Research map / 3D vision / 3D generation
DiGS-Avatar: Single-Image Animatable 3D Human Reconstruction via UV-Space Diffusion
jiakun li, Li Fang, Hao Zhu, Fei Hu, Long Ye, Yuan Zhang, Jinyao Yan
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 multi-view teacher supervision (ablation) | DiGS-Avatar · HuGe100K | PSNR: 19.03 -> 25.32 | Improved |
| add multi-view teacher supervision (ablation) | DiGS-Avatar · HuGe100K | LPIPS: 0.168 -> 0.039 | Improved |
| add GASA semantic injection (ablation) | DiGS-Avatar · HuGe100K | PSNR: 22.52 -> 25.32 | Improved |
| add GASA semantic injection (ablation) | DiGS-Avatar · HuGe100K | LPIPS: 0.084 -> 0.039 | Improved |
| add conditioning branch (ablation) | DiGS-Avatar · HuGe100K | PSNR: 23.26 -> 25.32 | Improved |
| add conditioning branch (ablation) | DiGS-Avatar · HuGe100K | LPIPS: 0.061 -> 0.039 | Improved |
| use adapted diffusion backbone instead of standard LDM-UNet (ablation) | DiGS-Avatar · HuGe100K | PSNR: 17.39 -> 25.32 | Improved |
| use adapted diffusion backbone instead of standard LDM-UNet (ablation) | DiGS-Avatar · HuGe100K | LPIPS: 0.204 -> 0.039 | Improved |
| add denoising loss L_diff (ablation) | DiGS-Avatar · HuGe100K | PSNR: 15.98 -> 25.32 | Improved |
| add denoising loss L_diff (ablation) | DiGS-Avatar · HuGe100K | LPIPS: 0.226 -> 0.039 | Improved |
| single-view student instead of multi-view teacher | Teacher vs student · HuGe100K | PSNR: 26.18 -> 25.32 | Mixed |
| DiGS-Avatar instead of LHM zero-shot | LHM · SIZER (zero-shot) | PSNR: 23.45 -> 23.54 | 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.
- IDOL — IDOL: Inertial Deep Orientation-Estimation and Localization (AAAI 2021) · better on 6 of 6 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
- Human-3Diffusion: Realistic Avatar Creation via Explicit 3D Consistent Diffusion Models (NeurIPS 2024 · similarity 0.91)
- DiffHuman: Probabilistic Photorealistic 3D Reconstruction of Humans (CVPR 2024 · similarity 0.89)
- CAP4D: Creating Animatable 4D Portrait Avatars with Morphable Multi-View Diffusion Models (CVPR 2025 · similarity 0.89)
- Morphable Diffusion: 3D-Consistent Diffusion for Single-image Avatar Creation (CVPR 2024 · similarity 0.88)
- 3DGS-Avatar: Animatable Avatars via Deformable 3D Gaussian Splatting (CVPR 2024 · similarity 0.88)
- AniGS: Animatable Gaussian Avatar from a Single Image with Inconsistent Gaussian Reconstruction (CVPR 2025 · 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.
