Deep Feature Learning for Image-Based Kinship Verification

Facial image-based kinship verification is one of the challenging tasks in computer vision. It has many potential applications, such as human trafficking, studying human genetics, generating family maps, family photo albums, etc. Therefore, we propose a deep feature learning method (DFLKV) which can extract more discriminative features for kinship verification. For a pair of facial images, we firstly design a network with multi-scale channel attention for the features extraction; then, select four methods for feature fusion; finally, infer kinship based on the fused features. We jointly adopt the contrastive loss and the binary cross-entropy loss to compute matching degree for paired samples. The experimental results on four widely used datasets KinFaceW-I, KinFaceW-II, Cornell KinFace and TS KinFace to validate the effectiveness of our approach.