LAGA: Layered 3D Avatar Generation and Customization via Gaussian Splatting

TL;DR

LAGA is a novel framework for creating customizable, high-fidelity 3D avatars with detachable garments, enabling flexible editing and transfer of clothing using layered Gaussian modeling and a dual loss strategy.

Contribution

The paper introduces a layered Gaussian modeling approach with a dual-SDS loss for high-quality, decomposable 3D avatars that can be freely customized and transferred.

Findings

Outperforms existing methods in 3D clothed human generation

Enables flexible garment editing and transfer

Maintains coherence between avatar components

Abstract

Creating and customizing a 3D clothed avatar from textual descriptions is a critical and challenging task. Traditional methods often treat the human body and clothing as inseparable, limiting users' ability to freely mix and match garments. In response to this limitation, we present LAyered Gaussian Avatar (LAGA), a carefully designed framework enabling the creation of high-fidelity decomposable avatars with diverse garments. By decoupling garments from avatar, our framework empowers users to conviniently edit avatars at the garment level. Our approach begins by modeling the avatar using a set of Gaussian points organized in a layered structure, where each layer corresponds to a specific garment or the human body itself. To generate high-quality garments for each layer, we introduce a coarse-to-fine strategy for diverse garment generation and a novel dual-SDS loss function to maintain coherence between the generated garments and avatar components, including the human body and other garments. Moreover, we introduce three regularization losses to guide the movement of Gaussians for garment transfer, allowing garments to be freely transferred to various avatars. Extensive experimentation demonstrates that our approach surpasses existing methods in the generation of 3D clothed humans.