Research map / 3D vision / Face and shape
Global-correlated 3D-decoupling Transformer for Clothed Avatar Reconstruction
Zechuan Zhang, Li Sun, Zongxin Yang, Ling Chen, Yi Yang
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 |
|---|---|---|---|
| replace transformer with UNet convolution (ablation) | GTA · CAPE | Chamfer: 0.861 -> 0.991 | Hurt |
| remove cross-attention and learnable embeddings (ablation) | GTA · CAPE | Chamfer: 0.861 -> 0.937 | Hurt |
| remove refinement module (ablation) | GTA · CAPE | Chamfer: 0.861 -> 0.890 | Hurt |
| use 2D features with spatial query instead of 3D (ablation) | GTA · CAPE | Chamfer: 0.861 -> 1.054 | Hurt |
| use 2D features with prior-enhanced query (ablation) | GTA · CAPE | Chamfer: 0.861 -> 1.133 | Hurt |
| use 2D features with hybrid fusion (ablation) | GTA · CAPE | Chamfer: 0.861 -> 1.008 | Hurt |
| 3D features with spatial query only (no hybrid prior fusion) (ablation) | GTA · CAPE | Chamfer: 0.861 -> 0.987 | Hurt |
| 3D features with prior-enhanced query only (ablation) | GTA · CAPE | Chamfer: 0.861 -> 1.059 | Hurt |
| spatial query vs prior-enhanced query (ablation) | GTA · CAPE | Chamfer: 1.059 -> 0.987 | Improved |
| GTA vs ECON on THuman2.0 | GTA · THuman2.0 | Chamfer: 2.120 -> 0.814 | Improved |
| GTA vs ECON on CAPE normals | GTA · CAPE-NFP | Normals: 0.035 -> 0.042 | Hurt |
| GTA vs ICON on CAPE-FP | GTA · CAPE-FP | Chamfer: 0.969 -> 0.763 | 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.
- ICON — ICON: Implicit Clothed Humans Obtained From Normals (CVPR 2022) · better on 9 of 9 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
- ARCH++: Animation-Ready Clothed Human Reconstruction Revisited (ICCV 2021 · similarity 0.89)
- SIFU: Side-view Conditioned Implicit Function for Real-world Usable Clothed Human Reconstruction (CVPR 2024 · similarity 0.87)
- HiLo: Detailed and Robust 3D Clothed Human Reconstruction with High-and Low-Frequency Information of Parametric Models (CVPR 2024 · similarity 0.87)
- FRESA: Feedforward Reconstruction of Personalized Skinned Avatars from Few Images (CVPR 2025 · similarity 0.87)
- 3DGS-Avatar: Animatable Avatars via Deformable 3D Gaussian Splatting (CVPR 2024 · similarity 0.86)
- VS: Reconstructing Clothed 3D Human from Single Image via Vertex Shift (CVPR 2024 · similarity 0.86)
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.
