Understanding Beauty via Deep Facial Features

TL;DR

This paper uses deep learning to analyze facial features and objectively quantify beauty, revealing correlations consistent with psychological theories and demonstrating beauty enhancement through generative models.

Contribution

It introduces a data-driven approach combining CNNs and GANs to analyze and enhance facial beauty based on large-scale datasets.

Findings

Correlations between facial features and attractiveness confirmed statistically

Beauty enhancements are visually compelling and statistically verified

Findings align with psychological studies on facial attractiveness

Abstract

The concept of beauty has been debated by philosophers and psychologists for centuries, but most definitions are subjective and metaphysical, and deficit in accuracy, generality, and scalability. In this paper, we present a novel study on mining beauty semantics of facial attributes based on big data, with an attempt to objectively construct descriptions of beauty in a quantitative manner. We first deploy a deep convolutional neural network (CNN) to extract facial attributes, and then investigate correlations between these features and attractiveness on two large-scale datasets labelled with beauty scores. Not only do we discover the secrets of beauty verified by statistical significance tests, our findings also align perfectly with existing psychological studies that, e.g., small nose, high cheekbones, and femininity contribute to attractiveness. We further leverage these high-level representations to original images by a generative adversarial network (GAN). Beauty enhancements after synthesis are visually compelling and statistically convincing verified by a user survey of 10,000 data points.