Research map / Image generation / Text-to-image diffusion
HeadSculpt: Crafting 3D Head Avatars with Text
Xiao Han, Yukang Cao, Kai Han, Xiatian Zhu, Jiankang Deng, Yi-Zhe Song, Tao Xiang, Kwan-Yee K. Wong
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 |
|---|---|---|---|
| remove landmark-based ControlNet control (ablation) | HeadSculpt · coarse-stage head generation (qualitative) | view consistency | Hurt |
| remove learned back-view textual inversion embedding (ablation) | HeadSculpt · coarse-stage head generation (qualitative) | Janus problem on back view | Hurt |
| use vanilla InstructPix2Pix instead of ControlNet-based one for editing (B4) (ablation) | HeadSculpt editing · 3D head editing (qualitative) | view consistency and identity preservation | Hurt |
| add edit scale blending of two scores (IESD) over omega_e=1 (B5) (ablation) | HeadSculpt editing · 3D head editing (CLIP-DS, 10 examples) | identity preservation and edit fidelity | Improved |
| edit by re-optimizing with altered text description (B1-B3) instead of IESD | HeadSculpt editing · 3D head editing (qualitative) | editing strength and identity | Hurt |
| use HeadSculpt instead of Latent-NeRF, 3DFuse, Fantasia3D, DreamFace | text-to-3D head generation · user study, 42 volunteers | rank for text consistency, texture and geometry | Improved |
| use HeadSculpt instead of competing text-to-3D methods | text-to-3D head generation · 30 text prompts | CLIP-R and CLIP-S | Improved |
Closest papers by meaning
- AvatarVerse: High-Quality & Stable 3D Avatar Creation from Text and Pose (AAAI 2024 · similarity 0.89)
- GeneAvatar: Generic Expression-Aware Volumetric Head Avatar Editing from a Single Image (CVPR 2024 · similarity 0.89)
- Gaussian Head Avatar: Ultra High-fidelity Head Avatar via Dynamic Gaussians (CVPR 2024 · similarity 0.89)
- DiffPortrait360: Consistent Portrait Diffusion for 360 View Synthesis (CVPR 2025 · similarity 0.89)
- MeGA: Hybrid Mesh-Gaussian Head Avatar for High-Fidelity Rendering and Head Editing (CVPR 2025 · similarity 0.88)
- DiffusionGAN3D: Boosting Text-guided 3D Generation and Domain Adaptation by Combining 3D GANs and Diffusion Priors (CVPR 2024 · similarity 0.88)
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.
