Implemetion of a face swap application in images

Read full text Summarise with VIRAT
Authors: Γεώργιος Ν. Αντωνίου
Cited by: 0
References: 23

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This undergraduate thesis aims to implement a face recognition and face swapping application using computer vision and deep learning techniques.Its main goal is to compare a user's image with a celebrity image database (dataset), identify the most similar celebrity, and finally replace the celebrity's face with the user's face to create a final image that is a composite of facial features from both images.The implementation is done using the Python programming language and primarily uses the OpenCV, dlib, and face recognition libraries, as they are the most useful for such applications.Initially, face detection and feature extraction (face encodings) are performed on the user's image and the images in the database.Subsequently, the similarity distance between the features is calculated to identify the closest match based on similarity to the user's face.For the face replacement process, Delaunay triangulation is used, which is based on facial landmarks and seamless cloning techniques for the best visual result.The final system offers the user a complete experience, displaying the original image, the closest celebrity match, and the final face swap result.This paper presents the capabilities of modern computer vision and image processing techniques, as well as the challenges related to the accuracy, performance, and quality of the results.

Fields of Study & Subjects

Face recognition and analysis Face and Expression Recognition Generative Adversarial Networks and Image Synthesis

Publication details

Journal / source
University of Thessaly Institutional Repository (University of Thessaly)
Publication date
1 January 2026
Language
EL
Document type
Thesis