Akrabalık doğrulamasında yaş ve cinsiyet normalizasyonu

Kinship verification from facial images using deep learning is an interesting problem that is unsolved and gains growing attention of the research community. However, the most recent kinship verification systems suffer from age- and gender-related facial attributes that cause problems in kinship verification between subjects of different age and gender. In this study, we propose various methods to reduce the negative effect of the age- and gender-related facial attributes in kinship verification to achieve a more robust verification model. The proposed approach utilizes the comprehensive modeling capabilities of the recent generative adversarial network architectures to model the age and gender of subjects and reduce their effect in kinship verification, if not remove entirely. Furthermore, we conduct a thorough analysis over individual and combined effects of age and gender normalization, performed in both image and latent space of the generative models. Lastly, we investigate the impact of additional emphasis on the facial identity information during the normalization process. Taking one of the most recent kinship verification models as our baseline, we show that gender normalization has reduced the verification performance gap between subject pairs with the same and different gender, up to 6%. Furthermore, joint normalization of age and gender improves the kinship verification accuracy up to 5% and 10% on two different in-the-wild kinship datasets. Therefore, this thesis proposes generic approaches to improve the reliability and robustness of kinship verification by normalizing the age and gender attributes without making changes in the core architecture of the employed kinship verification system.