Families in the Wild (FIW)

Families in the Wild (FIW) 是由东北大学电气与计算机工程系和计算机与信息科学学院联合创建的大型亲属识别数据集。该数据集包含1,000个家庭的11,163张不受限制的家庭照片，旨在通过提供丰富的家庭信息标签来吸引研究社区的兴趣。FIW数据集通过专门设计的标注工具快速有效地标记复杂的层次关系和局部标签信息，涵盖了11种亲属关系类型，总计418,060对图像。该数据集的应用领域包括亲属验证、自动照片库管理、历史血统和家谱研究、社交媒体分析以及多种安全应用，如寻找失踪人员和犯罪现场调查，旨在通过这些应用提升监控系统的整体性能。

Families in the Wild (FIW) is a large-scale kinship recognition dataset jointly created by the Department of Electrical and Computer Engineering and the College of Computer and Information Science of Northeastern University. This dataset contains 11,163 unrestricted family photos from 1,000 families, aiming to attract the research community's interest by providing rich family information annotations. FIW uses a specially designed annotation tool to quickly and efficiently mark complex hierarchical relationships and local label information, covering 11 types of kinship relations and totaling 418,060 image pairs. Its application scenarios include kinship verification, automated photo library management, historical ancestry and genealogical research, social media analysis, and various security applications such as missing person search and crime scene investigation, with the goal of improving the overall performance of monitoring systems through these applications.