Research map / 3D vision / Face and shape
HairNeRF: Geometry-Aware Image Synthesis for Hairstyle Transfer
Seunggyu Chang, Gihoon Kim, Hayeon Kim
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 head registration (alignment via deformed rays) to HairNeRF (ablation) | HairNeRF · VoxCeleb2 reconstruction | MSE: 0.980 -> 0.957 | Improved |
| add head registration to HairNeRF (ablation) | HairNeRF · VoxCeleb2 reconstruction | PSNR: 31.77 -> 31.84 | Improved |
| use StyleNeRF-based HairNeRF instead of StyleGAN2-based Barbershop for reconstruction | hairstyle transfer · VoxCeleb2 reconstruction | LPIPS: 0.127 -> 0.106 | Improved |
| use NeRF-based generator (StyleNeRF) instead of StyleGAN2 for hairstyle transfer | hairstyle transfer · FFHQ hairstyle transfer | FID: 48.00 -> 57.26 (vs Barbershop) | Hurt |
| use NeRF-based HairNeRF instead of StyleGAN2-based baselines | hairstyle transfer · FFHQ hairstyle transfer | NIQE: 12.00 -> 9.57 (vs StyleYourHair) | Improved |
| use HairNeRF instead of Barbershop | hairstyle transfer · FFHQ hairstyle transfer | ArcFace identity similarity: 0.77 -> 0.70 | Hurt |
| use HairNeRF instead of StyleYourHair | hairstyle transfer · FFHQ hairstyle transfer | ArcFace identity similarity: 0.55 -> 0.70 | Improved |
| use HairNeRF instead of Barbershop | hairstyle transfer · FFHQ hairstyle transfer | Realism: 1.06 -> 1.40 | Improved |
| use HairNeRF instead of Barbershop | hairstyle transfer · UF (Unsplash) hairstyle transfer | Realism: 1.91 -> 1.71 | Hurt |
Closest papers by meaning
- StyleGAN Salon: Multi-View Latent Optimization for Pose-Invariant Hairstyle Transfer (CVPR 2023 · similarity 0.92)
- HairNet: Hairstyle Transfer with Pose Changes (ECCV 2022 · similarity 0.90)
- HairFastGAN: Realistic and Robust Hair Transfer with a Fast Encoder-Based Approach (NeurIPS 2024 · similarity 0.90)
- What to Preserve and What to Transfer: Faithful, Identity-Preserving Diffusion-based Hairstyle Transfer (AAAI 2025 · similarity 0.89)
- Stable-Hair: Real-World Hair Transfer via Diffusion Model (AAAI 2025 · similarity 0.88)
- Text-Conditioned Generative Model of 3D Strand-based Human Hairstyles (CVPR 2024 · 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.
