CapMIT1003

该数据集名为CapMIT1003，包含了带注释的图片以及与点击相关的图像探索，旨在研究人类在标注任务过程中的注意力情况。注释要求使用一到两句话的英文描述，并且不会收集任何个人信息。该数据集的规模涉及27,865次点击，分布在4,573个观察样本上，其任务是进行标注任务。

This dataset, named CapMIT1003, comprises annotated images and click-associated image exploration content, aiming to investigate human attention during annotation tasks. Annotations are required to be described in one or two English sentences, and no personal information is collected. The dataset includes 27,865 clicks distributed across 4,573 observational samples, and it is centered around annotation tasks.