Computer Sciences Commons^™

Cross-View Graph Embedding, Zhiwu Huang, S. Shan, H. Zhang, S. Lao, X. Chen

Research Collection School Of Computing and Information Systems

Recently, more and more approaches are emerging to solve the cross-view matching problem where reference samples and query samples are from different views. In this paper, inspired by Graph Embedding, we propose a unified framework for these cross-view methods called Cross-view Graph Embedding. The proposed framework can not only reformulate most traditional cross-view methods (e.g., CCA, PLS and CDFE), but also extend the typical single-view algorithms (e.g., PCA, LDA and LPP) to cross-view editions. Furthermore, our general framework also facilitates the development of new cross-view methods. In this paper, we present a new algorithm named Cross-view Local Discriminant Analysis (CLODA) …

Snap-And-Ask: Answering Multimodal Question By Naming Visual Instance, Wei Zhang, Lei Pang, Chong-Wah Ngo

Research Collection School Of Computing and Information Systems

In real-life, it is easier to provide a visual cue when asking a question about a possibly unfamiliar topic, for example, asking the question, “Where was this crop circle found?”. Providing an image of the instance is far more convenient than texting a verbose description of the visual properties, especially when the name of the query instance is not known. Nevertheless, having to identify the visual instance before processing the question and eventually returning the answer makes multimodal question-answering technically challenging. This paper addresses the problem of visual-totext naming through the paradigm of answering-by-search in a two-stage computational framework, which …

A Theoretical Framework For Serious Game Design: Exploring Pedagogy, Play And Fidelity And Their Implications For The Design Process, Pauline Rooney

Journal Articles

It is widely acknowledged that digital games can provide an engaging, motivating and “fun” experience for students. However an entertaining game does not necessarily constitute a meaningful, valuable learning experience. For this reason, experts espouse the importance of underpinning serious games with a sound theoretical framework which integrates and balances theories from two fields of practice: pedagogy and game design (Kiili, 2005; Seeney & Routledge, 2009). Additionally, with the advent of sophisticated, immersive technologies, and increasing interest in the opportunities for constructivist learning offered by these technologies, concepts of fidelity and its impact on student learning and engagement, have emerged …

3dme: 3d Media Express From Rgb-D Images, Tam Nguyen, Lusong Li, Jun Tan, Shuicheng Yan

Computer Science Faculty Publications

Considering the continuously increasing availability and accessibility of 3D media and the depth camera such as Kinect, we demonstrate an innovative 3D media system called 3DME. The objective of this demo is three-fold.

First, the demo exhibits the creation of 3D images from RGB-D images. Second, 3DME allows a user to insert impressive effects to the produced 3D content. Last but not least, our demo is one of the first attempts toward advertising with 3D content.

Hi, Magic Closet, Tell Me What To Wear!, Si Liu, Tam Nguyen, Jiashi Feng, Meng Wang, Shuicheng Yan

Computer Science Faculty Publications

In this demo, we present a practical system, "magic closet," for automatic occasion-oriented clothing pairing. Given a user-input occasion, e.g., wedding or shopping, the magic closet intelligently and automatically pairs the user-specified reference clothing (upper body or lower body) with the most suitable one from online shops. Two key criteria are explicitly considered for the magic closet system. One criterion is to dress properly, e.g., compared to suit pants, it is more decent to wear a cocktail dress for a banquet occasion. The other criterion is to dress aesthetically, e.g., a red T-shirt matches better with white pants than with …

Sense Beauty Via Face, Dressing, And/Or Voice, Tam Nguyen, Si Liu, Bingbing Ni, Jun Tan, Yong Rui, Shuicheng Yan

Computer Science Faculty Publications

Discovering the secret of beauty has been the pursuit of artists and philosophers for centuries. Nowadays, the computational model for beauty estimation has been actively explored in computer science community, yet with the focus mainly on facial features. In this work, we perform a comprehensive study of female attractiveness conveyed by single/multiple modalities of cues, i.e., face, dressing, and/or voice, and aim to uncover how different modalities individually and collectively affect the human sense of beauty. To this end, we collect the first Multi-Modality Beauty (M²B) dataset in the world for female attractiveness study, which is thoroughly annotated …

The Shanghai-Hongkong Team At Mediaeval2012: Violent Scene Detection Using Trajectory-Based Features, Yu-Gang Jiang, Qi Dai, Chun Chet Tan, Xiangyang Xue, Chong-Wah Ngo

Research Collection School Of Computing and Information Systems

The Violent Scene Detection task offers a very practical challenge in detecting complex and diverse violent video clips in movies. In this working note paper, we will briefly describe our system and discuss the results, which achieved top performance in mAP@201 and runner-up in mAP@100, among all 35 submissions worldwide. The central component of our system is a set of features derived from the appearance and motion of local patch trajectories [2]. We use these features and SVM classifier as the baseline approach and add in a few other components to further improve the performance. Our findings indicate that the …

Understanding Engagement In Educational Computer Games, Fiona Fui-Hoon Nah, Yunjie Zhou, Adeline Boey, Hanji Li

Research Collection School Of Computing and Information Systems

This paper presents an empirical study to understand engagement in educational computer games. Engagement is defined as an experience that occupies an individual’s attention and captures one’s interest. The nature of engagement is viewed as comprising conditions, actions, and outcomes of engagement. The data collection method includes in-depth interviews with 12 educational computer game players who have experienced engagement in playing these games. We used the Grounded Theory (GT) approach to develop an understanding of user engagement in the computer game-based learning context.

Eating Alone, Together: New Forms Of Commensality, Catherine Grevet, Anthony Tang, Elizabeth Mynatt

Research Collection School Of Computing and Information Systems

Eating with others, or commensality, is an enjoyable activity that serves many important social functions; however, many individuals eat meals alone due to life circumstances, meaning that they miss out on these social benefits. We developed and deployed a simple technology probe providing social awareness around mealtimes to explore how social systems might help alleviate the loneliness of solitary dining. Our findings suggest that these systems can convey a sense of connectedness around a meal; further, our analysis revealed three themes relevant to systems of this type: that contextually-located peripheral awareness engenders connectedness; that such tools can foster a feeling …

Trajectory-Based Modeling Of Human Actions With Motion Reference Points, Yu-Gang Jiang, Qi Dai, Xiangyang Xue, Wei Liu, Chong-Wah Ngo

Research Collection School Of Computing and Information Systems

Human action recognition in videos is a challenging problem with wide applications. State-of-the-art approaches often adopt the popular bag-of-features representation based on isolated local patches or temporal patch trajectories, where motion patterns like object relationships are mostly discarded. This paper proposes a simple representation specifically aimed at the modeling of such motion relationships. We adopt global and local reference points to characterize motion information, so that the final representation can be robust to camera movement. Our approach operates on top of visual codewords derived from local patch trajectories, and therefore does not require accurate foreground-background separation, which is typically a …

Extending A Mobile Device's Interaction Space Through Body-Centric Interaction, Xiang Chen, Nicolai Marquardt, Anthony Tang, Sebastian Boring, Saul Greenberg

Research Collection School Of Computing and Information Systems

Modern mobile devices rely on the screen as a primary input modality. Yet the small screen real-estate limits interaction possibilities, motivating researchers to explore alternate input techniques. Within this arena, our goal is to develop Body-Centric Interaction with Mobile Devices: a class of input techniques that allow a person to position and orient her mobile device to navigate and manipulate digital content anchored in the space on and around the body. To achieve this goal, we explore such interaction in a bottom-up path of prototypes and implementations. From our experiences, as well as by examining related work, we discuss and …

The Fat Thumb: Using The Thumb's Contact Size For Single-Handed Mobile Interaction, Sebastian Boring, David Ledo, Xiang ‘Anthony’ Chen, Anthony Tang, Anthony Tang, Saul Greenberg

Research Collection School Of Computing and Information Systems

Modern mobile devices allow a rich set of multi-finger interactions that combine modes into a single fluid act. Such gestures may require the use of both hands: one holding the device while the other is interacting. While on the go, however, only one hand may be available to both hold the device and interact with it. In this demo, we present the Fat Thumb interaction technique, which uses the thumb's contact size as a form of simulated pressure. We present how this can be used, for example, to integrate panning and zooming into a single interaction. Contact size determines the …

Sampling And Ontologically Pooling Web Images For Visual Concept Learning, Shiai Zhu, Chong-Wah Ngo, Yu-Gang Jiang

Research Collection School Of Computing and Information Systems

Sufficient training examples are essential for effective learning of semantic visual concepts. In practice, however, acquiring noise-free training examples has always been expensive. Recently the rapid popularization of social media websites, such as Flickr, has made it possible to collect training exemplars without human assistance. This paper proposes a novel and efficient approach to collect training samples from the noisily tagged Web images for visual concept learning, where we try to maximize two important criteria, relevancy and coverage, of the automatically generated training sets. For the former, a simple method named semantic field is introduced to handle the imprecise and …

Eliciting A Sensemaking Process From Verbal Protocols Of Reverse Engineers, Adam R. Bryant, Robert F. Mills, Gilbert L. Peterson, Michael R. Grimaila

Faculty Publications

A process of sensemaking in reverse engineering was elicited from verbal protocols of reverse engineers as they investigated the assembly code of executable programs. Four participants were observed during task performance and verbal protocols were collected and analyzed from two of the participants to determine their problem-solving states and characterize likely transitions between those states. From this analysis, a high-level process of sensemaking is described which represents hypothesis generation and information-seeking behaviors in reverse engineering within a framework of goal-directed planning. Future work in validation and application of the process is discussed.

User Assisted Tree Reconstruction From Point Clouds, William P. Leavenworth Ii

Department of Computer Graphics Technology Degree Theses

LIDAR is a useful tool for quickly digitalizing real world objects, but it usually takes some effort to produce a recognizable object from the raw input. When the object is a tree, the challenge is to create a three‐dimensional model that represents its general shape, while avoiding the influence of undersampling and noise. In the method developed for this research, the user creates a sketch overlaying a display of the raw input data. Each node in the sketch creates an estimated slice of the tree skeleton at that point, and the slices are connected according to the connectivity of the …

The Utilization Of Context To Adapt The Interface Of A Computer, Ludovic Delaveau

Purdue Polytechnic Masters Theses

This research examines the possibility for a computer to adapt its interface and its settings to the current environment, through the notion of situations and corresponding profiles. These situations are group of context elements and tasks or activities performed in that operating conditions, such as the activities conducted at the workplace or the relaxation time at home. The profiles are set of configuration values for the computer, such as the network services. Adapting the computer serves the user as it takes care of putting his computer in the right mode to be efficient in the activity performed. In order to …

Real Time Stereo Vision Using Dynamic Programming On A Graphics Processing Unit (Gpu), Janet Marie Mendoza, Rodrigo Antonio Romero, Olac Fuentes

COURI Symposium Abstracts, Summer 2012

Stereo vision is an important research topic in computer vision. It attempts to emulate the human vision system by using two or more images of the same scene to estimate depth information in three dimensions. The development of this topic is of much importance to many applications that vary from medical to entertainment. Currently, many researchers are searching for a method that will increase the speed of the process and still obtain high-quality results. Dynamic programming, which is a method for solving complex problems by breaking them down into simpler sub problems, has been successfully applied in stereo vision. We …

The Effects Of Diagrams And Relational Complexity On User Performance In Conditional Probability Problems In A Non-Learning Context, Vincent J. Kellen

College of Computing and Digital Media Dissertations

Many disciplines in everyday life depend on improved performance in conditional probability problems. Most adults struggle with conditional probability problems and several prior studies have shown participant accuracy is less than 50%. This study examined user performance when aided with computer-generated Venn and Euler type diagrams in a non-learning context. Despite the prevalence of research into diagrams and extensive research into conditional probability problem solving, this study is one of the only studies to apply theories of working memory to predict user performance in conditional probability problems with diagrams. Following relational complexity theory, this study manipulated problem complexity in computer …

Tacky Golf, Robert Crosby

Computer Science and Software Engineering

This project implements a simple miniature golf game in 3d for the iPhone. Using a modular approach the game engine was written in several modules to handle its various functions and for cross platform potability. Four major modules are described in detail about their design choices and how they work. Ultimately the game and its engine are intended to be sold on the Apple App store and to be used for future games and applications on the iPhone.

Beaglebone Webcam Server, Alexander Corcoran

Computer Engineering

The Beaglebone Webcam Server is a Linux based IP webcam, based on an inexpensive ARM development board, which hosts its own web server to display the webcam feed. The server has the ability to either connect to a wired router, or to act as a wireless access point in order for users to connect and control its functions via any Wi-Fi enabled device.

Aligning Images In The Wild, Wen-Yan Lin, Linlin Liu, Yasuyuki Matsushita, Kok-Lim Low, Siying Liu

Research Collection School Of Computing and Information Systems

Aligning image pairs with significant appearance change is a long standing computer vision challenge. Much of this problem stems from the local patch descriptors’ instability to appearance variation. In this paper we suggest this instability is due less to descriptor corruption and more the difficulty in utilizing local information to canonically define the orientation (scale and rotation) at which a patch’s descriptor should be computed. We address this issue by jointly estimating correspondence and relative patch orientation, within a hierarchical algorithm that utilizes a smoothly varying parameterization of geometric transformations. By collectively estimating the correspondence and orientation of all the …

Gpu-Accelerated Point-Based Color Bleeding, Ryan Daniel Schmitt

Master's Theses

Traditional global illumination lighting techniques like Radiosity and Monte Carlo sampling are computationally expensive. This has prompted the development of the Point-Based Color Bleeding (PBCB) algorithm by Pixar in order to approximate complex indirect illumination while meeting the demands of movie production; namely, reduced memory usage, surface shading independent run time, and faster renders than the aforementioned lighting techniques.

The PBCB algorithm works by discretizing a scene’s directly illuminated geometry into a point cloud (surfel) representation. When computing the indirect illumination at a point, the surfels are rasterized onto cube faces surrounding that point, and the constituent pixels are combined …

Semi-Supervised Hierarchical Clustering For Personalized Web Image Organization, Lei Meng, Ah-Hwee Tan

Research Collection School Of Computing and Information Systems

Existing efforts on web image organization usually transform the task into surrounding text clustering. However, Current text clustering algorithms do not address the problem of insufficient statistical information for image representation and noisy tags which greatly decreases the clustering performance while increases the computational cost. In this paper, we propose a two-step semi-supervised hierarchical clustering algorithm, Personalized Hierarchical Theme-based Clustering (PHTC), for web image organization. In the first step, the Probabilistic Fusion ART (PF-ART) is proposed for grouping semantically similar images and simultaneously learning the probabilistic distribution of tag occurrence for mining the key tags/topics of clusters. In this way, …

Fast Semantic Diffusion For Large-Scale Context-Based Image And Video Annotation, Yu-Gang Jiang, Qi Dai, Jun Wang, Chong-Wah Ngo

Research Collection School Of Computing and Information Systems

Exploring context information for visual recognition has recently received significant research attention. This paper proposes a novel and highly efficient approach, which is named semantic diffusion, to utilize semantic context for large-scale image and video annotation. Starting from the initial annotation of a large number of semantic concepts (categories), obtained by either machine learning or manual tagging, the proposed approach refines the results using a graph diffusion technique, which recovers the consistency and smoothness of the annotations over a semantic graph. Different from the existing graph-based learning methods that model relations among data samples, the semantic graph captures context by …

Flexrender: A Distributed Rendering Architecture For Ray Tracing Huge Scenes On Commodity Hardware., Robert Edward Somers

Master's Theses

As the quest for more realistic computer graphics marches steadily on, the demand for rich and detailed imagery is greater than ever. However, the current "sweet spot" in terms of price, power consumption, and performance is in commodity hardware. If we desire to render scenes with tens or hundreds of millions of polygons as cheaply as possible, we need a way of doing so that maximizes the use of the commodity hardware we already have at our disposal.

Techniques such as normal mapping and level of detail have attempted to address the problem by reducing the amount of geometry in …

A Modular Framework For Home Healthcare Monitoring, Blake Puryear

Computer Science and Computer Engineering Undergraduate Honors Theses

Many patients with chronic health problems have multiple ailments but different patients may have different such ailments. Home monitoring systems for individual ailments exist but a patient may have multiple of these, all designed independently. There are no standard architectures so this leads to unmanageable diversity which causes problems for patients in having to learn to use a variety of monitors and for physicians in trying to monitor many patients. The purpose of this project was to design and prototype a next generation modular remote healthcare monitoring system capable of monitoring multiple ailments and extensible to new ailments in order …

Simple Touch-Based Gaming: Project Pond, Taylor Yust

Computer Science and Computer Engineering Undergraduate Honors Theses

The mainstream success of smartphone technology has prompted the video game industry to invest resources into new types of mobile games. While seemingly successful on the surface, there are unseen issues and risks that may potentially harm the long-term sustainability of the market. In response, I’ve developed my own smartphone gaming app that addresses what I find to be some of the largest problems in the market. My intent is to test my design theories while also challenging the status quo in mobile game design. The fruit of my labor is Pond, a smartphone gaming app for Windows Phone 7. …

Cogtool-Helper: Leveraging Gui Functional Testing Tools To Generate Predictive Human Performance Models, Amanda Swearngin

School of Computing: Dissertations, Theses, and Student Research

Numerous tools and techniques for human performance modeling have been introduced in the field of human-computer interaction. With such tools comes the ability to model legacy applications. Models can be used to compare design ideas to existing applications, or to evaluate products against those of competitors. One such mod- eling tool, CogTool, allows user interface designers and analysts to mock up design ideas, demonstrate tasks, and obtain human performance predictions for those tasks. This is one step towards a simple and complete analysis process, but it still requires a large amount of manual work. Graphical user interface (GUI) testing tools …

Converting Three-Component To Four-Component Image (2012), Ronald S. Cok, Michael E. Miller

AFIT Patents

A method of converting a three-or-more-color-component image input signal to an image output signal includes acquiring an input signal having a plurality of pixel signals, each pixel signal having three, or more, color components; determining a residual difference for each color component of each pixel signal; determining a limit value of the residual differences; calculating a common scale factor for each of the color components based upon the limit value; and applying the common scale factor to the image input signal to produce the image output signal.

Analyzing Financial Data Through Interactive Visualization, Fizi Yadav

Purdue Polytechnic Masters Theses

This investigation explored the role data visualization in the scientific community and its effect on the cognitive ability of an individual. The research attempted to answer the question, “ Whether appropriate data visualization helps in better understanding of corporate financial data?” and used quantitative methods to gain an insight. Participants in the study were divided into two groups of 30 each, with one group receiving the treatment devised for the research and the other using the common prevalent method. Both the groups were subjected to a test, based upon the analysis of which a conclusion was derived. The subjects were …