Skip to Main content Skip to Navigation
New interface
Journal articles

Kinship verification through facial images using multiscale and multilevel handcrafted features

Abstract : We address kinship verification, which is a challenging problem in computer vision and pattern discovery. It has several applications, such as organizing photoalbums, recognizing resemblances among humans, and finding missing children. We present a system for facial kinship verification based on several kinds of texture descriptors (local binary patterns, local ternary patterns, local directional patterns, local phase quantization, and binarized statistical image features) with pyramid multilevel (PML) face representation for feature extraction along with our proposed paired feature representation and our proposed robust feature selection to reduce the number of features. The proposed approach consists of the following three main stages: (1) face preprocessing, (2) feature extraction and selection, and (3) kinship verification. Extensive experiments are conducted on five publicly available databases (Cornell, UB KinFace, Family 101, KinFace W-I, and KinFace W-II). Additionally, we provided a wide experiment for each stage to find the best and most suitable settings. We present many comparisons with state-of-the-art methods and through these comparisons, it appears that our experiments show stable and good results.
Complete list of metadata
Contributor : Sébastien MAVROMATIS Connect in order to contact the contributor
Submitted on : Monday, February 22, 2021 - 10:18:28 PM
Last modification on : Sunday, June 26, 2022 - 10:15:34 AM
Long-term archiving on: : Sunday, May 23, 2021 - 7:17:22 PM


Publisher files allowed on an open archive






Abdelhakim Chergui, Salim Ouchtati, Sébastien Mavromatis, Salah Eddine Bekhouche, Jean Sequeira, et al.. Kinship verification through facial images using multiscale and multilevel handcrafted features. Journal of Electronic Imaging, 2020, 29 (02), pp.1. ⟨10.1117/1.JEI.29.2.023017⟩. ⟨hal-02531188⟩



Record views


Files downloads