DGH: Dynamic Gaussian Hair

TL;DR

DGH introduces a data-driven, scalable framework for realistic, dynamic hair modeling that learns hair motion and appearance directly from data, outperforming traditional physics-based methods.

Contribution

The paper presents a novel coarse-to-fine model and strand-guided optimization for learning dynamic hair appearance and motion, enabling scalable, view-consistent rendering.

Findings

Achieves promising geometry and appearance results.

Supports integration into 3D Gaussian avatar frameworks.

Generalizes across diverse hairstyles and motions.

Abstract

The creation of photorealistic dynamic hair remains a major challenge in digital human modeling because of the complex motions, occlusions, and light scattering. Existing methods often resort to static capture and physics-based models that do not scale as they require manual parameter fine-tuning to handle the diversity of hairstyles and motions, and heavy computation to obtain high-quality appearance. In this paper, we present Dynamic Gaussian Hair (DGH), a novel framework that efficiently learns hair dynamics and appearance. We propose: (1) a coarse-to-fine model that learns temporally coherent hair motion dynamics across diverse hairstyles; (2) a strand-guided optimization module that learns a dynamic 3D Gaussian representation for hair appearance with support for differentiable rendering, enabling gradient-based learning of view-consistent appearance under motion. Unlike prior simulation-based pipelines, our approach is fully data-driven, scales with training data, and generalizes across various hairstyles and head motion sequences. Additionally, DGH can be seamlessly integrated into a 3D Gaussian avatar framework, enabling realistic, animatable hair for high-fidelity avatar representation. DGH achieves promising geometry and appearance results, providing a scalable, data-driven alternative to physics-based simulation and rendering.