DynAvatar: Dynamic 3D Head Avatar Deformation With Expression Guided Gaussian Splatting
Lists
Only you see this
Abstract
Generating high-fidelity, expressive, and realistic 3D head avatars remains a fundamental challenge for immersive applications such as virtual reality, gaming, and telepresence. This task requires not only precise modeling of non-rigid facial deformations but also semantically controllable expression synthesis under diverse viewpoints and motion contexts. We present DynAvatar, a novel framework that integrates expression-guided deformation into the 3D Gaussian splatting pipeline to produce photorealistic and emotionally resonant head avatars. Our method introduces two key innovations: (1) an expression-guided Gaussian deformation module that tightly couples geometric displacement with high-level semantic cues, enabling fine-grained and anatomically meaningful facial animation; and (2) a spatial context embedding mechanism that encodes the canonical position of each Gaussian to preserve semantic coherence and spatial consistency during expression generation. Extensive experiments on both controlled and in-the-wild datasets demonstrate that DynAvatar significantly outperforms state-of-the-art methods in terms of visual realism, expression fidelity, and rendering quality.
BibTeX
@article{song2026dynavatar,
author = {Wenfeng Song and Zhongyong Ye and Zhenyu Wu and Shuai Li and Xia Hou and Aimin Hao},
title = {{DynAvatar: Dynamic 3D Head Avatar Deformation With Expression Guided Gaussian Splatting}},
journal = {IEEE Trans. Vis. Comput. Graph.},
year = {2026},
doi = {10.1109/tvcg.2025.3640423},
}Loading related content...