HairWeaver: Few-Shot Photorealistic Hair Motion Synthesis with Sim-to-Real Guided Video Diffusion
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912898030239744 |
|---|---|
| author | Chang, Di Hou, Ji Bozic, Aljaz Neuberger, Assaf Juefei-Xu, Felix Maury, Olivier Lin, Gene Wei-Chin Stuyck, Tuur Roble, Doug Soleymani, Mohammad Grabli, Stephane |
| author_facet | Chang, Di Hou, Ji Bozic, Aljaz Neuberger, Assaf Juefei-Xu, Felix Maury, Olivier Lin, Gene Wei-Chin Stuyck, Tuur Roble, Doug Soleymani, Mohammad Grabli, Stephane |
| contents | We present HairWeaver, a diffusion-based pipeline that animates a single human image with realistic and expressive hair dynamics. While existing methods successfully control body pose, they lack specific control over hair, and as a result, fail to capture the intricate hair motions, resulting in stiff and unrealistic animations. HairWeaver overcomes this limitation using two specialized modules: a Motion-Context-LoRA to integrate motion conditions and a Sim2Real-Domain-LoRA to preserve the subject's photoreal appearance across different data domains. These lightweight components are designed to guide a video diffusion backbone while maintaining its core generative capabilities. By training on a specialized dataset of dynamic human motion generated from a CG simulator, HairWeaver affords fine control over hair motion and ultimately learns to produce highly realistic hair that responds naturally to movement. Comprehensive evaluations demonstrate that our approach sets a new state of the art, producing lifelike human hair animations with dynamic details. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_11117 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | HairWeaver: Few-Shot Photorealistic Hair Motion Synthesis with Sim-to-Real Guided Video Diffusion Chang, Di Hou, Ji Bozic, Aljaz Neuberger, Assaf Juefei-Xu, Felix Maury, Olivier Lin, Gene Wei-Chin Stuyck, Tuur Roble, Doug Soleymani, Mohammad Grabli, Stephane Computer Vision and Pattern Recognition We present HairWeaver, a diffusion-based pipeline that animates a single human image with realistic and expressive hair dynamics. While existing methods successfully control body pose, they lack specific control over hair, and as a result, fail to capture the intricate hair motions, resulting in stiff and unrealistic animations. HairWeaver overcomes this limitation using two specialized modules: a Motion-Context-LoRA to integrate motion conditions and a Sim2Real-Domain-LoRA to preserve the subject's photoreal appearance across different data domains. These lightweight components are designed to guide a video diffusion backbone while maintaining its core generative capabilities. By training on a specialized dataset of dynamic human motion generated from a CG simulator, HairWeaver affords fine control over hair motion and ultimately learns to produce highly realistic hair that responds naturally to movement. Comprehensive evaluations demonstrate that our approach sets a new state of the art, producing lifelike human hair animations with dynamic details. |
| title | HairWeaver: Few-Shot Photorealistic Hair Motion Synthesis with Sim-to-Real Guided Video Diffusion |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2602.11117 |