Gaze-Informed Vision Transformers: Predicting Driving Decisions Under Uncertainty

TL;DR

This paper enhances Vision Transformers for driving decision prediction by integrating human eye gaze data, introducing a novel loss function, and demonstrating improved accuracy and attention alignment under uncertain conditions.

Contribution

We introduce FAX loss to incorporate eye gaze into ViT, improving its focus and accuracy in driving scenarios with uncertainty, a novel approach in human-centered AI.

Findings

Gaze data improves ViT attention alignment with human focus.

FAX loss significantly boosts prediction accuracy under uncertainty.

Gaze-informed ViT outperforms baseline models in driving decision tasks.

Abstract

Vision Transformers (ViT) have advanced computer vision, yet their efficacy in complex tasks like driving remains less explored. This study enhances ViT by integrating human eye gaze, captured via eye-tracking, to increase prediction accuracy in driving scenarios under uncertainty in both real-world and virtual reality scenarios. First, we establish the significance of human eye gaze in left-right driving decisions, as observed in both human subjects and a ViT model. By comparing the similarity between human fixation maps and ViT attention weights, we reveal the dynamics of overlap across individual heads and layers. This overlap demonstrates that fixation data can guide the model in distributing its attention weights more effectively. We introduce the fixation-attention intersection (FAX) loss, a novel loss function that significantly improves ViT performance under high uncertainty conditions. Our results show that ViT, when trained with FAX loss, aligns its attention with human gaze patterns. This gaze-informed approach has significant potential for driver behavior analysis, as well as broader applications in human-centered AI systems, extending ViT's use to complex visual environments.