Relight Lab
Usability and Interface Redesign
This usecase showcases the process for redesigning the user experience for the desktop application.
Relight Lab Desktop Application
RelightLab is an open-source application to create relightable images also known as Reflectance Transformation Imaging (RTI). I worked on this software in collaboration with Dr. Federico Ponchio from ISTI-CNR while my research associated period at University of Cagliari (UniCa).
Challenges 🫥
The target audience for Relight Lab is pretty diverse, spanning from experienced users who professionally capture and process RTI as a service to museum curators with extremely limited knowledge of both RTI and computers in general. The workflow can be pretty complex and follow different paths and the choice of the most appropriate basis depends on the acquisition mode, the material appearance of the subject, the intended use.
Goal 🎯
Introduce a semi-intrusive flow of actions to the user to allow a blend of both reasonable degrees-of-freedom to switch between various tools-and-options, while also proposing a firm flow of actions from start to end. This will allow a wizard-like experience for users to help them stay on track with the process. This approach also saves a lot of time which would otherwise be lost in several attempts to switch back and forth between various steps without a particular order.
Wireframes
What we did (principles applied)
The design of the interface addresses these challenges following a set of well established principles:
Discoverability: the application starts with a screen presenting the user with a choice of the main tasks available, and the tasks are organized in a natural flow of operations.
Spatial organization: the RTI processing is organized in a set of functional tasks, each in a different page. This provides plenty of space while only the relevant parameters and information are presented to the user. The user can easily create a mental map of the tasks.
Ease of navigation: a streamlined task flow must not come in the way of the ability to quickly navigate to a different step and change some parameter. Experienced users will know exactly which parameter to change and not necessarily follow the established sequence. The interface is designed in such a way that every parameter is at max 2 mouse clicks away, and dependencies between parameters live in the same page.
Efficiency: RTI processing can be a pretty repetitive process, we aim to minimize the number of parameters the user needs to specify. The interface strives to remember all of the user's previous parameters and save a predefined set of parameters, in order to easily specify an entire configuration.
Background processing: RTI processing can be computational intensive, the interface is designed to run the processing in background while the user can start working on a new dataset. The user needs to be able to monitor and control the background processing.
Visual feedback: the exported RTI should be immediately available to the user for inspection: we use the browser as the viewer, and the 'casting' metaphor: a local https:// server is started and the browser pointed to the dataset visualization webpage.
Interface Ui
Contributions
My major contributions were focused on understanding the existing flow of the software, gathering current requirements and redesigning the interface to simplify the processes, to make usability enhancements, user-flow corrections and draw/design wireframe for further development references (see section of screenshots). I revised the user interface elements considering the UI recommendations and suggestions from the existing expert users of the software.
Acknowledgements
This work is in collaboration with Dr. Federico Ponchio from ISTI-CNR while my research associated period at University of Cagliari. Part of the software contributions are from ViDiC Group of CRS4.
Reference: Grant 9.2022 / TDM – Tessuto Digitale Metropolitano CUP F23C17000010006