Computer Science Showcase 2021

Zeke Edriss and team (undergraduate students)
Class: CPSC 329 project
Description: In this video we go over the history of security and malware (most destructive viruses, value-wise), in an attempt to show the values of security. For each attack, we went over the specific details of what occurred, including who launched the attacks, how they did it (what vulnerabilities were exploited), what their motivation was, and the specific type of malware used. We also talk about how each company handled the security threat, how much damage was done and how it was resolved. We also made a simple quiz based on the slide deck for viewers to attempt (using google forms).

Braden Foxcroft (undergraduate student)
Class: CPSC 329
Description: I created a game where you play as a virus, trying to outplay the antivirus and achieve several objectives.  The game is made on an html canvas, using JavaScript, built for Google Chrome.  Presented is a video walk-through of it, explaining how the game works and some of the challenges faced when making it. The GitHub link to the project is: github.com/braden-foxcroft/Virus_game

Lucia Fu, Vineet Kour, Lorena Young and Michael Wahba (undergraduate students)
and Owen Brierly (graduate student)
Class:  UNIV 401.32
Description: The immune system is a complex process which can be visualized and modeled through an interactive simulation of agent-based modelling. Our work uses Lifebrush as a tool to allow a dynamic portrait of the interactive particles and cells within the human immune system at a microscopic dimension to be viewed. Our work is primarily be focused on SARS-CoV-2 and the immune biological processes that follow in response to infection which we simulate through LifeBrush. We present a poster and a paper outlining how the immune system works and how Lifebrush simulates immune responses post COVID-19 infection.  View the project's PDF Poster  and PDF Paper.

Ahmet Numan Aral and Phillip Aubrecht (undergraduate students)
Class:  CPSC 405
Description: SlideScott seeks to help educators create interactive slides that measure student performance and engagement, with minimal time and effort. SlideScott uses interactive elements like multiple choice quizzes and auto marking written responses to keep students engaged and focused. In the future SlideScott plans to provide educators with the analytics of student’s performance, usage; utilizing a proprietary algorithm to suggest improvements.  Also visit numanaral.github.io/slide-scott.

Ahmet Numan Aral (undergraduate student)
Individual project
Description: A bomberman game made with pure web (powered by typescript & react) without using a WebGL or any game engine. It's still work in progress. The online-multi-player mode is up there but there is no UI yet for you to get there, I will be showcasing that on the demo day. Keep checking the prod site or the git repo/milestone for more updates at numanaral.github.io/bomba-man.

Sabrina Lakhdhir (undergraduate student)
Class:  CPSC 502.07 honours research project
Description: The near reality of fully autonomous vehicle’s (AVs) production and usage has already been established. AVs driverless nature has led to several previous research projects focused on studying and designing interactions between them and vulnerable road users. Thus far, much of the work has focused on exploring interfaces that communicate the vehicle’s intent and awareness to pedestrians through mounts on vehicles and street infrastructure. An under-explored area is envisioning interfaces that can be placed on pedestrians themselves to provide more personal forms of feedback. We report on a study wherein participants proposed designs for future wearables to aid in AV-pedestrian interactions by providing relevant alerts to wearers, helping them make safer crossing decisions. Informed by our participants’ designs and analyzed against the Design for Wearability Guidelines, we propose and discuss three promising designs for future wearables as interfaces for receiving information from AVs. Additionally, we introduce early-stage prototypes of the three designs, followed by a brief technical evaluation of their functionality.

Michael Wahba (undergraduate student)
Class: CPSC 503
Description: Lifebrush is an open-source virtual reality environment used to model multi-agent systems. I used Lifebrush to create a model demonstrating the spread of a pandemic through a population. To house this simulation I created a virtual reality museum environment in Unreal Engine where users can learn more about the Coronavirus in an interactive environment. The hope is that users can learn more about the virus not through reading, but through interacting.

Colin Au Yeung (undergraduate student)
Class:  CPSC 584
Description: Virtual Youtubers, or VTubers for short, are an upcoming genre of livestreamers who rather than having camera feeds of themselves, they use technologies such as face tracking to livestream through virtual avatars like 3D and Live2D models. However, for these individuals, the characters they take on only exist in the virtual space. While for many livestreaming tasks this is fine, there are still many areas where physicality is important. In this project we propose how robots could be used as avatars for livestreamers. We developed a tracking api for designing interactions with a robot avatar and did high fidelity prototyping using puppets. Finally we did a survey to determine the ability of the robot avatar prototypes to act as avatars.

Jake Kurtz (undergraduate student)
Class: CPSC 591
Description: Many games with large open worlds have dynamic weather and lighting systems to give players a greater sense of immersion. A big part of creating convincing and beautiful worlds is creating convincing and beautiful clouds. In this project, I developed a shader which procedurally generates volumetric clouds using noise functions and ray marching techniques.

Zahra Ghavasieh and Shamim Khalili (undergraduate students)
Class:  CPSC 599
Description:  This video describes a demo of the prototype of the Augmented Book, an augmented reality browser application that combines multiple aspects of electronic and physical books to create a greater reading experience. While many other projects have explored the projection of enhancements onto physical books or added image augmentations in picture books, few have reattempted these concepts in recent years using WebAR technology. Inspired by these previous works, the Augmented Book attempts to provide an alternative and more accessible interface using modern means. Using this initial concept, our team has constructed a small prototype demonstrating some of the more basic and readily available features of the Augmented Book such as book recognition, bookmarking, and the audiobook player, among others.  Try out the prototype yourself at gnvqeewj.8thwall.app/augmented-book.

Tian Xia and Kai Huang (undergraduate students)
Class:  CPSC 599 project
Description: When people visit new places and need information about such places, they must search on multiple platforms to obtain business information, Google ratings, GPS navigation, etc. TourAR addresses this issue by displaying multiple sources of information in one application. TourAR is an augmented reality application built with WebXR frameworks to improve user experience in tourism. It provides users with features: 1) displaying target locations’ names and Google ratings as labels floating or attached to the target locations based on their geolocations; 2) highlighting target locations by dimming the background and displaying the target locations with green highlight or normal lighting; 3) summoning a virtual tour guide to commentate on the target locations; 4) navigating to a target location. Furthermore, we conducted user studies to improve the design and user experience of TourAR.

Alexander (Sasha) Ivanov (graduate student)
Thesis project
Description: We present EvoIsland, a scalable interactive evolutionary user interface framework inspired by the spatially isolated land masses seen on Earth. Our generalizable interaction system encourages creators to spatially explore a wide range of design possibilities through the combination, separation, and rearrangement of hexagonal tiles on a grid. As these tiles are grouped into island- like clusters, localized populations of designs form through an underlying evolutionary system. The interactions that take place within EvoIsland provide content creators with new ways to shape evolutionary populations suited for displaying, assessing, and manipulating systems that produce a wide range of solutions with visual phenotype outputs.

Dante Bencivenga and Xining Chen (graduate students)
Thesis project
Description:  Quantum walks are quantum algorithms that can find marked vertices on a graph quadratically faster than random walks. By controlling the operation of a quantum walk using an additional qubit, we have developed a new algorithm to solve the related yet distinct task of sampling from a probability distribution over vertices. Efficient sampling is an important computational task used in applications like Monte Carlo simulation. Our controlled quantum walk algorithm can sample quadratically faster than the optimal classical algorithm which uses random walks.

Preston Haffey and Sepideh Avizheh (graduate students)
Research project
Description:  As more economic activity moves online, it is important that mutually distrusting parties are able to conduct a fair exchange of goods over the internet. We examine a protocol called FairSwap that enables fair exchange between two parties using a smart contract run on a public blockchain. We identity information leakage that exposes sensitive information about the exchange to the public. We also present the problem and discuss some solutions that help preserve privacy for FairSwap.  View PDF Poster.

Christopher Smith (graduate student)
Class: CPSC 601
Description: This project called “Lively Words” aims to help those learning a language to increase learning retention by providing a more engaging and memorable experience. This is done by utilizing augmented reality to enhance a word with characteristics or actions associated with the word’s meaning while retaining the environmental context of the word. The goal of this is for the user to create a direct association between a word and its meaning instead of first converting the word to a known context or language. To demonstrate this idea, a total of four prototype were designed for the words: “water”, “dog”, and “grow”. Through these designs, an early design space was created and resulted in each word’s augmentation falling into the categories: invocative, transformative, active, and interactive. Although still at an early stage, this project shows promise as a new way of enhancing language learning with many possibilities for the future.  View all of my project videos.

Mohammad Navid Masahati and Sarah Shah (graduate students)
Class:  SENG 607 project
Description:  Before the COVID-19 pandemic, the City of Calgary municipal organization faced financial pressures due to the aftermath of the 2015 oil price collapse. The COVID-19 pandemic has further amplified the challenge and urgency to identify citizens' needs correctly and promptly to channelize funds in enhancing services efficiently. Moreover, traditional feedback tools like surveys, hotlines, and committee meetings aren't sufficient for decision-making anymore.  On the other hand, social media and specifically Twitter is a go-to solution when people choose to express their daily life challenges and demands. Thus, it would be a very good resource to dig deep and extract those requirements that are interrupting the normal life a citizen deserves to have. These requirements could help the decision-makers of the city to make data-oriented decisions and better target their investments in city services to improve the living standards of citizens to its utmost.  Here in this research, we have developed a requirement engineering and analysis pipeline using state-of-the-art Natural Language Processing (NLP) techniques to extract the most demanding needs of people from millions of tweets. These are the vital and hidden requirements, that if not investigated, could not have been understood by the city. The results will further be accessible by the proper authorities to help make the City of Calgary a more pleasant place to live in, even during these challenging times in the pandemic.  Moreover, the requirement engineering tool is not bound to the City of Calgary and the requirements regarding the pandemic, but it also can easily be applied on any other context and be powerful support of the smart decisions in any other industry.

Albert Leung (graduate student)
Class:  DATA 607 / 608
Description: Each year thousands of people are murdered across the United States and many cases remain unsolved. Violent crimes are a danger to everyone in society as they threaten the trust and security of every individual. Though violent crimes on a per capita basis are in decline, the percentage of unsolved murder cases is increasing. This can be attributed to several factors as criminals become savvier in how they commit crimes and information about obscuring evidence can be easily found online. Besides the clearance rate of identifying offenders decreasing, evidence shows that policing agencies also carry bias in terms of how the cases are handled. This investigation aims to identify patterns (like victimology, geography, and crime characteristics to name a few) and use them to review these crimes with the ultimate goals of holding law enforcement agencies accountable and resolving unsolved cases. Using the data from the Murder Accountability Project, classical machine learning models were created as well as deep learning models to try to predict whether a murder case will be solved or unsolved and to predict the offender profile so that the probability of an offender is convicted would increase. Both techniques created models that gave > 90% accuracy. Since deep learning is a greedy algorithm and it takes time and effort to build a robust model, one should consider using classical machine learning model as a starting point.

Scout Windsor (graduate student)
Class:  ARCH 680
Description:  Boreal Moss Studio is a science fiction-inspired, modern design set in the mountains of Alberta, Canada. I created this project as a final assignment for Dr. Alicia Nahmad Vazquez's class, ARCH 680 - LEARNING MACHINES {THAT LEARN}. The model features building, furniture, and surface textures that I created using images generated by a machine learning model. I trained the model on 504 photographs captured in West Confederation Park, Calgary, Alberta. I turned the generated images into 3D tileable textures using Adobe's Substance Alchemist, a software package that also uses machine learning for texture generation. While creating the Boreal Moss studio's design, I was inspired by architecture from Science Fiction, such as the Death Star Bridge and Naboo Lake Country from Star Wars, and Starfleet Headquarters from the Star Trek TV series. Real-life architecture also influenced the design, such as the Lucas Museum of Narrative Art and Rudolph Steiner's Goetheanum. The Lucas Museum of Narrative Art combines a modern sci-fi aesthetic with natural lighting and integrated natural areas. Rudolph Steiner's Goetheanum required shipbuilding techniques to create its beautiful exterior, but a fire sadly destroyed it in 1922. The nine ML-generated textures were versatile, and I applied them to kitchen counters, drawers, rugs, crossbeams, tables, cushions, bookshelves, and more. I also created a material display that hangs on the wall, which was inspired by the domed paint displays at car shows. I rendered the final studio design and materials using Twinmotion, an architectural rending engine that runs on the Unreal Engine architecture. In this video, I present my creation process, work-in-progress images, and final renders. I hope you enjoy the Boreal Moss Studio as much as I enjoyed creating it!

Ryo Suzuki (faculty member)
Description: We present RealitySketch, an augmented reality interface for sketching interactive graphics and visualizations. In recent years, an increasing number of AR sketching tools enable users to draw and embed sketches in the real world. However, with the current tools, sketched contents are inherently static, floating in mid-air without responding to the real world. This video introduces a new way to embed dynamic and responsive graphics in the real world. In RealitySketch, the user draws graphical elements on a mobile AR screen and binds them with physical objects in real-time and improvisational ways, so that the sketched elements dynamically move with the corresponding physical motion. The user can also quickly visualize and analyze real-world phenomena through responsive graph plots or interactive visualizations. This video contributes to a set of interaction techniques that enable capturing, parameterizing, and visualizing real-world motion without pre-defined programs and configurations. Finally, we demonstrate our tool with several application scenarios, including physics education, sports training, and in-situ tangible interfaces.

Christian Jacob (faculty member)
Description: This video highlights a small number of summer research projects and graduate theses that members of the LINDSAY Virtual Human lab have been working on over the last few years.