Research map / 3D vision / Face and shape
Get3DHuman: Lifting StyleGAN-Human into a 3D Generative Model Using Pixel-Aligned Reconstruction Priors
Zhangyang Xiong, Di Kang, Derong Jin, Weikai Chen, Linchao Bao, Shuguang Cui, Xiaoguang Han
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 prior losses to adversarial-only training (ablation) | Get3DHuman · Pseudo-GT human eval set | FID: 89.39 -> 59.95 | Improved |
| add adversarial loss to prior-only training (ablation) | Get3DHuman · Pseudo-GT human eval set | COV (%): 31.51 -> 39.22 | Improved |
| add refinement module (ablation) | Get3DHuman · Pseudo-GT human eval set | FID: 59.95 -> 54.39 | Improved |
| add refinement module (ablation) | Get3DHuman · Pseudo-GT human eval set | FID_3D: 74.38 -> 69.70 | Improved |
| add prior losses to adversarial-only training (ablation) | Get3DHuman · Pseudo-GT human eval set | FPD: 0.63 -> 0.85 | Hurt |
| add normal-map supervision on intermediate features (ablation) | Get3DHuman (prior-only) · qualitative | geometry detail | Improved |
| use 2D-generator and reconstruction priors instead of differentiable-rendering supervision (Get3DHuman vs GET3D) | 3D human generator · Pseudo-GT human eval set | FID_3D: 88.15 -> 69.70 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- GET3D — GET3D: A Generative Model of High Quality 3D Textured Shapes Learned from Images (NeurIPS 2022) · better on 5 of 5 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
- SyncHuman: Synchronizing 2D and 3D Generative Models for Single-view Human Reconstruction (NeurIPS 2025 · similarity 0.91)
- En3D: An Enhanced Generative Model for Sculpting 3D Humans from 2D Synthetic Data (CVPR 2024 · similarity 0.91)
- SiTH: Single-view Textured Human Reconstruction with Image-Conditioned Diffusion (CVPR 2024 · similarity 0.89)
- PSHuman: Photorealistic Single-image 3D Human Reconstruction using Cross-Scale Multiview Diffusion and Explicit Remeshing (CVPR 2025 · similarity 0.89)
- GeneMAN: Generalizable Single-Image 3D Human Reconstruction from Multi-Source Human Data (NeurIPS 2025 · similarity 0.89)
- DiffHuman: Probabilistic Photorealistic 3D Reconstruction of Humans (CVPR 2024 · similarity 0.89)
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.
