RoadSceneVQA: Benchmarking Visual Question Answering in Roadside Perception Systems for Intelligent Transportation System
RoadSceneVQA: 用于智能交通系统 roadside 系统的视觉问答基准测试
专题命中 视觉问答 :visual question answering(title,abstract);MLLM(abstract);分类 cs.CV
AI总结 RoadSceneVQA 通过大规模标注数据和 CogniAnchor 融合模块,提升多模态大语言模型在交通感知与推理任务中的性能。
Comments 10 pages, 6 figures, accepted by AAAI 2026. The model is also called Dream, to the other me in the world forever