HeadGaS: Real-Time Animatable Head Avatars via 3D Gaussian Splatting

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Dhamo, Helisa, Nie, Yinyu, Moreau, Arthur, Song, Jifei, Shaw, Richard, Zhou, Yiren, Pérez-Pellitero, Eduardo
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909285489836032
author Dhamo, Helisa
Nie, Yinyu
Moreau, Arthur
Song, Jifei
Shaw, Richard
Zhou, Yiren
Pérez-Pellitero, Eduardo
author_facet Dhamo, Helisa
Nie, Yinyu
Moreau, Arthur
Song, Jifei
Shaw, Richard
Zhou, Yiren
Pérez-Pellitero, Eduardo
contents 3D head animation has seen major quality and runtime improvements over the last few years, particularly empowered by the advances in differentiable rendering and neural radiance fields. Real-time rendering is a highly desirable goal for real-world applications. We propose HeadGaS, a model that uses 3D Gaussian Splats (3DGS) for 3D head reconstruction and animation. In this paper we introduce a hybrid model that extends the explicit 3DGS representation with a base of learnable latent features, which can be linearly blended with low-dimensional parameters from parametric head models to obtain expression-dependent color and opacity values. We demonstrate that HeadGaS delivers state-of-the-art results in real-time inference frame rates, surpassing baselines by up to 2dB, while accelerating rendering speed by over x10.
format Preprint
id arxiv_https___arxiv_org_abs_2312_02902
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle HeadGaS: Real-Time Animatable Head Avatars via 3D Gaussian Splatting
Dhamo, Helisa
Nie, Yinyu
Moreau, Arthur
Song, Jifei
Shaw, Richard
Zhou, Yiren
Pérez-Pellitero, Eduardo
Computer Vision and Pattern Recognition
3D head animation has seen major quality and runtime improvements over the last few years, particularly empowered by the advances in differentiable rendering and neural radiance fields. Real-time rendering is a highly desirable goal for real-world applications. We propose HeadGaS, a model that uses 3D Gaussian Splats (3DGS) for 3D head reconstruction and animation. In this paper we introduce a hybrid model that extends the explicit 3DGS representation with a base of learnable latent features, which can be linearly blended with low-dimensional parameters from parametric head models to obtain expression-dependent color and opacity values. We demonstrate that HeadGaS delivers state-of-the-art results in real-time inference frame rates, surpassing baselines by up to 2dB, while accelerating rendering speed by over x10.
title HeadGaS: Real-Time Animatable Head Avatars via 3D Gaussian Splatting
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2312.02902