GETAvatar: Generative Textured Meshes for Animatable Human Avatars

TL;DR

GETAvatar introduces a novel generative model that creates high-quality, animatable 3D human meshes with realistic textures and details, enabling efficient high-resolution rendering for virtual human avatars.

Contribution

It proposes an explicit textured 3D mesh generation approach with surface detail learning and rasterization-based rendering, surpassing volumetric methods in quality and efficiency.

Findings

Achieves state-of-the-art 3D-aware human generation quality.

Generates 512x512 images at 17FPS and 1024x1024 at 14FPS.

Outperforms previous methods by 2x in resolution and speed.

Abstract

We study the problem of 3D-aware full-body human generation, aiming at creating animatable human avatars with high-quality textures and geometries. Generally, two challenges remain in this field: i) existing methods struggle to generate geometries with rich realistic details such as the wrinkles of garments; ii) they typically utilize volumetric radiance fields and neural renderers in the synthesis process, making high-resolution rendering non-trivial. To overcome these problems, we propose GETAvatar, a Generative model that directly generates Explicit Textured 3D meshes for animatable human Avatar, with photo-realistic appearance and fine geometric details. Specifically, we first design an articulated 3D human representation with explicit surface modeling, and enrich the generated humans with realistic surface details by learning from the 2D normal maps of 3D scan data. Second, with the explicit mesh representation, we can use a rasterization-based renderer to perform surface rendering, allowing us to achieve high-resolution image generation efficiently. Extensive experiments demonstrate that GETAvatar achieves state-of-the-art performance on 3D-aware human generation both in appearance and geometry quality. Notably, GETAvatar can generate images at 512x512 resolution with 17FPS and 1024x1024 resolution with 14FPS, improving upon previous methods by 2x. Our code and models will be available.