VRD (Visual Relationship Detection dataset)

The Visual Relationship Dataset (VRD) contains 4000 images for training and 1000 for testing annotated with visual relationships. Bounding boxes are annotated with a label containing 100 unary predicates. These labels refer to animals, vehicles, clothes and generic objects. Pairs of bounding boxes are annotated with a label containing 70 binary predicates. These labels refer to actions, prepositions, spatial relations, comparatives or preposition phrases. The dataset has 37993 instances of visual relationships and 6672 types of relationships. 1877 instances of relationships occur only in the test set and they are used to evaluate the zero-shot learning scenario.

视觉关系数据集（Visual Relationship Dataset，VRD）包含4000张训练图像与1000张测试图像，所有图像均完成了视觉关系标注。边界框（bounding box）被标注有包含100种一元谓词的标签，此类标签涵盖动物、交通工具、衣物以及通用物体。成对边界框则被标注有包含70种二元谓词的标签，这类标签对应动作、介词、空间关系、比较关系或介词短语。该数据集总计包含37993个视觉关系实例与6672种关系类型，其中1877个关系实例仅出现在测试集中，用于评估零样本（Zero-shot）学习场景。