Facial Expression Editing of Untrained Subject Image Using Landmarks

View at publisher Summarise with VIRAT
Authors: Kanade Sumino, Shota Harada, Ikuhisa Mitsugami, Ryusuke Sagawa
Cited by: 0

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

Many applications require methods for intuitively controlling facial expressions in images by editing the facial landmark points. However, most existing methods, such as Landmark-GAN, have limitations and can only be applied to trained subject face images. In this paper, we propose a method that can overcome the problem of those existing works. This method is based on First Order Motion Model (FOMM), which can transfer the facial expression of a person (reference video) to a single face image of another person. It uses keypoint features that describe changes in facial expressions acquired through unsupervised learning from face image datasets, but the correspondence between the values of these keypoint features and facial expressions is not clear, which makes intuitive expression editing difficult. This method estimates the change of FOMM keypoint features from that of facial landmark points acquired by OpenFace.

Fields of Study & Subjects

Face recognition and analysis Generative Adversarial Networks and Image Synthesis Digital Media Forensic Detection

Publication details

Journal / source
CRC Press eBooks
Pages
12–21
Publication date
27 March 2026
Language
EN
Document type
Book chapter