Driving Innovation in Geneva: The ALFIE Project showcases cutting-edge research at ITU Kaleidoscope 2026

A speaker stands at a podium delivering a presentation at an ITU Kaleidoscope conference. Behind him, a large projection screen displays a slide titled "Why It Matters: Safety & Regulation." The slide contains statistics on drowsy driving (91,000 US crashes annually) and details about the EU Mandate (GSR 2019/2144), alongside a colorful, geometric network graphic. Another panelist sits at a desk to the left, and the backs of audience members are visible in the foreground.
Zenon Lambrou from Catalink presenting at the ITU Kaleidoscope conference.

From 7 to 9 July, 2026, the global tech and policy community gathered in Geneva, Switzerland, for the ITU Kaleidoscope conference. Bringing together leading thinkers from industry, policy, and academia, this year’s conference served as a forum for discussing the future for standardised, trustworthy AI. 


Representing the ALFIE project, representatives from Universitat Autonoma de Barcelona (UAB) and Catalink presented research addressing two important applications of AI: driver drowsiness detection and sign language interpreting.


Enhancing well-being and public safety: Driver Drowiness Detection

 

Zenon Lambrou from Catalink presented a research paper titled “Driver Drowsiness Detection on the Edge.” Featured as part of the conference’s specialized track on AI in Healthcare and Well-being Applications, the paper tackles a critical safety issue using next-generation AI applications.


This research focused on evaluating different drowsiness models alongside the custom trained model provided from Catalink Limited on 3 different datasets. It also tested 2 different methods and tested the mobile version of each model. The findings showed that the mobile versions underperformed compared to the computer version. Finally is highlighted that the raw performance of these models might be misleading and that a bias based evaluation might be needed.

 

Sign Language Interpreting in the Digital Age 

 

Shifting the lens to sign language interpreting, Dr Estel·la Oncins (UAB) presented a poster and paper on Artificial Intelligence Driven Sign Language Interpreting Avatars. 

 

Dr. Estella Oncins and a colleague reviewing a research poster on "Artificial Intelligence Driven Sign Language Interpreting Avatars" during an academic conference exhibition.
Dr Estel·la Oncins presenting a poster at the ITU Kaleidoscope conference

Dr. Oncins’ research explores how AI can be leveraged to create realistic, accurate, and automated sign language avatars. By integrating accessibility standards with advanced AI, this initiative aims to make digital content, broadcasting, and virtual environments seamlessly accessible to the Deaf community. The presentation sparked vital conversations among policy-makers and standardisation experts on how to ensure future digital ecosystems are inclusive by design, rather than as an afterthought.

 

The ALFIE Mission: Connecting Research, Standards, and Society

 

The ITU Kaleidoscope conference is unique because of its close ties to international standardization bodies. For the ALFIE project, participating in this event is a crucial step toward turning academic research into global benchmarks.

By showcasing innovations in both media accessibility (Sign Language Avatars) and AI safety (Drowsiness Detection), ALFIE continues to demonstrate its core mission: fostering technological ecosystems that are trustworthy, safe, and accessible to everyone, everywhere.

 

Stay Connected with ALFIE

 

Want to dive deeper into the papers and posters presented at ITU Kaleidoscope 2026? We will be uploading the official publications and research materials to our Zenodo Community shortly.

Follow us on social media for more live updates from our global research engagements!

 

Scroll to Top
Privacy Overview

This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.

You can review our privacy policy here.