February 22, 2024
This is some text inside of a div block.

Case Study: Developing AI Solutions for Visual Merchandising

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse varius enim in eros elementum tristique. Duis cursus, mi quis viverra ornare, eros dolor interdum nulla, ut commodo diam libero vitae erat. Aenean faucibus nibh et justo cursus id rutrum lorem imperdiet. Nunc ut sem vitae risus tristique posuere.This is some text inside of a div block.

Giving second-hand clothing a new life with artificial intelligence visual merchandising.

The team at the Hartree Centre North East Hub worked alongside Throneware on their Regrant project to help to explore of Virtual Try-On systems to develop an artificial intelligence visual merchandising solution to give pre-loved garments a second life.

The Challenge

The team at Throneware were looking for support with their Regrant project. They wanted to find an artificial intelligence (AI) or digital solution to help the pre-loved clothing industry to compete with the fast fashion, first-hand clothing industry. Chris and Rebecca from Regrant were successful in our first round of open calls and granted twelve weeks of support with our data scientists. The project centred around the use of AI to develop a solution for visual merchandising, to make selling second-hand clothing easier, with the solution being to take basic images of garments from sellers, and put them onto a model in a stylised image that makes shopping second-hand feel like shopping new

The Support

The team at the Hartree Centre North East Hub worked with Regrant project members to explore the burgeoning world of Virtual Try-On (VTON or VITON) systems. Initial stages of the project involved examining the requirements and workflow already set up by Regrant, using the InvokeAI framework and critically appraising this system, before moving on to look at open-source VTON offerings developed within the past few years by the academic research community. This activity revealed a wealth of models, typically reliant on deep-learning approaches, but also identified issues with the availability of datasets used to train such models. Work then followed a two-fold path: One strand looking at the behaviour and relative performance of pre-trained models, and the other looking at tools to help Regrant train their own deep learning models, based on established techniques, and new image data sets.

Benefits of the Support

The project has lead to a deeper understanding of the complexities of applying deep learning and generative AI to the exciting and growing domain of Virtual Try-On. Over the course of the project the feasibility of Virtual Try-On approaches have been explored, allowing Regrant a better understanding of the benefits and limitations of current techniques, that will assist in the development and growth of their current and future product offerings.

“Working with the Hartree Centre North East Hub allowed us to evaluate our current solution and their expertise helped us to plan the best way forward. The team then tested these approaches and gave us actionable steps to train and improve our model in future. We would thoroughly recommend working with the experts from the Hartree Centre North East Hub to any organisation looking to use innovative technologies to solve a problem."

- Chris Grant, Throneware / Regrant

Further Information

This work was completed as part of one of our collaborative data projects. The projects are up to 12 weeks in duration and give you access to a wide range of expertise across our team of data scientists and data engineers. We will work alongside your team to scope your data science or engineering project, build a prototype solution, and explore options to deploy it within your organisation. You can learn more about them on our webpage here.

If you would like to learn more about the Hartree Centre North East Hub or our collaborative data projects, please get in touch with us at: