aialliance/GEOBench-VLM
收藏Hugging Face2025-06-30 更新2024-12-14 收录
下载链接:
https://hf-mirror.com/datasets/aialliance/GEOBench-VLM
下载链接
链接失效反馈官方服务:
资源简介:
GEOBench-VLM是一个专门用于评估视觉-语言模型(VLMs)在地理空间任务中表现的基准数据集。该数据集包含多个子集,每个子集对应不同的任务,如场景理解、物体计数、定位、细粒度分类和时间分析等。数据集包含超过10,000条手动验证的指令,涵盖了视觉条件、物体类型和尺度的多样性。数据集的结构包括多个文件夹,每个文件夹包含特定任务的图像和相关问题。数据实例展示了问题的结构,包括图像路径、正确答案、选项列表、任务类型等。
The GEOBench-VLM dataset is a benchmark specifically designed to evaluate Vision-Language Models (VLMs) on geospatial tasks. It contains over 10,000 manually verified instructions, covering a diverse set of variations in visual conditions, object type, and scale, involving tasks such as scene understanding, object counting, localization, fine-grained categorization, and temporal analysis.
提供机构:
aialliance



