Vision-Language Models can Identify Distracted Driver Behavior from Naturalistic Videos

arXiv:2306.10159v4 Announce Type: replace
Abstract: Recognizing the activities causing distraction in real-world driving scenarios is critical for ensuring the safety and reliability of both drivers and pedestrians on the roadways. Conventional computer vision techniques are typically data-intensive and require a large volume of annotated training data to detect and classify various distracted driving behaviors, thereby limiting their efficiency and scalability. We aim to develop a generalized framework that showcases robust performance with access to limited or no annotated training data. …

abstract arxiv behavior computer computer vision cs.cv data driver drivers driving identify language language models pedestrians reliability safety training training data type videos vision vision-language models world