XR CALIBRATION
DISGUISE
Overview

XR, an advanced workflow found within Disguise Designer, enables its users to create a virtual stage or set extension with the help of LED video and external hardware. During the beta phase of the XR calibration workflow's development, usability was drastically improved by its redesign. This case study delves into the project's success.

The XR studio, powered by Designer, features a 3D previsualization of the stage behind consisting of cameras, tracking system, and LED wall in the background

Role

As a UX designer, I partnered with the development team and senior stakeholders to provide user research, on-site design input, workflow documentation and managed user feedback during the rapid re-design of an extended reality calibration workflow. The project, which was completed over five months during the 2021 UK lockdown, took place in Disguise's XR Studio in London.

Problem

During XR calibration, widgets in the Designer UI clutter the 3D previsualization in the background.


During the beta phase of the product development, feedback gathered from a slack channel of users revealed that they struggled with the existing XR workflow due to several issues, such as a lack of documentation, a cluttered and unhelpful user interface, and a lack of critical feedback. These problems made it difficult for users to set up the system correctly, understand what was happening during the calibration process, and diagnose and fix any issues that arose. As a result, the XR workflow was slow and unreliable, causing delays and errors in the pre-production phase of a show.

Research

As XR was a new, technical, and niche product, during the learning phase we formed a team including developers, QA engineers, and myself to share knowledge of the workflow and become in-house experts on a purpose-built stage at Disguise HQ. This helped us overcome the lack of internal knowledge and expertise on the product. We spent several weeks learning the workflow ourselves and trying to identify the specific challenges and pain points that users were experiencing. Based on our initial observations and experiences, we conducted user interviews to gather more detailed and comprehensive feedback from users about their experiences. We then analyzed the feedback in a research repository and compiled the insights to inform the design process.

At Disguise HQ's XR Studio, I collaborated with the XR team to better understand their workflow. In the foreground, operators used Designer, while a XR test pattern illuminated the LED wall in the background to aid in the calibration process.

Findings
A snippet taken from the research repository, the image shows the user raw data, the inisght gleamed from it and the tagging categories used, to bucket the insight by area of ui it exists within and the sentiment.

The results were an exhaustive list of potential feature improvements, but some key issues stood out. For example, the lack of documentation made it difficult for users to set up the system correctly and understand what was happening during the calibration process. One user commented, "the plot calibration error looks fascinating but I don't know what it does." Without clear instructions, users had to rely on trial and error to figure out how to use the system, which was time-consuming and frustrating. This added a significant burden on the support team, who had to spend more time diagnosing users' problems. Additionally, the user interface was cluttered and unhelpful, making it hard for users to find the information they needed and navigate the workflow. Another user said, "more feedback, it's taken me so long to figure out how to learn the observation debugger." This added to users' frustration and made it difficult for them to use the system effectively. Before the redesign, the workflow relied on a disjointed collection of widgets with little logical structure, which added to the mental load in an already stressful environment. One user commented, "Everything felt a bit all over the place, every window is so spread out but they have to work together"

UNEXPECTED LEARNING - API INTERFACE

One unexpected insight we learned was the need for a portable interface on the studio floor for taking camera observations. Currently, this requires one person to operate the camera and another to click the observation take button on the software, which is often located elsewhere in the building. As one user noted, "I could walk around with an iPad and trigger all this and zooms from where I am on stage. Right now we need multiple people for this, which is more expensive and slower." This can be cumbersome and inefficient, and having a portable interface on the studio floor would make it easier and more efficient to take camera observations.

Disguise HQ's XR Studio, we can see the cameras and designer operater physically apart, when we think about enviroments like theatres or tv studios this distance is amplified.

Based on this insight, I designed a companion interface in Figma for tablets that could be used on the studio floor to take camera observations. This interface was integrated into the existing workflow, allowing users to take camera observations more easily and efficiently. It is currently in the product roadmap waiting for development. Once it is implemented, it will provide a more streamlined and efficient way for users to take camera observations on the studio floor.

XR camera calibration interface designed to allow the take of camera observations on the studio floor; reducing the amount of operators needed.
Results

To address the challenges and pain points identified during the beta phase, we implemented several changes to the XR workflow. Firstly, we created clear and comprehensive documentation in the form of a checklist, on how to perform the calibration stage, which included step-by-step instructions and troubleshooting tips. This made it easier for users to set up the system correctly and understand what was happening during the calibration process.


Secondly, we redesigned the user interface to make it more intuitive and user-friendly, with a clear and logical structure that helped users find the information they needed and navigate the workflow (see diagram n). This reduced the mental load on users and made it easier for them to use the system effectively. Finally, we added critical feedback during the calibration process, which provided real-time information on the status of the system and helped users diagnose and fix any issues that arose. These improvements made the XR workflow faster and more reliable, reducing delays and errors in the pre-production phase of a show.

We improved designer camera calibration widget by consolidating multiple other widgets into it, thus reducing the amount of widgets on screen. Additionally, we had added multiple smaller user experience features such as visual feedback on camera observations, blob grid preview, zoom lock and focus amongst others.

Lessons

During this project, I learned several valuable lessons. Firstly, I learned how to deliver a project with ambiguous requirements at a crucial time and tight deadline for a company that was relying on XR as a key business pivot. I had to work closely with stakeholders to understand their hopes and fears for the project and adapt my approach to meet their needs and expectations. Through regular communication and collaboration, I was able to build trust and rapport with stakeholders, and deliver a successful project that met their needs and expectations.


Secondly, I learned how to form a research repository and leverage it to generate user insights that can inform the design of a project. I conducted user interviews and analysed the feedback to identify the challenges and pain points that users were experiencing. This helped me understand the specific needs and requirements of users, and enabled me to design a more user-friendly and efficient XR workflow.


Thirdly, I learned that not every user experience outcome is UI-focused, and that in some cases, documentation can be a key deliverable. For this project, I identified gaps in multiple areas of the business that were in need of one source of truth, and delivered a comprehensive workflow checklist that provided clear and concise instructions for users. This helped improve the overall user experience and ensured that users had the information they needed to use the XR workflow effectively.


Finally, I learned that some projects don't call for traditional wireframes and prototypes phases. In this case, they were counter-intuitive because the UI was based on C++ and not backed by a design system, making it difficult to change. Instead, I adopted a more hands-on approach, becoming the user myself and working with developers on site to hard-code improvements on the fly. This approach proved to be more successful, allowing us to iterate quickly and deliver a more user-friendly and efficient XR workflow. Overall, this project taught me valuable skills and lessons that I can apply to future projects.

Metrics & Measurement

When disguise XR was released, it quickly became a leader in the extended reality space for live experiences, it has been used on countless AAA projects such as Disney’s Mandalorian & MTV Katy Perry Live an many more. It's difficult to find metrics on the system's success because it doesn't have any analytical output, but feedback from users on the Designer r18 release, which contained the XR workflow, was overwhelmingly positive.

“some never believed in XR and suddenly its main stream"


"That’s a dope feature for sure!"


"I think we should throw disguise a party to say thank you."