jasonzhango/SPAR-7M
收藏Hugging Face2025-09-28 更新2025-10-25 收录
下载链接:
https://hf-mirror.com/datasets/jasonzhango/SPAR-7M
下载链接
链接失效反馈官方服务:
资源简介:
SPAR-7M是一个大型的视觉语言数据集,专为空间感知和推理任务设计。包含超过700万的问题回答对,涵盖33种不同的空间任务,由4500多个丰富的三维室内场景生成。支持单视图、多视图和基于视频的图像输入,并包含面向感知和推理的问题类型。
SPAR-7M is a large-scale vision-language dataset designed for spatial perception and reasoning. It contains over 7 million QA pairs across 33 diverse spatial tasks, generated from more than 4,500 richly annotated 3D indoor scenes. It supports single-view, multi-view, and video-based image inputs, and features both perception and reasoning-oriented question types.
提供机构:
jasonzhango



