Research map / 3D vision / Face and shape
COSMU: Complete 3D human shape from monocular unconstrained images
Marco Pesavento, Marco Volino, Adrian Hilton
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 |
|---|---|---|---|
| select references with the full energy (orientation, visibility, pose similarity) instead of random (ablation) | COSMU · THuman3.0 | Chamfer distance: 1.485 -> 1.304 | Improved |
| add body part-based registration of references (ablation) | COSMU · THuman3.0 | Chamfer distance: 3.546 -> 1.304 | Improved |
| use rigid per-part registration instead of non-rigid full-body registration (ablation) | COSMU · THuman3.0 | Chamfer distance: 6.932 -> 1.304 | Improved |
| select N_p references per body part instead of a single reference (ablation) | COSMU · THuman3.0 | Chamfer distance: 1.314 -> 1.304 | Improved |
| feed both RGB and segmentation masks to part classifier (ablation) | COSMU body part classifier · THuman3.0 | Chamfer distance: 2.249 -> 1.304 | Improved |
| use 4 views (3 references) instead of 2 views | COSMU · THuman3.0 | Chamfer distance: 1.355 -> 1.304 | Improved |
| leverage real reference views instead of estimating unseen views from the target | COSMU-2 · THuman3.0 | Chamfer distance: COSMU-2 1.355 vs ECON 1.368 CD; P2S 1.373 vs 1.365 | Mixed |
Closest papers by meaning
- Moulding Humans: Non-Parametric 3D Human Shape Estimation From Single Images (ICCV 2019 · similarity 0.90)
- Recovering 3D Human Mesh From Monocular Images: A Survey (TPAMI 2023 · similarity 0.89)
- MultiGO: Towards Multi-level Geometry Learning for Monocular 3D Textured Human Reconstruction (CVPR 2025 · similarity 0.88)
- MUC: Mixture of Uncalibrated Cameras for Robust 3D Human Body Reconstruction (AAAI 2025 · similarity 0.88)
- PSHuman: Photorealistic Single-image 3D Human Reconstruction using Cross-Scale Multiview Diffusion and Explicit Remeshing (CVPR 2025 · similarity 0.87)
- ODHSR: Online Dense 3D Reconstruction of Humans and Scenes from Monocular Videos (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.
