Research map / 3D vision / Face and shape
ETCH-X: Robustify Expressive Body Fitting to Clothed Humans with Composable Synthetic Data
Xiaoben Li, Jingyi Wu, Zeyu CAI, YU Siyuan, Boqian Li, Yuliang Xiu
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 ETCH with ETCH-X (dense correspondence, SMPL-X) | ETCH · 4D-Dress | V2V (All): 2.408 -> 1.897 | Improved |
| replace ETCH with ETCH-X | ETCH · CAPE | V2V (Hands): 3.449 -> 2.215 | Improved |
| replace ETCH with ETCH-X | ETCH · BEDLAM2.0 (OOD) | V2V: 12.209 -> 3.429 | Improved |
| replace ETCH with ETCH-X | ETCH · CAPE | V2V (Head): 1.236 -> 1.120 | Improved |
| add partial-scan augmentation in training (ablation) | ETCH-X · 4D-Dress partial scans | V2V: 13.861 -> 3.808 | Improved |
| add partial-scan augmentation in training (ablation) | ETCH-X · 4D-Dress full scans | V2V: 1.897 -> 2.135 | Hurt |
| add partial-scan augmentation in training (ablation) | ETCH-X · CAPE partial scans | V2V: 10.056 -> 2.898 | Improved |
| add tightness masking (ablation) | ETCH-X · 4D-Dress | V2V: 4.256 -> 4.169 | Improved |
| add tightness masking (ablation) | ETCH-X · CAPE | V2V: 1.975 -> 1.894 | Improved |
| add hand LVD for hand refinement (ablation) | ETCH-X · CAPE | V2V (Hands): 2.607 -> 2.417 | Improved |
| add hand classifier to hand refinement (ablation) | ETCH-X · CAPE | V2V (Hands): 2.417 -> 2.278 | Improved |
| add hand data augmentation from invisible-vertex distribution (ablation) | ETCH-X · CAPE | V2V (Hands): 2.278 -> 2.215 | Improved |
Closest papers by meaning
- ETCH: Generalizing Body Fitting to Clothed Humans via Equivariant Tightness (ICCV 2025 · similarity 0.95)
- Monocular Expressive Body Regression through Body-Driven Attention (ECCV 2020 · similarity 0.89)
- SCALE: Modeling Clothed Humans with a Surface Codec of Articulated Local Elements (CVPR 2021 · similarity 0.88)
- Learning To Regress Bodies From Images Using Differentiable Semantic Rendering (ICCV 2021 · similarity 0.88)
- Cloth2Body: Generating 3D Human Body Mesh from 2D Clothing (ICCV 2023 · similarity 0.88)
- FreeCloth: Free-form Generation Enhances Challenging Clothed Human Modeling (CVPR 2025 · similarity 0.87)
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.
