Kinship Verification
Kinship verification, the automated determination of familial relationships from images or other data, aims to improve accuracy and robustness in identifying kinship ties. Current research focuses on developing advanced algorithms, including convolutional neural networks (CNNs), vision transformers, and contrastive learning methods, often incorporating multi-task learning and feature fusion techniques to enhance performance. These advancements are driven by the need for improved accuracy in applications such as forensic science, social media analysis, and missing person identification, while also addressing challenges like bias mitigation and handling diverse data sources, including facial images and biosignals. The field is also exploring the use of linguistic data to understand the diverse ways kinship is represented across languages and cultures.
Papers
A new method color MS-BSIF Features learning for the robust kinship verification
Rachid Aliradi, Abdealmalik Ouamane, Abdeslam Amrane
Fusion of Deep and Shallow Features for Face Kinship Verification
Belabbaci El Ouanas, Khammari Mohammed, Chouchane Ammar, Mohcene Bessaoudi, Abdelmalik Ouamane, Akram Abderraouf Gharbi