Our Research team is interested in capturing data about people’s experience of the built environment, as hinted at in the last post (despite the fact it was actually for April 1st, as anyone who read it closely hopefully realized).
One aspect of this work has been covered in a recent publication that has been accepted as “late breaking work” at CHI 2024 – the Human-Computer Interaction conference being held next month in Hawai'i.
Here’s a brief video that describes the work:
The project looked at whether Vision-Language Models (VLMs) could be used to help evaluate designs along human-centric dimensions. To learn more, go ahead and read the paper.
Congratulations to the team involved in the project itself, creating the publication and getting it accepted to CHI! Bon Adriel Aseniero, Michael Lee, Yi Wang, Qian Zhou, Nastaran Shahmansouri and Rhys Goldstein.