FHAvatar: Fast and High-Fidelity Reconstruction of Face-and-Hair Composable 3D Head Avatar from Few Casual Captures

TL;DR

FHAvatar is a fast, high-fidelity framework for reconstructing 3D face-and-hair avatars from minimal casual captures, enabling real-time animation and editing with state-of-the-art quality.

Contribution

It introduces a novel decoupled face and hair representation with an aggregated transformer backbone for efficient multi-view learning from few captures.

Findings

Achieves state-of-the-art reconstruction quality from limited views.

Supports real-time avatar animation and editing.

Reconstructs avatars within minutes from casual captures.

Abstract

We present FHAvatar, a novel framework for reconstructing 3D Gaussian avatars with composable face and hair components from an arbitrary number of views. Unlike previous approaches that couple facial and hair representations within a unified modeling process, we explicitly decouple two components in texture space by representing the face with planar Gaussians and the hair with strand-based Gaussians. To overcome the limitations of existing methods that rely on dense multi-view captures or costly per-identity optimization, we propose an aggregated transformer backbone to learn geometry-aware cross-view priors and head-hair structural coherence from multi-view datasets, enabling effective and efficient feature extraction and fusion from few casual captures. Extensive quantitative and qualitative experiments demonstrate that FHAvatar achieves state-of-the-art reconstruction quality from only a few observations of new identities within minutes, while supporting real-time animation, convenient hairstyle transfer, and stylized editing, broadening the accessibility and applicability of digital avatar creation.