SBU Captions Dataset

该数据集包含了与SBU数据集中的图片相关联的标题，这些标题被用于图像文本检索的预训练。此外，该数据集还是预训练数据集的一部分，它与COCO、VG和CC数据集结合使用。其任务是进行图像文本检索的预训练。

This dataset contains captions associated with images from the SBU dataset, which are used for pre-training in image-text retrieval. Furthermore, this dataset is part of a pre-training corpus and is used in conjunction with the COCO, VG, and CC datasets. Its core objective is to support pre-training for image-text retrieval.