Research map / 3D vision / Face and shape
Head Avatars with Dynamic Explicit Hair
Vanessa Sklyarova, Haonan Chen, Berna Kabadayi, Tobias Kirschstein, Zicong Fan, Xi Wang, Gerard Pons-Moll, Matthias Niessner, Marc Pollefeys, Michael Black, Justus Thies
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add L_elastic stretch regularization (ablation) | DynHair · Physical realism ablation (2 scenes, 3 motions) | VER: 283.75 -> 26.88 | Improved |
| add relative gravity conditioning (ablation) | DynHair · Physical realism ablation (2 scenes, 3 motions) | IoU hair: 0.852 -> 0.883 | Improved |
| add head acceleration conditioning (ablation) | DynHair · Physical realism ablation (2 scenes, 3 motions) | IoU hair: 0.878 -> 0.883 | Improved |
| add FiLM modulation of strand features (ablation) | DynHair · Physical realism ablation (2 scenes, 3 motions) | VER: 40.46 -> 26.88 | Improved |
| use LSTM instead of MLP head-pose encoder (ablation) | DynHair · Physical realism ablation (2 scenes, 3 motions) | tLPIPS_ex: -0.0010 -> 0.0024 | Improved |
| condition on absolute pose instead of velocity/acceleration (ablation) | DynHair · Physical realism ablation (2 scenes, 3 motions) | VER: 26.88 -> 36.27 | Hurt |
| add non-rigid deformation module (ablation) | DynHair · Cross-reenactment | qualitative | Improved |
| add strand-based dynamic hair vs GHA | DynHair · Self-reenactment (3 subjects) | FID (hair region): 81.07 -> 37.51 | Improved |
| add strand-based dynamic hair vs GHA | DynHair · Self-reenactment (3 subjects) | PSNR: 22.33 -> 21.60 | Hurt |
| learn hair dynamics vs Maya simulator | DynHair · Self-reenactment (3 subjects) | IoU hair: 0.776 -> 0.878 | Improved |
| add increased VGG loss weight late in training (ablation) | DynHair · Extended ablation (2 subjects) | PSNR (full image): 20.88 -> 21.21 | Improved |
| add color regularization (ablation) | DynHair · Extended ablation (2 subjects) | PSNR (full image): 20.95 -> 21.21 | Improved |
Closest papers by meaning
- DGH: Dynamic Gaussian Hair (NeurIPS 2025 · similarity 0.93)
- LUCAS: Layered Universal Codec Avatars (CVPR 2025 · similarity 0.90)
- MeGA: Hybrid Mesh-Gaussian Head Avatar for High-Fidelity Rendering and Head Editing (CVPR 2025 · similarity 0.89)
- Quaffure: Real-Time Quasi-Static Neural Hair Simulation (CVPR 2025 · similarity 0.88)
- SimAvatar: Simulation-Ready Avatars with Layered Hair and Clothing (CVPR 2025 · similarity 0.87)
- Perm: A Parametric Representation for Multi-Style 3D Hair Modeling (ICLR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
