Research map / 3D vision / Face and shape
SCALE: Modeling Clothed Humans with a Surface Codec of Articulated Local Elements
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 explicit articulation of local surface elements (ablation) | SCALE · CAPE blazerlong, shortlong, and skirt | Chamfer-L2: 1.92 -> 1.07; 1.43 -> 0.89; 2.71 -> 2.69 | Helped |
| Remove the UV-based global patch descriptor (ablation) | SCALE · CAPE blazerlong, shortlong, and skirt | Chamfer-L2: 1.07 -> 1.08; 0.89 -> 0.89; 2.69 -> 2.71 | No clear effect |
| Use a UNet instead of PointNet for pose embedding (ablation) | SCALE with global descriptor · CAPE blazerlong, shortlong, and skirt | Chamfer-L2: 1.46 -> 1.07; 1.03 -> 0.89; 2.72 -> 2.69 | Helped |
| Add the global patch descriptor when using PointNet pose embedding (ablation) | SCALE with PointNet pose embedding · CAPE blazerlong, shortlong, and skirt | Chamfer-L2: 1.92 -> 1.46; 1.42 -> 1.03; 3.06 -> 2.72 | Helped |
| Add joint surface-normal prediction (ablation) | SCALE · CAPE blazerlong, shortlong, and skirt | Chamfer-L2: 1.08 -> 1.07; 0.99 -> 0.89; 2.72 -> 2.69 | Helped |
| Use local pose-dependent shape codes instead of global shape codes (ablation) | SCALE versus global-code surface-element methods · CAPE blazerlong, shortlong, and skirt | Chamfer-L2: 1.43 -> 1.07; 1.09 -> 0.89; 3.07 -> 2.69 | Helped |
| Add explicit articulation to AtlasNet (ablation) | AtlasNet · CAPE blazerlong, shortlong, and skirt | Chamfer-L2: 6.20 -> 1.32; 1.54 -> 0.99; 4.59 -> 2.95 | Helped |
| Use surface elements rather than vertex offsets (ablation) | SCALE · blazerlong, shortlong, and skirt | Chamfer-L2: 1.13 -> 1.07; 0.91 -> 0.89; 2.78 -> 2.69 | Helped |
Closest papers by meaning
- TailorNet: Predicting Clothing in 3D as a Function of Human Pose, Shape and Garment Style (CVPR 2020 · similarity 0.92)
- VS: Reconstructing Clothed 3D Human from Single Image via Vertex Shift (CVPR 2024 · similarity 0.90)
- SCULPT: Shape-Conditioned Unpaired Learning of Pose-dependent Clothed and Textured Human Meshes (CVPR 2024 · similarity 0.90)
- FreeCloth: Free-form Generation Enhances Challenging Clothed Human Modeling (CVPR 2025 · similarity 0.90)
- MultiGO: Towards Multi-level Geometry Learning for Monocular 3D Textured Human Reconstruction (CVPR 2025 · similarity 0.89)
- Garment Recovery with Shape and Deformation Priors (CVPR 2024 · 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.