Master’s Thesis

Posted on February 4, 2019 by 5lcht

Master's Thesis

Master's Thesis [Work-In-Progress]

Analyzing the social VR landscape to better understand how to make co-located VR experiences more social.

My thesis is an exploration into designing interactions and hardware for collocated collaborative Virtual Reality experiences for the purpose of creating design guidelines that support better communication and learning between collaborators.

My motivation behind this work stemmed from the lack of commercial device support for collocated VR experiences. For example, I have experienced this issue personally while watching VR Netflix using my phone and a Daydream headset. If there is someone else in the room with you while you are watching TV in a VR headset, it is difficult to pay attention to or acknowledge the other person. We imagine that it will be very difficult for the average consumer to adapt to watching long-format media in VR until there is a way to comfortably share this experience with family and friends.

For this project, two bachelor’s students are assisting with the work. Together, we are developing hardware prototypes (in cardboard), interaction designs (in sketch), and designing scenarios.

Survey of Social Applications

Method, Themes, Activities [TBD]

Literature Review

Method, Time-space Matrix, Themes, Measurements, Opportunities [TBD]

Prototypes

We will use these demos as a part of a user study to discover if body representation in co-located social VR experiences encourages collaboration and increases feelings of togetherness. Scenarios will be developed to study multiple co-located VR participants and co-located VR + non-VR participants.

Low-cost body tracking for virtual reality

This purpose of this demo is to provide a way for mobile VR users to engage in VR experiences together simply by placing their phones in a low-cost VR device like Google cardboard. By using phone camera-based body tracking, VR users do not have to instrument an environment to get realistic body representation in the virtual environment.

Additional work to be done

  • Use additional sensor or marker to identify z coordinate.
  • Animate a 3D model with poses.
  • Add facial expression tracking of both VR and non-VR participants.
  • Provide multi-user support.
  • Prototype will be configurable to support multiple study activities.

github

live Demo

Low-cost face tracking for virtual reality

This demo instruments the user’s computer or phone camera to track lower facial movements of co-located participants while both users are wearing a head-mounted VR device in order to visualize participant’s facial expressions in VR.

Additional work to be done

  • Animate a 3D model with face.
  • Prototype will be configurable to support multiple study activities.

github

live Demo

Kinect body tracking for virtual reality

This demo instruments the user’s environment to provide robust multi-user body tracking in the virtual environment.

Additional work to be done

  • Animate a 3D model with poses.
  • Add facial expression tracking of both VR and non-VR participants.
  • Provide multi-user support.
  • Prototype will be configurable to support multiple study activities.

github

live Demo

Study

Tests & Methods [TBD]

Results [TBD]

Reflection

Impact of research [TBD]

Credits

Research conducted with WeAreVR group from the University of Michigan Information Interaction Lab and thesis advisior, Professor Michael Nebeling.

WeAreVR Group Members: Katy Madier, Rhea Kulkarni, Sindhu Giri, Sophie Linn


Posted under: Uncategorized

Mission CTRL

Posted on by 5lcht

Mission CTRL

Mission CTRL

A tool to help space tourists find resources, fun, and friends.

The Team: Katy Madier, Alexandra Carey, Jacob Berman

My Role: design, three js programming

This project is a concept platform to help space travellers find resources and connect with other travelers while adventuring through space.

Wireframes

In this project we started with identifying key tasks and designing master/detail pages for each important section of the app. We utilized a wireframe template for this app called moon, to set up our initial information layout.

missionctrl   wireframes
#walter mobile

Prototype

My team and I built this tool with Ionic (Angular) and a firebase database. Since this app utilizes 3-dimensional space for navigation and planet discovery, I lead the development of the VR features with Three js.

View the project Github

Reflection

This project was fun and challenging. The implementation of Three js within an ionic app was not entirely smooth, and I found that there were compromises we had to make to meet the course deadline. We also did not get to bring the app design as far as we had originally planned. This was a fun exercise in imagining future consumer navigation applications.


Posted under: Uncategorized

Experiments & Personal Projects

Posted on August 12, 2018 by 5lcht

Experiments

Experiments & Personal Projects

Explorations into AR/VR design, programming, and mobile interactions

These are a combination of personal projects, design sprints, programming practice, and project demos.

Mobile

camera orientation filter

Processing camera input and motion sensing

This demo adds an image filter to the phone camera stream when the orientaton of the screen has been changed. This demo requires access to a mobile device camera.

View the project Github

Live Demo

Filtering pixels from a video feed

This demo selects a pixel color from a video feed and then changes or removes that color from the video. This can be used for chroma keying video.

View the project Github

Live Demo

Augmented Reality

A-Frame XR – placing OBJ model in a scene

This demo adds allows a user to add a bunny object to an AR scene when tapping on the phone screen. This demo requires access to a mobile device camera and a WebXR browser.

View the project Github

Live Demo

AR.js with OBJ model

This demo adds a bunny object to an AR scene when viewing an AR marker. This demo requires access to a device camera and fiducial markers.

View the project Github

Live Demo

A-Frame XR & AR.js with positional sound

This demo adds a boombox and music to the scene when viewing an AR marker. The sound is emitted from the object, and tapers off as the phone moves away from the object. This demo requires access to a device camera and fiducial markers.

View the project Github

Live Demo

Virtual Reality

A-Frame Poptarts

A-Frame & Sketchup Poptarts

This demo was made to play with gifs and a 3D modeled poptart I made in Sketchup.

View the project Github

Live Demo

A-Frame with motion & sound

A-Frame with motion and sound

This demo was practice using sound, animation, and a VR controller.

View the project Github

Live Demo

Bunny Toss

A-Frame with physics

This demo uses the A-Frame physics library and daydream controls to throw a bunny.

View the project Github

Live Demo


Posted under: Uncategorized

Prototyping for Augmented and Virtual Reality

Posted on August 5, 2018 by 5lcht

Master's Thesis

360proto

Prototyping for Augmented and Virtual Reality with Paper

Research conducted at the University of Michigan Information Interaction Lab by Professor Michael Nebeling and Katy Madier.

This paper has been conditionally accepted at CHI 2019. Details coming soon.


Posted under: Uncategorized

AR Kitchen Design Studio

Posted on by 5lcht

arkitchen

ARKitchen

An Augmented Reality Kitchen Cabinet Design Studio

The Team: Katy Madier, Sonali Tandon, Michael Nebeling

A well-known kitchen manufacturer has commissioned the lab to create an immersive showroom experience for their dealership locations. The client’s main focus is to expand on the physical experience of selecting design through product samples to include full-size previews of style choices for the goal of providing more immersive visualizations of the installed finished product.

Our prototypes specifically attempt to aid customers during the style and finish selection process that generally happens during the first and second visit to the dealership. Instead of a customer relying solely on physical color samples to make finish decisions, physical samples are complemented by an augmented reality display. This display allows the customer to view selections in a realistic way, providing better context and helping the customer feel more confident in their decision.

My role in this project included paper and digital prototyping, conducting stakeholder and dealer demos, and author for a project specification document.

Below is an edited recap of our research, with findings and recommendations removed as a courtesy to our stakeholders.

Sketches & Ideation

Guided by Professor Michael Nebeling, Sonali Tanden and I were tasked with researching and designing prototypes. We created four different prototypes of the augmented reality display. Each projected prototype can be controlled by a tablet interface, prototype #0..

Projection Tests

Tablet Controller Mockups

Prototypes

Prototype #1: Tablet Interface

  • The interface allows the user to select a style, finish, and hardware.
  • Can save favorite selections in a cart style menu.
  • Can share favorite selections via email or QR code.
  • Can compare two favorite selections at a time with a side-by-side view or upper and lower cabinets view.

Prototype #2: Mobile AR App

  • AR application for phones or tablets that allows a user to select kitchen style, finish, and hardware to then view the selections in the AR environment.
  • The app utilizes the same tablet interface assets, web-based code, and interface design of the tablet controller.
  • The user can walk around the AR image to see a 360-degree view of their selections.

Prototype #3: Full-scale Kitchen Projection

  • The projection shows customer style, finish, and hardware selections onto real full-sized cabinet doors in the Design Studio.
  • The projection system can be customized to accommodate design studios with different cabinet configurations.

Prototype #4: Mini-Kitchen Projection

  • Prototype projects customer style, finish and hardware selections from the tablet onto a small kitchen replica on the Design Studio table-top.

Demos & Interviews

To test our prototypes, we conducted three demos for stakeholders, from leadership to showroom designers and salespeople. With each demo, we collected feedback and improved our interface design and project concepts by incorporating the insights.

Reflection

Impact of research


Posted under: Uncategorized

Together Chicago Resource Mapping

Posted on June 15, 2018 by 5lcht

Together Chicago

Together Chicago

An Augmented Reality Kitchen Cabinet Design Studio

The Team: Yazmon Ector, Benjamin Heath, Karan Kwatra, Ryan Ma, Katy Madier, Rachael Park, Mohammed Zebdi

Together Chicago s a non-profit focused on improving economic development, providing access to education, reducing violence, and mobilizing faith communities within the city of Chicago.

Our team has created an interactive web-based mapping application that displays open-sourced data of the City of Chicago in a searchable and filterable format. Data can also be analyzed and exported from the platform.

This project helps individuals and non-profits make better data-driven decisions that will provide a positive impact on the Chicago community.

System Goals

  • Allow users to easily search and compile demographic data from a query of a specific geographic area.
  • Encourage greater collaboration among nonprofits, city, and service organizations in the Chicago area.
  • Enable organizations to make better decisions regarding how best to deploy resources and services.

My roles in this project are Interaction Design, User Experience Research, and Front-End Development. This project ran from January 2018 – December 2019 as a part of the Multidisciplinary Design Program at the University of Michigan School of Engineering.

Sketches & Ideation

Use Cases, Personas, User Flows

view use cases

Use Cases 1
Use Cases 2
Use Cases 3

Paper Prototypes & Initial Code

TC Paper Prototype
TC High Fidelity 2

Prototypes

TC Medium Fidelity

Medium Fidelity

Live Version

TC high fidelity

High Fidelity

Prototype built with React JS, Django, and Database. My role for this prototype was in front-end styling and React development.

Validation Methoodology

Interviews

Design and feature decisions will be defined based on stakeholder and user interviews.

  • 2 stakeholders, who are also system users
  • 15 minutes
  • Audio recording & notes

Findings

  • Users do not want to be shown curated data correlations.
  • Success is encouraging users to embrace unexpected conclusions.
  • Biggest perceived user issues: navigation and intimidating datasets.

User Evlauations

Usability Testing with a paper prototype

  • 5 users
  • 10-15 minutes
  • 3 tasks
  • Recordings of moderate & critical errors
  • User feedback session

Findings

  • Most critical errors happened in the search filter sidebar.
  • Testers showed confusion with default information and search capabilities.
  • Default map view should include the entire city of Chicago and the search bar should be obvious.

User Survey

Survey to gather data analysis and technical skills and habits of potential system users.

  • 49 responses
  • Non-profit leaders and CEOs

Findings

  • Technically saavy
  • Want to use data to solve business problems
  • Need better resources in their daily business process

Posted under: Uncategorized

Needs Assessment and Usability Evaluation for The Google Home Mini

Posted on April 15, 2018 by 5lcht

Google Home Mini

Usability Evaluation of the Google Home Mini

The Team: Sonorous: Tori Cox, Bonnie Lee, Katy Madier, Ruchi Ookalkar, Yu Qin

Team Sonorous evaluated the usability of the Google Home Mini for University of Michigan college students. The team evaluated the product during the spring 2018 course of SI622 with Professor Joyojeet Pal. Project progress was presented to a UX research team at Google monthly. Partial research results and recommendations are available below.

Research Questions

  • How do college students consume media on different devices?
  • What are the student’s expectations and perceptions of the Google Home Mini?
  • How do students prefer to discover new features and what features could assist students?
  • Is the Google Home Mini an effective tool for students?

Needs Assessments

Google Home Mini Interaction Map

Interaction Map

System evaluation by creating a map of possible user interaction paths.

Google Home Mini Interaction Map

Interviews

  • Recruiting: Identify students that fit population criteria, high proximity bias.
  • Interview Process: 20-45 minutes, 1 note taker, 1 audio recording
  • Demographics: Ages 20-29, 5 different majors, 4 masters & 1 bachelor
  • Data Analysis: Affinity Diagram, personas, and spreadsheet pivot tables
Comparative Evaluation

Comparative Evaluation

  • Evaluated 5 different categories of comparative products.
  • 10 competitors identified based on similarities in functionality, size, price, and usability.
  • Measured 10 metrics defined by data from user interviews and feedback from project stakeholders.
  • Scored products based on online reviews and product specs.
Google Home Mini Interaction Map

Survey

  • Delivery: University Email Lists & Facebook Groups (~ 14,000 people).
  • Requirements: UM Student
  • Respondents: 100
  • Age: 18-30
  • Degree: 43% Undergraduate / 46% Graduate

Usability Evaluation

Heuristic Evaluation

Evaluation using Nielsen’s Heuristics:

  • Feedback: What happens when it doesn’t understand the user?
  • Accessibility: Can it understand user’s accent?
  • Navigation: User control and freedom
  • Consistency: Consistency in response personality
  • Prevention: Error prevention
  • Memory: Support follow-up queries?
  • Efficiency: Mini v.s Phone/Laptop

Usability Tests

  • Target Population: University of Michigan Student Google Home Mini Users
  • Recruiting: Email Questionnaire + $10 gift card
  • Demographics: Age 23-29, Graduate Students, School of Information & School of Pharmacy

Test Setup

  • Introduction
  • Pre-questionnaire
  • 4 Tasks
  • Debriefing
  • Post-questionnaire

Recommendations

Support University Wifi connections and Enterprise Google Accounts

Students are unable to connect the Google Home Mini through the University wifi accounts. Additionally, Enterprise Google Accounts do not have all the available features, like adding a shopping item or an event.

[Interviews][Survey][Heuristic Evaluation][Usability Test]

Create a way for users to send friends feature recommendations

Current Mini users rely on recommendations from friends to discover new features.

[Interview][Survey]

Support Social Communication

Expand feature options on the mini to include more social communication.

[Interviews][Survey]

Create a Privacy Mode

Provide a way for the account holder to turn a privacy mode on and off or give the account holder an interaction log or request history for their account.

[Heuristic Evaluation][Usability Tests]

Reflection

Impact of research


Posted under: Uncategorized

#walter

Posted on December 15, 2017 by 5lcht

Master's Thesis

#walter

A concept browser extension and debate platform that brings fact checking and verified experts to twitter conversations.

The Team: Katy Madier, Caitlin Endyke, Jacob Berman, Innocent Ndubuisi-Obi

My Role: research, poster design, video and prototype creation

A Pew Research Center Survey suggests that 62% of Americans get their news from social media. At the same time, social media platforms run algorithms that recommend and show only similar and related information in a user's feed. What does that mean for this country’s political discourse? Are the conversations we’re having online helping us better understand the issues, or are they confirming already held viewpoints?

This project explores how an innovative annotation tool can elevate the debates Americans are having about political news on social media. #walter is a news annotation tool that we hope will take users out of their social media echo-chambers. Its horizontal presentation of user comments, discussion from vetted experts, and on-demand fact-checking will help users read news articles more critically and have more diverse, productive conversations about America’s most pressing topics.

Full Research Paper

#walter desktop

Desktop Prototype

#walter mobile

Mobile Prototype

#walter poster

Poster

Video Submission

Survey

We surveyed 50 social media users to better understand how they evaluate expert qualification and trust in fact-checking mechanisms.

  • 64% of our survey participants said they would like fact-checked information in articles they read online, and 84% of respondents said that they trust the information that has been fact-checked.
  • 60% of people surveyed said they thought "many years of experience" would qualify someone as an expert.

Reflection

Impact of research


Posted under: Uncategorized

The Kid Cooperative

Posted on by 5lcht

The Kid Cooperative

Kid Cooperative

A management and messaging system to assist parent cooperatives in scheduling childcare.

When my daughter and I moved to Ann Arbor in the fall of 2017, we found that the afterschool childcare program at her elementary school was full. Even the alternative childcare options in the area were completely booked. The local YMCA even had a 2-year waitlist. We do not have the budget for daily babysitters, so we decided to try something different. After brainstorming on this idea in SI582, Interaction Design, I came up with a solution.

Having been involved in a co-op before, I thought one way to solve our problem would be to create a childcare co-op with other parents in the area. After looking into this possibility, I realized that there are not many tools available for managing all of the information that co-ops need to share. During the SI588 course, I was able to create a prototype of a platform that would solve my problem. The Kid Cooperative is a co-op management tool that would support parenting groups by managing communication, point systems, and a shared calendar.

Competitive Analysis

A large part of my design process relies heavily on research, fact-checking assumptions, and diagramming processes. I developed a deeper understanding of what may be required by a co-op from background research and competitive analysis. With this information, I was able to create tools and functions to address the specific user’s needs. During this phase, my design changed in concept from being a platform for one Ann Arbor-based co-op to a co-op management platform. I could not justify the need for a location-specific site and instead thought that a site for anyone, anywhere made more sense.

Sketches & Ideation

From initial sketches of potential solutions to sketching user stories, I tested a wide variety of diverse concepts. The brainstorming process provided me with many opportunities to refine my original co-op idea as well as try out less practical solutions. Through testing many different perspectives on the problem, I became better at understanding the idea as a whole.

Brainstorming

Kid Cooperative - Sketching Alternatives 1
Kid Cooperative - Sketching Alternatives 2
Kid Cooperative - Sketching Alternatives 3

Personas

Personas 1
Personas 2
Personas 3

Storymaps

Storymap 1
Storymap 2
Storymap 3

Prototypes

Kid Coop paper prototype

Paper Prototypes

kid coop wireframes

Wireframes

Kid Cooperative Highfidelity

High-Fidelity Prototype

live Demo

Usability Testing

These prototypes were tested during in-class usability tests with 3 Interaction Design students. The results of these tests were used to iterate on the design of the prototype interface.

Reflection

Impact of research

TBD

Credits

This was a individual project created in the University of Michigan's School of Information Interaction Design course.


Posted under: Uncategorized