ChatBEV：通过视觉-语言模型赋能交通场景理解与仿真
ChatBEV: Empowering Traffic Scene Understanding and Simulation via Vision-Language Model
AI总结：
本文提出ChatBEV及其基准ChatBEV-QA，通过微调视觉-语言模型实现鸟瞰图场景理解，并集成到语言引导的交通仿真中，显著降低轨迹误差和碰撞率。
摘要：
全面的交通场景理解是智能交通系统（ITS）的基础能力，支撑着交通仿真等应用。尽管视觉-语言模型（VLMs）已展现出强大的推理潜力，但其在交通场景中的鸟瞰图（BEV）应用仍受限于狭窄的任务定义和稀缺的标注数据。我们提出了ChatBEV-QA，一个大规模BEV视觉问答基准，包含超过13.7万个问答对，旨在评估复杂交通环境中的全局场景理解、车辆-车道交互以及车辆-车辆交互。在此基础上，我们微调了ChatBEV，一个专门的视觉-语言模型，能够准确解读来自BEV地图的多样化场景理解查询。为了展示在ITS应用中的下游实用性，我们将ChatBEV集成到一个语言引导的交通仿真框架中。其全局理解和导航推理提供了关键的上下文感知指导，与仅文本的基线相比，轨迹位移误差降低了最多20.9%，场景碰撞率降低了最多37.9%。
英文摘要：
Comprehensive traffic scene understanding is a foundational capability for Intelligent Transportation Systems (ITS) underpinning applications such as traffic simulation. While VisionLanguage Models (VLMs) have demonstrated strong reasoning potential, their application to Bird's-Eye View (BEV) maps in traffic contexts remains limited by narrow task definitions and scarce annotated data. We introduce ChatBEV-QA, a large-scale BEV VQA benchmark of 137K+ QA pairs, designed to evaluate global scene understanding, vehicle-lane interactions, and vehiclevehicle interactions within complex traffic environments. Building on this, we fine-tune ChatBEV, a specialized VLM that accurately interprets diverse scene understanding queries from BEV maps. To demonstrate downstream utility in ITS applications, we integrate ChatBEV into a language-guided traffic simulation framework. Its global understanding and navigation reasoning provide crucial context-aware guidance, reducing trajectory displacement error by up to 20.9% and scenario collision rates by up to 37.9% over text-only baselines.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
