration energy harvesting (VEH) as an energy-efficient receiver for short-range acoustic data communication has been investigated recently. When data was encoded in acoustic signal within the energy harvesting frequency band and transmitted through a speaker, a VEH receiver was capable of decoding the data by processing the harvested energy signal. Although previous work created new opportunities for simultaneous energy harvesting and communication using the same hardware, the communication makes annoying sounds as the energy harvesting frequency band lies within the sensitive region of human auditory system. In this work, we present a novel modulation scheme to …

Go to article

Visualizing Research Impact Through Citation Data, Yong Wang, Conglei Shi, Liangyue Li, Hanghang Tong, Huamin Qu Mar 2018

Visualizing Research Impact Through Citation Data, Yong Wang, Conglei Shi, Liangyue Li, Hanghang Tong, Huamin Qu

Research Collection School Of Computing and Information Systems

Research impact plays a critical role in evaluating the research quality and influence of a scholar, a journal, or a conference. Many researchers have attempted to quantify research impact by introducing different types of metrics based on citation data, such as h-index, citation count, and impact factor. These metrics are widely used in the academic community. However, quantitative metrics are highly aggregated in most cases and sometimes biased, which probably results in the loss of impact details that are important for comprehensively understanding research impact. For example, which research area does a researcher have great research impact on? How does …

Go to article

Compressive Representation For Device-Free Activity Recognition With Passive Rfid Signal Strength, Lina Yao, Quan Z. Sheng, Xue Li, Tao Gu, Mingkui Tan, Xianzhi Wang, Sen Wang, Wenjie Ruan Feb 2018

Compressive Representation For Device-Free Activity Recognition With Passive Rfid Signal Strength, Lina Yao, Quan Z. Sheng, Xue Li, Tao Gu, Mingkui Tan, Xianzhi Wang, Sen Wang, Wenjie Ruan

Research Collection School Of Computing and Information Systems

Understanding and recognizing human activities is a fundamental research topic for a wide range of important applications such as fall detection and remote health monitoring and intervention. Despite active research in human activity recognition over the past years, existing approaches based on computer vision or wearable sensor technologies present several significant issues such as privacy (e.g., using video camera to monitor the elderly at home) and practicality (e.g., not possible for an older person with dementia to remember wearing devices). In this paper, we present a low-cost, unobtrusive, and robust system that supports independent living of older people. The system …

Go to article

Things You May Not Know About Android (Un)Packers: A Systematic Study Based On Whole-System Emulation, Yue Duan, Mu Zhang, Abhishek Vasist Bhaskar, Heng Yin, Xiaorui Pan, Tongxin Li, Xueqiang Wang, Xiaofeng Wang Feb 2018

Things You May Not Know About Android (Un)Packers: A Systematic Study Based On Whole-System Emulation, Yue Duan, Mu Zhang, Abhishek Vasist Bhaskar, Heng Yin, Xiaorui Pan, Tongxin Li, Xueqiang Wang, Xiaofeng Wang

Research Collection School Of Computing and Information Systems

The prevalent usage of runtime packers has complicated Android malware analysis, as both legitimate and malicious apps are leveraging packing mechanisms to protect themselves against reverse engineer. Although recent efforts have been made to analyze particular packing techniques, little has been done to study the unique characteristics of Android packers. In this paper, we report the first systematic study on mainstream Android packers, in an attempt to understand their security implications. For this purpose, we developed DROIDUNPACK, a whole-system emulation based Android packing analysis framework, which compared with existing tools, relies on intrinsic characteristics of Android runtime (rather than heuristics), …

Go to article

Building Deep Networks On Grassmann Manifolds, Zhiwu Huang, J. Wu, Gool L. Van Feb 2018

Building Deep Networks On Grassmann Manifolds, Zhiwu Huang, J. Wu, Gool L. Van

Research Collection School Of Computing and Information Systems

Learning representations on Grassmann manifolds is popular in quite a few visual recognition tasks. In order to enable deep learning on Grassmann manifolds, this paper proposes a deep network architecture by generalizing the Euclidean network paradigm to Grassmann manifolds. In particular, we design full rank mapping layers to transform input Grassmannian data to more desirable ones, exploit re-orthonormalization layers to normalize the resulting matrices, study projection pooling layers to reduce the model complexity in the Grassmannian context, and devise projection mapping layers to respect Grassmannian geometry and meanwhile achieve Euclidean forms for regular output layers. To train the Grassmann networks, …

Go to article

Enhanced Vireo Kis At Vbs 2018, Phuong Anh Nguyen, Yi-Jie Lu, Hao Zhang, Chong-Wah Ngo Feb 2018

Enhanced Vireo Kis At Vbs 2018, Phuong Anh Nguyen, Yi-Jie Lu, Hao Zhang, Chong-Wah Ngo

Research Collection School Of Computing and Information Systems

The VIREO Known-Item Search (KIS) system has joined the Video Browser Showdown (VBS) [1] evaluation benchmark for the first time in year 2017. With experiences learned, the second version of VIREO KIS is presented in this paper. Considering the color-sketch based retrieval, we propose a simple grid-based approach for color query. This method allows the aggregation of color distributions in video frames into a shot representation, and generates the pre-computed rank list for all available queries which reduces computational resources and favors a recommendation module. With focusing on concept based retrieval, we modify our multimedia event detection system at TRECVID …

Go to article

First Prev Next Last