Skip to content

Using MEBeauty dataset to analyse and predict facial attractiveness ratings

Notifications You must be signed in to change notification settings

Pantalaymon/facial-attractiveness-ratings

Repository files navigation

Overview

Facial attractiveness plays a significant role in various aspects of life, from social interactions to professional opportunities. This project delves into the complex interplay of facial features and their impact on perceived attractiveness.

In our pursuit of understanding facial ratings, we delve into the Multi-Ethnic Dataset, an expansive repository capturing the diverse spectrum of human faces. Through data-driven exploration, we aim to unravel the intricacies of facial aesthetics and perceptions of attractiveness. Our journey unfolds across four distinct domains, each offering unique insights into the realm of facial ratings:

Analysing Facial Ratings

We conduct a comprehensive examination of the ratings assigned to faces within the dataset, employing statistical tools and visualization techniques to uncover underlying patterns and trends.

File : Facial_ratings_analysis.ipynb

Identifying Facial Features Associated with Attractiveness/Unattractiveness

Utilizing feature extraction algorithms and machine learning methodologies, we seek to identify the subtle facial attributes that correlate with varying degrees of attractiveness.

File : Incoming

Predicting Attractiveness Rating of Faces

Through the construction of predictive models, we endeavor to estimate facial attractiveness ratings with precision, leveraging the wealth of data encapsulated within the Multi-Ethnic Dataset.

File : incoming

Generating Faces with a Given Level of Attractiveness

Employing generative algorithms and deep learning architectures, we explore the realm of synthetic facial generation, crafting virtual visages imbued with predetermined levels of attractiveness.

File: incoming

About

Using MEBeauty dataset to analyse and predict facial attractiveness ratings

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published