Research map / 3D vision / 3D generation
High-Quality Full-Head 3D Avatar Generation from Any Single Portrait Image
Yujie Gao, ChenCheng Wang, Xianbing Sun, Jiahui Zhan, Wentao Wang, Yiyi Zhang, Haohua Zhao, Liqing Zhang, Jianfu Zhang
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 ID guidance in low-resolution stage (ablation) | ID-TS diffusion (256x256) · Digital Human ablation | PSNR: 18.6968 -> 18.9627 | Improved |
| add ID guidance in low-resolution stage (ablation) | ID-TS diffusion (256x256) · Digital Human ablation | CSIM: 0.8860 -> 0.8961 | Improved |
| add ID guidance in low-resolution stage (ablation) | ID-TS diffusion (256x256) · Digital Human ablation | DreamSim: 0.1217 -> 0.1106 | Improved |
| add ID guidance to fine-tuned 512x512 model (ablation) | ID-TS diffusion · Digital Human ablation | PSNR: 17.5717 -> 18.2606 | Improved |
| add ID guidance to fine-tuned 512x512 model (ablation) | ID-TS diffusion · Digital Human ablation | CSIM: 0.8799 -> 0.9039 | Improved |
| add two-stage diffusion to fine-tuned model (ablation) | ID-TS diffusion · Digital Human ablation | PSNR: 17.5717 -> 19.1919 | Improved |
| add two-stage diffusion to fine-tuned model (ablation) | ID-TS diffusion · Digital Human ablation | LPIPS: 0.2446 -> 0.2092 | Improved |
| add two-stage diffusion on top of ID guidance (ablation) | ID-TS diffusion · Digital Human ablation | PSNR: 18.2606 -> 19.9974 | Improved |
| add ID guidance on top of two-stage diffusion (ablation) | ID-TS diffusion · Digital Human ablation | CSIM: 0.9164 -> 0.9261 | Improved |
| add fine-tuning on digital human head dataset (ablation) | Gaussian head generation · Digital Human ablation | PSNR: 15.6821 -> 18.3844 | Improved |
| add per-subject optimization (ablation) | Gaussian head generation · Digital Human ablation | PSNR: 18.3844 -> 20.4287 | Improved |
| add per-subject optimization (ablation) | Gaussian head generation · Digital Human ablation | LPIPS: 0.2556 -> 0.2315 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- ID-Sculpt — ID-Sculpt: ID-aware 3D Head Generation from Single In-the-wild Portrait Image (AAAI 2025) · better on 6 of 6 shared results
- PanoHead — PanoHead: Geometry-Aware 3D Full-Head Synthesis in 360deg (CVPR 2023) · better on 6 of 6 shared results
- GPAvatar — GPAvatar: High-fidelity Head Avatars by Learning Efficient Gaussian Projections (CVPR 2025) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- GAF: Gaussian Avatar Reconstruction from Monocular Videos via Multi-view Diffusion (CVPR 2025 · similarity 0.92)
- Human-3Diffusion: Realistic Avatar Creation via Explicit 3D Consistent Diffusion Models (NeurIPS 2024 · similarity 0.91)
- Morphable Diffusion: 3D-Consistent Diffusion for Single-image Avatar Creation (CVPR 2024 · similarity 0.91)
- Arc2Avatar: Generating Expressive 3D Avatars from a Single Image via ID Guidance (CVPR 2025 · similarity 0.90)
- CAP4D: Creating Animatable 4D Portrait Avatars with Morphable Multi-View Diffusion Models (CVPR 2025 · similarity 0.90)
- DiffusionAvatars: Deferred Diffusion for High-fidelity 3D Head Avatars (CVPR 2024 · similarity 0.90)
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.
