Research map / 3D vision / Face and shape
ATLAS: Decoupling Skeletal and Shape Parameters for Expressive Parametric Human Modeling
Jinhyung Park, Javier Romero, Shunsuke Saito, Fabian Prada, Takaaki Shiratori, Yichen Xu, Federica Bogo, Shoou-I Yu, Kris Kitani, Rawal Khirodkar
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 skeleton parameters to shape-only fitting (ablation) | ATLAS · 3DBodyTex | fitting error (mm): 6.47 -> 2.48 | Improved |
| add shape parameters to skeleton-only fitting (ablation) | ATLAS · 3DBodyTex | fitting error (mm): 3.17 -> 2.48 | Improved |
| add skeleton parameters to shape-only fitting (ablation) | ATLAS · Goliath-Test | fitting error (mm): 4.76 -> 2.34 | Improved |
| add shape parameters to skeleton-only fitting (ablation) | ATLAS · Goliath-Test | fitting error (mm): 2.67 -> 2.34 | Improved |
| use non-linear instead of linear pose correctives (ablation) | ATLAS · SMPL pose dataset | fitting error (mm): 1.82 -> 1.61 | Improved |
| add relative depth term to single-image fitting (ablation) | ATLAS fitting · Goliath-Test single-image | vertex error (mm): 60.7 -> 55.4 | Improved |
| add relative depth term to single-image fitting (ablation) | ATLAS fitting · Goliath-Test single-image | joint error (mm): 54.5 -> 53.7 | Improved |
| add mask term to fitting without depth (ablation) | ATLAS fitting · Goliath-Test single-image | vertex error (mm): 61.8 -> 60.7 | Improved |
| add mask term to fitting without depth (ablation) | ATLAS fitting · Goliath-Test single-image | joint error (mm): 55.7 -> 54.5 | Improved |
| use ATLAS decoupled model instead of SMPL-X | body model fitting · Goliath-Test | vertex-to-vertex error (mm): 2.78 -> 2.34 | Improved |
| use ATLAS instead of SMPL-X at 32 components | body model fitting · 3DBodyTex | vertex-to-vertex error: 21.6% lower vertex-to-vertex error | Improved |
| use ATLAS fitting pipeline instead of SMPLify-X | single-image mesh fitting · Goliath-Test | vertex error (mm): 87.7 -> 55.4 | Improved |
Closest papers by meaning
- BLSM: A Bone-Level Skinned Model of the Human Mesh (ECCV 2020 · similarity 0.91)
- Human Mesh Modeling for Anny Body (ECCV 2026 · similarity 0.89)
- SAM 3D Body: Robust Full-Body Human Mesh Recovery (CVPR 2026 · similarity 0.89)
- Accurate 3D Body Shape Regression Using Metric and Semantic Attributes (CVPR 2022 · similarity 0.89)
- Reconstructing Humans with a Biomechanically Accurate Skeleton (CVPR 2025 · similarity 0.88)
- MultiGO: Towards Multi-level Geometry Learning for Monocular 3D Textured Human Reconstruction (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.
