AnyAvatar: High-Fidelity Gaussian Head Avatars under Uncalibrated Camera Settings

Accepted · ACM MM 2026

Yujian Liu^#, Dongxu Shen^#, Haoran Li^#, Yuting Liu, Chuang Chen, Xinyi Jiang, Zhupeng Jiang, Peng Cao^*, Shidang Xu^*, Xiaoli Liu^*

^* Corresponding author ^# Co-first author (equal contribution)

Type

Conference paper

Publication

ACM International Conference on Multimedia, 2026 — Accepted

Accepted for ACM International Conference on Multimedia (ACM MM) 2026. Proceedings details will be added when available.

Visit the authors’ project page.

The authors' pipeline connects ray-based localization and multi-view FLAME fitting to Gaussian binding, camera-pose optimization, structured triplane appearance and head rendering from new viewpoints. — AnyAvatar: camera-pose refinement, Gaussian head representation and novel-view rendering.
Yujian Liu et al., AnyAvatar authors' project figure. Full figure reproduced; display size and image format only are adjusted. Noncommercial reuse. · Authors’ project figure · CC BY-NC 4.0
View full-size figure