Saved in:
Bibliographic Details
Main Authors: Liu, Xinqi, Wu, Chenming, Liu, Jialun, Liu, Xing, Wu, Jinbo, Zhao, Chen, Feng, Haocheng, Ding, Errui, Wang, Jingdong
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2402.16607
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910372753047552
author Liu, Xinqi
Wu, Chenming
Liu, Jialun
Liu, Xing
Wu, Jinbo
Zhao, Chen
Feng, Haocheng
Ding, Errui
Wang, Jingdong
author_facet Liu, Xinqi
Wu, Chenming
Liu, Jialun
Liu, Xing
Wu, Jinbo
Zhao, Chen
Feng, Haocheng
Ding, Errui
Wang, Jingdong
contents In this paper, we present a novel method that facilitates the creation of vivid 3D Gaussian avatars from monocular video inputs (GVA). Our innovation lies in addressing the intricate challenges of delivering high-fidelity human body reconstructions and aligning 3D Gaussians with human skin surfaces accurately. The key contributions of this paper are twofold. Firstly, we introduce a pose refinement technique to improve hand and foot pose accuracy by aligning normal maps and silhouettes. Precise pose is crucial for correct shape and appearance reconstruction. Secondly, we address the problems of unbalanced aggregation and initialization bias that previously diminished the quality of 3D Gaussian avatars, through a novel surface-guided re-initialization method that ensures accurate alignment of 3D Gaussian points with avatar surfaces. Experimental results demonstrate that our proposed method achieves high-fidelity and vivid 3D Gaussian avatar reconstruction. Extensive experimental analyses validate the performance qualitatively and quantitatively, demonstrating that it achieves state-of-the-art performance in photo-realistic novel view synthesis while offering fine-grained control over the human body and hand pose. Project page: https://3d-aigc.github.io/GVA/.
format Preprint
id arxiv_https___arxiv_org_abs_2402_16607
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle GVA: Reconstructing Vivid 3D Gaussian Avatars from Monocular Videos
Liu, Xinqi
Wu, Chenming
Liu, Jialun
Liu, Xing
Wu, Jinbo
Zhao, Chen
Feng, Haocheng
Ding, Errui
Wang, Jingdong
Computer Vision and Pattern Recognition
In this paper, we present a novel method that facilitates the creation of vivid 3D Gaussian avatars from monocular video inputs (GVA). Our innovation lies in addressing the intricate challenges of delivering high-fidelity human body reconstructions and aligning 3D Gaussians with human skin surfaces accurately. The key contributions of this paper are twofold. Firstly, we introduce a pose refinement technique to improve hand and foot pose accuracy by aligning normal maps and silhouettes. Precise pose is crucial for correct shape and appearance reconstruction. Secondly, we address the problems of unbalanced aggregation and initialization bias that previously diminished the quality of 3D Gaussian avatars, through a novel surface-guided re-initialization method that ensures accurate alignment of 3D Gaussian points with avatar surfaces. Experimental results demonstrate that our proposed method achieves high-fidelity and vivid 3D Gaussian avatar reconstruction. Extensive experimental analyses validate the performance qualitatively and quantitatively, demonstrating that it achieves state-of-the-art performance in photo-realistic novel view synthesis while offering fine-grained control over the human body and hand pose. Project page: https://3d-aigc.github.io/GVA/.
title GVA: Reconstructing Vivid 3D Gaussian Avatars from Monocular Videos
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2402.16607