DiffAttn: Diffusion-Based Drivers' Visual Attention Prediction with LLM-Enhanced Semantic Reasoning

TL;DR

DiffAttn is a diffusion-based model that predicts drivers' visual attention by integrating transformer-based scene understanding and LLM-enhanced semantic reasoning, achieving state-of-the-art results.

Contribution

The paper introduces DiffAttn, a novel diffusion framework combining transformer encoding and LLM reasoning for improved driver attention prediction.

Findings

Achieves state-of-the-art performance on four public datasets.

Effectively models local and global scene features.

Enhances safety-critical cue sensitivity through LLM integration.

Abstract

Drivers' visual attention provides critical cues for anticipating latent hazards and directly shapes decision-making and control maneuvers, where its absence can compromise traffic safety. To emulate drivers' perception patterns and advance visual attention prediction for intelligent vehicles, we propose DiffAttn, a diffusion-based framework that formulates this task as a conditional diffusion-denoising process, enabling more accurate modeling of drivers' attention. To capture both local and global scene features, we adopt Swin Transformer as encoder and design a decoder that combines a Feature Fusion Pyramid for cross-layer interaction with dense, multi-scale conditional diffusion to jointly enhance denoising learning and model fine-grained local and global scene contexts. Additionally, a large language model (LLM) layer is incorporated to enhance top-down semantic reasoning and improve sensitivity to safety-critical cues. Extensive experiments on four public datasets demonstrate that DiffAttn achieves state-of-the-art (SoTA) performance, surpassing most video-based, top-down-feature-driven, and LLM-enhanced baselines. Our framework further supports interpretable driver-centric scene understanding and has the potential to improve in-cabin human-machine interaction, risk perception, and drivers' state measurement in intelligent vehicles.