Learning to Decouple the Lights for 3D Face Texture Modeling

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Huang, Tianxin, Zhang, Zhenyu, Tai, Ying, Lee, Gim Hee
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866912152451809280
author Huang, Tianxin
Zhang, Zhenyu
Tai, Ying
Lee, Gim Hee
author_facet Huang, Tianxin
Zhang, Zhenyu
Tai, Ying
Lee, Gim Hee
contents Existing research has made impressive strides in reconstructing human facial shapes and textures from images with well-illuminated faces and minimal external occlusions. Nevertheless, it remains challenging to recover accurate facial textures from scenarios with complicated illumination affected by external occlusions, e.g. a face that is partially obscured by items such as a hat. Existing works based on the assumption of single and uniform illumination cannot correctly process these data. In this work, we introduce a novel approach to model 3D facial textures under such unnatural illumination. Instead of assuming single illumination, our framework learns to imitate the unnatural illumination as a composition of multiple separate light conditions combined with learned neural representations, named Light Decoupling. According to experiments on both single images and video sequences, we demonstrate the effectiveness of our approach in modeling facial textures under challenging illumination affected by occlusions. Please check https://tianxinhuang.github.io/projects/Deface for our videos and codes.
format Preprint
id arxiv_https___arxiv_org_abs_2412_08524
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Learning to Decouple the Lights for 3D Face Texture Modeling
Huang, Tianxin
Zhang, Zhenyu
Tai, Ying
Lee, Gim Hee
Computer Vision and Pattern Recognition
Existing research has made impressive strides in reconstructing human facial shapes and textures from images with well-illuminated faces and minimal external occlusions. Nevertheless, it remains challenging to recover accurate facial textures from scenarios with complicated illumination affected by external occlusions, e.g. a face that is partially obscured by items such as a hat. Existing works based on the assumption of single and uniform illumination cannot correctly process these data. In this work, we introduce a novel approach to model 3D facial textures under such unnatural illumination. Instead of assuming single illumination, our framework learns to imitate the unnatural illumination as a composition of multiple separate light conditions combined with learned neural representations, named Light Decoupling. According to experiments on both single images and video sequences, we demonstrate the effectiveness of our approach in modeling facial textures under challenging illumination affected by occlusions. Please check https://tianxinhuang.github.io/projects/Deface for our videos and codes.
title Learning to Decouple the Lights for 3D Face Texture Modeling
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.08524