Threls x face:LIFT

Deliverables

Business Analysis, UX/UI Design, Backend, Web App, Mobile App

Tech Used

Flutter, Laravel

Service

Develop your Software Product,

Project Year

2021 - 2024

Website

facelift.research.um.edu.mt

The University of Malta approached us in 2021 to collaborate on the faceLIFT project funded by the Malta Council for Science and Technology (MCST). This project explores innovative applications of artificial intelligence in multimodal information processing. Its primary goal is to create advanced text-to-image technology that generates facial images based on linguistic descriptions.

The Challenge

The main challenge was to create a reliable, scalable and user-friendly application that could integrate with their proprietary AI model, allowing users to generate facial images from text inputs across mobile and web platforms. FaceLIFT sought to have this complex technology aimed solely for entertainment while also paving the way for broader, long-term applications in fields like forensic science and education. The vision was to revolutionise AI interactions through an intuitive, cross-platform app that would set a foundation for both immediate engagement and future expansion.

Our Solution

Design & User Experience: We began by collaborating closely with FaceLIFT to ensure the app's design and user interface aligned with their vision. This phase involved creating wireframes and prototypes that would provide a seamless experience for users, making it simple to input text and receive generated facial images in real time.

Cross-Platform Development (Flutter) Given the need for both mobile (iOS and Android) and web applications, we opted to use Flutter, a single codebase that allows for efficient cross-platform development. This enabled us to maintain consistency in the UX across platforms, while reducing time and resource demands for development.

Backend Integration with AI Models The heart of the application was FaceLIFT’s AI models, which utilised autoencoders and Generative Adversarial Networks (GANs) to generate facial images from text descriptions. Our backend developers worked closely with the client’s AI developer to ensure seamless integration of these models with the app’s backend. This collaboration was key to enabling smooth communication between the user interface and the AI, allowing users to input text and receive accurate, AI-generated facial images quickly.

The Result

The app now provides users with an intuitive interface both on Mobile and on Web integrated with FaceLIFT’s AI technology model. This project has not only met the initial goal of entertainment and personal engagement but has also opened possibilities for expansion into more serious applications, such as forensic science and educational tools.