Transferring Rich Deep Features for Facial Beauty Prediction

TL;DR

This paper introduces a method that transfers deep features from pretrained face verification models to predict facial beauty, achieving improved or comparable results and enhancing interpretability.

Contribution

It proposes a novel approach combining deep feature transfer with Bayesian regression for facial beauty prediction, demonstrating effectiveness and interpretability.

Findings

Improved performance on SCUT-FBP and ECCV HotOrNot datasets

Effective feature fusion strategy enhances prediction accuracy

Provides insights into facial beauty perception

Abstract

Feature extraction plays a significant part in computer vision tasks. In this paper, we propose a method which transfers rich deep features from a pretrained model on face verification task and feeds the features into Bayesian ridge regression algorithm for facial beauty prediction. We leverage the deep neural networks that extracts more abstract features from stacked layers. Through simple but effective feature fusion strategy, our method achieves improved or comparable performance on SCUT-FBP dataset and ECCV HotOrNot dataset. Our experiments demonstrate the effectiveness of the proposed method and clarify the inner interpretability of facial beauty perception.