AUTOPILOT VQA: Benchmarking Vision-Language Models for Incident-Centric Dashcam Understanding

论文概要

研究领域: 自动驾驶 作者: Siddharth Damodharan, Radhika Gupta, Ali Alshami 发布时间: 2025-07-12 arXiv: 2507.08722

中文摘要

我们提出了AUTOPILOT-VQA，一个以事件为中心的视觉问答基准，用于行车记录仪视频理解。该数据集通过围绕真实驾驶事件和近事件设计的结构化问题来评估不同系统。基准涵盖多样化的安全相关类别，包括天气和光照条件、交通环境、道路布局、路面状态、标志、涉及实体、事故发生、撞击位置以及可避免性相关推理。通过要求模型回答关于上下文场景属性和事件级事件细节的基于事实的问题，AUTOPILOT-VQA超越了物体识别，迈向时间定位的、安全意识的推理。该数据集作为AUTOPILOT CVPR 2026竞赛的一部分发布。

原文摘要

Recent advances in Vision-Language Models, Large Language Models, and Multimodal Large Language Models have improved autonomous driving tasks. However, evaluating whether these models can reliably reason about safety-critical incidents remains challenging. To address this gap, we present AUTOPILOT-VQA, an incident-centric visual question answering benchmark for dashcam video understanding. The dataset evaluates different systems through structured questions designed around real-world driving incidents and near-incidents. The benchmark covers diverse safety-relevant categories, including weather and lighting conditions, traffic environment, road layout, road surface state, signage, involved entities, accident occurrence, impact location, and avoidability-related reasoning.

--- *自动采集于 2025-07-13*

#论文 #arXiv #自动驾驶 #小凯