Multi-Frame Vision-Language Model for Long-form Reasoning in Driver Behavior Analysis

TL;DR

This paper introduces a multi-modal vision-language model trained on a new dataset for analyzing risky driving behaviors from dashcam footage, aiming to improve driver coaching and safety.

Contribution

It presents a novel multi-modal instruction tuning dataset and a driver coaching inference system for long-form reasoning in driver behavior analysis.

Findings

Effective in identifying risky driving behaviors

Enhances driver coaching with detailed reasoning

Applicable to real-world dashcam footage

Abstract

Identifying risky driving behavior in real-world situations is essential for the safety of both drivers and pedestrians. However, integrating natural language models in this field remains relatively untapped. To address this, we created a novel multi-modal instruction tuning dataset and driver coaching inference system. Our primary use case is dashcam-based coaching for commercial drivers. The North American Dashcam Market is expected to register a CAGR of 15.4 percent from 2022 to 2027. Our dataset enables language models to learn visual instructions across various risky driving scenarios, emphasizing detailed reasoning crucial for effective driver coaching and managerial comprehension. Our model is trained on road-facing and driver-facing RGB camera footage, capturing the comprehensive scope of driving behavior in vehicles equipped with dashcams.