Digital makeup

In today's world, where smartphones are equipped with advanced cameras and social media is deeply intertwined with daily life, the ability to capture and share images has become second nature. Along with the rise of social media platforms, the beauty industry has also adapted to the digital era...

Full description

Saved in:
Bibliographic Details
Main Author: Lau, Yong Jie
Other Authors: He Ying
Format: Final Year Project
Language:English
Published: Nanyang Technological University 2024
Subjects:
Online Access:https://hdl.handle.net/10356/181167
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:In today's world, where smartphones are equipped with advanced cameras and social media is deeply intertwined with daily life, the ability to capture and share images has become second nature. Along with the rise of social media platforms, the beauty industry has also adapted to the digital era. One notable innovation is the development of makeup application tools that allow users to enhance or completely transform their selfies through “filters”. This research aims to explore the technology behind such filters and develop a similar algorithm. The author will examine and implement a segmentation artificial intelligence (AI) model, specifically the Fully Convolutional Network (FCN). The author will then apply two colour transfer techniques - Reinhard’s colour transfer and Principal Component Colour Matching (PCCM) - to the segmented images and evaluate the results in the context of makeup transfer. The study aims to evaluate and deliver a functioning algorithm, tested with several models for its effectiveness and usability. While discussing the results, the author will also discuss the limitations of each approach and provide recommendations for further improvement.