ControlHair: Physically-based Video Diffusion for Controllable Dynamic Hair Rendering

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lin, Weikai, Li, Haoxiang, Zhu, Yuhao
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914063850668032
author Lin, Weikai
Li, Haoxiang
Zhu, Yuhao
author_facet Lin, Weikai
Li, Haoxiang
Zhu, Yuhao
contents Hair simulation and rendering are challenging due to complex strand dynamics, diverse material properties, and intricate light-hair interactions. Recent video diffusion models can generate high-quality videos, but they lack fine-grained control over hair dynamics. We present ControlHair, a hybrid framework that integrates a physics simulator with conditional video diffusion to enable controllable dynamic hair rendering. ControlHair adopts a three-stage pipeline: it first encodes physics parameters (e.g., hair stiffness, wind) into per-frame geometry using a simulator, then extracts per-frame control signals, and finally feeds control signals into a video diffusion model to generate videos with desired hair dynamics. This cascaded design decouples physics reasoning from video generation, supports diverse physics, and makes training the video diffusion model easy. Trained on a curated 10K video dataset, ControlHair outperforms text- and pose-conditioned baselines, delivering precisely controlled hair dynamics. We further demonstrate three use cases of ControlHair: dynamic hairstyle try-on, bullet-time effects, and cinemagraphic. ControlHair introduces the first physics-informed video diffusion framework for controllable dynamics. We provide a teaser video and experimental results on our website.
format Preprint
id arxiv_https___arxiv_org_abs_2509_21541
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ControlHair: Physically-based Video Diffusion for Controllable Dynamic Hair Rendering
Lin, Weikai
Li, Haoxiang
Zhu, Yuhao
Graphics
Computer Vision and Pattern Recognition
I.3; I.2; I.4
Hair simulation and rendering are challenging due to complex strand dynamics, diverse material properties, and intricate light-hair interactions. Recent video diffusion models can generate high-quality videos, but they lack fine-grained control over hair dynamics. We present ControlHair, a hybrid framework that integrates a physics simulator with conditional video diffusion to enable controllable dynamic hair rendering. ControlHair adopts a three-stage pipeline: it first encodes physics parameters (e.g., hair stiffness, wind) into per-frame geometry using a simulator, then extracts per-frame control signals, and finally feeds control signals into a video diffusion model to generate videos with desired hair dynamics. This cascaded design decouples physics reasoning from video generation, supports diverse physics, and makes training the video diffusion model easy. Trained on a curated 10K video dataset, ControlHair outperforms text- and pose-conditioned baselines, delivering precisely controlled hair dynamics. We further demonstrate three use cases of ControlHair: dynamic hairstyle try-on, bullet-time effects, and cinemagraphic. ControlHair introduces the first physics-informed video diffusion framework for controllable dynamics. We provide a teaser video and experimental results on our website.
title ControlHair: Physically-based Video Diffusion for Controllable Dynamic Hair Rendering
topic Graphics
Computer Vision and Pattern Recognition
I.3; I.2; I.4
url https://arxiv.org/abs/2509.21541