Generalizable and Animatable 3D Full-Head Gaussian Avatar from a Single Image

2026-01-19

AI Summary

Existing methods struggle to generate high-quality, animatable full-head 3D avatars under large viewpoint variations. This work proposes an end-to-end framework that, for the first time, enables generalized, real-time animatable 360-degree full-head reconstruction from a single image. The approach integrates a parametric face model with Gaussian primitives represented in UV space, leveraging a full-head prior derived from a pre-trained 3D GAN. Through multi-view supervision and a symmetry-aware feature fusion mechanism, the method effectively combines global structure with local detail. Experiments demonstrate that our approach significantly outperforms state-of-the-art methods in terms of geometric and texture fidelity, multi-view consistency, and naturalness of facial dynamics.

Technology Category

• Computer Vision: 3D Computer Vision
• Machine Learning: Learning with Manifolds
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

Building 3D animatable head avatars from a single image is an important yet challenging problem. Existing methods generally collapse under large camera pose variations, compromising the realism of 3D avatars. In this work, we propose a new framework to tackle the novel setting of one-shot 3D full-head animatable avatar reconstruction in a single feed-forward pass, enabling real-time animation and simultaneous 360$^\circ$ rendering views. To facilitate efficient animation control, we model 3D head avatars with Gaussian primitives embedded on the surface of a parametric face model within the UV space. To obtain knowledge of full-head geometry and textures, we leverage rich 3D full-head priors within a pretrained 3D generative adversarial network (GAN) for global full-head feature extraction and multi-view supervision. To increase the fidelity of the 3D reconstruction of the input image, we take advantage of the symmetric nature of the UV space and human faces to fuse local fine-grained input image features with the global full-head textures. Extensive experiments demonstrate the effectiveness of our method, achieving high-quality 3D full-head modeling as well as real-time animation, thereby improving the realism of 3D talking avatars.

Problem

• 3D avatar
• single-image reconstruction
• animatable head
• full-head modeling
• real-time animation

Innovation

• Gaussian Avatar
• Single-Image 3D Reconstruction
• Animatable 3D Head
• UV Space Embedding
• 3D GAN Prior