Deep Kinship Verification via Appearance-shape Joint Prediction and Adaptation-based Approach

TL;DR

This paper introduces a deep learning approach that combines appearance and shape features for kinship verification, leveraging transfer learning and joint prediction to improve accuracy on benchmark datasets.

Contribution

It proposes a novel appearance-shape joint prediction pipeline with adaptation-based learning for kinship verification, outperforming existing methods.

Findings

Achieved superior accuracy on a popular kinship verification benchmark.

Effectively combined appearance and shape features for improved prediction.

Demonstrated the effectiveness of transfer learning from face recognition models.

Abstract

Kinship verification aims to identify the kin relation between two given face images. It is a very challenging problem due to the lack of training data and facial similarity variations between kinship pairs. In this work, we build a novel appearance and shape based deep learning pipeline. First we adopt the knowledge learned from general face recognition network to learn general facial features. Afterwards, we learn kinship oriented appearance and shape features from kinship pairs and combine them for the final prediction. We have evaluated the model performance on a widely used popular benchmark and demonstrated the superiority over the state-of-the-art.