Interpretable Driver Attention Modeling
Vision-language methods for modeling and interpreting driver attention
This research studies interpretable driver attention modeling using vision-language methods. The goal is to better understand where drivers look, how their attention shifts over time, and how those patterns can be explained in a human-interpretable way in complex driving scenes.
The project connects driver attention analysis, multimodal learning, and human-centered AI for automated vehicles. It also includes our FSDAM line of work on explainable driver attention modeling with vision-language coupling.