FAGhead: Fully Animate Gaussian Head from Monocular Videos

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xuan, Yixin, Li, Xinyang, Yao, Gongxin, Zhou, Shiwei, Sun, Donghui, Chen, Xiaoxin, Pan, Yu
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913406984912896
author Xuan, Yixin
Li, Xinyang
Yao, Gongxin
Zhou, Shiwei
Sun, Donghui
Chen, Xiaoxin
Pan, Yu
author_facet Xuan, Yixin
Li, Xinyang
Yao, Gongxin
Zhou, Shiwei
Sun, Donghui
Chen, Xiaoxin
Pan, Yu
contents High-fidelity reconstruction of 3D human avatars has a wild application in visual reality. In this paper, we introduce FAGhead, a method that enables fully controllable human portraits from monocular videos. We explicit the traditional 3D morphable meshes (3DMM) and optimize the neutral 3D Gaussians to reconstruct with complex expressions. Furthermore, we employ a novel Point-based Learnable Representation Field (PLRF) with learnable Gaussian point positions to enhance reconstruction performance. Meanwhile, to effectively manage the edges of avatars, we introduced the alpha rendering to supervise the alpha value of each pixel. Extensive experimental results on the open-source datasets and our capturing datasets demonstrate that our approach is able to generate high-fidelity 3D head avatars and fully control the expression and pose of the virtual avatars, which is outperforming than existing works.
format Preprint
id arxiv_https___arxiv_org_abs_2406_19070
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle FAGhead: Fully Animate Gaussian Head from Monocular Videos
Xuan, Yixin
Li, Xinyang
Yao, Gongxin
Zhou, Shiwei
Sun, Donghui
Chen, Xiaoxin
Pan, Yu
Computer Vision and Pattern Recognition
High-fidelity reconstruction of 3D human avatars has a wild application in visual reality. In this paper, we introduce FAGhead, a method that enables fully controllable human portraits from monocular videos. We explicit the traditional 3D morphable meshes (3DMM) and optimize the neutral 3D Gaussians to reconstruct with complex expressions. Furthermore, we employ a novel Point-based Learnable Representation Field (PLRF) with learnable Gaussian point positions to enhance reconstruction performance. Meanwhile, to effectively manage the edges of avatars, we introduced the alpha rendering to supervise the alpha value of each pixel. Extensive experimental results on the open-source datasets and our capturing datasets demonstrate that our approach is able to generate high-fidelity 3D head avatars and fully control the expression and pose of the virtual avatars, which is outperforming than existing works.
title FAGhead: Fully Animate Gaussian Head from Monocular Videos
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2406.19070