Fashion IQ
收藏资源简介:
Fashion IQ是由麻省理工学院-IBM沃森人工智能实验室创建的一个新型数据集,专注于交互式时尚图像检索。该数据集首次包含了人类生成的相对标题,这些标题区分了相似的服装图像对,并附带了真实世界的产品描述和视觉属性标签。Fashion IQ数据集包含多种时尚图像,如连衣裙、衬衫和上衣,以及详细的产品元数据和属性标签。该数据集旨在通过自然语言反馈,推动对话式时尚图像检索的研究,解决传统关键词搜索界面的局限性,提供更自然、用户友好的交互方式。
Fashion IQ is a novel dataset created by the MIT-IBM Watson AI Lab, focusing on interactive fashion image retrieval. This dataset is the first to include human-generated comparative captions that distinguish between pairs of similar clothing images, paired with real-world product descriptions and visual attribute labels. The Fashion IQ dataset contains various fashion images such as dresses, shirts and tops, along with detailed product metadata and attribute labels. This dataset aims to advance research on conversational fashion image retrieval through natural language feedback, addressing the limitations of traditional keyword-based search interfaces and providing a more natural and user-friendly interaction mode.




