人脸亲子关系验证研究综述EI北大核心CSCD

人脸亲子关系验证即通过给定的不同人的两幅人脸图像判断其是否具有亲子关系,是计算机视觉和机器学习领域中一个重要的研究问题,在丢失儿童寻找、社会媒体分析、图像自动标注等领域具有广泛的应用价值.随着人脸亲子关系验证问题受到越来越多的关注,其在多个方面都得到了相应的发展,本文对人脸亲子关系验证方法做了综述整理.首先,简要介绍了人脸亲子关系验证在近十年的研究现状,随后对问题进行了定义并讨论其面临的挑战.接下来,汇总了常用的亲子数据库,对数据库属性做了详细的总结和对比.然后,对人脸亲子关系验证方法进行了分类总结、对比,以及不同方法的性能表现.最后,展望了人脸亲子关系验证今后可能的研究方向.

Survey on Facial Kinship Verification

Facial kinship verification aims to verify whether two facial images have kin relations or not. It is an important research topic in computer vision and machine learning domain. Kinship verification has a wide range of applications in fields of missing children finding, social media analysis and image automatic labelling. It receives more and more attention and has been developed in many directions. In this paper, we do a survey on the kinship verification. First, we briefly introduce the kinship verification’s state of research within last ten years. Then we define the problem and discuss the challenges. Then, we summarize and compare the attributes of public kinship databases. Kinship verification algorithms are categorized and reviewed as well. We also analysis different kinship verification methods and compare their performance. Finally, we discuss the future possible research directions of kinship verification.