Research map / 3D vision / Face and shape
HUMBI: A Large Multiview Dataset of Human Body Expressions
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 HUMBI training data to 3DDFA | 3D face mesh prediction · 3DDFA test | reprojection error (pixels): 7.1±6.4 -> 4.3±6.6 | Helped |
| Add HUMBI training data to 3DDFA | 3D face mesh prediction · HUMBI Face test | reprojection error (pixels): 13.3±13.7 -> 8.4±12.2 | Helped |
| Add HUMBI training data to ObMan | 3D hand mesh prediction · ObMan test | reprojection error (pixels): 3.84±2.6 -> 3.5±2.4 | Helped |
| Add HUMBI training data to ObMan | 3D hand mesh prediction · HUMBI Hand test | reprojection error (pixels): 6.5±8.4 -> 4.8±5.8 | Helped |
| Add HUMBI training data to UP-3D | 3D body mesh prediction · UP-3D test | reprojection error (pixels): 22.7±18.6 -> 18.4±13.8 | Helped |
| Add HUMBI training data to UP-3D | 3D body mesh prediction · HUMBI Body test | reprojection error (pixels): 14.5±6.6 -> 12.5±8.4 | Helped |
| Use HUMBI alone rather than Human3.6M training data | 3D body pose prediction · Human3.6M test | AUC: 0.562 -> 0.434 | Hurt |
| Use HUMBI alone rather than MPI-INF-3DHP training data | 3D body pose prediction · MPI-INF-3DHP test | AUC: 0.377 -> 0.354 | Hurt |
| Add HUMBI training data to Human3.6M | 3D body pose prediction · Human3.6M test | AUC: 0.562 -> 0.551 | Hurt |
| Add HUMBI training data to MPI-INF-3DHP | 3D body pose prediction · MPI-INF-3DHP test | AUC: 0.377 -> 0.425 | Helped |
| Train with HUMBI in addition to either existing dataset | 3D gaze prediction · MPII-Gaze and UT-Multiview tests | mean gaze error (degrees): MPII test: 6.1±3.3 -> 7.4±4.1; UTMV test: 5.0±3.2 -> 5.4±3.2 | Mixed |
| Increase camera count toward the 107-camera setup (ablation) | garment reconstruction · HUMBI garment | reconstruction accuracy and density | No clear effect |
Closest papers by meaning
- MUC: Mixture of Uncalibrated Cameras for Robust 3D Human Body Reconstruction (AAAI 2025 · similarity 0.86)
- MVHumanNet: A Large-scale Dataset of Multi-view Daily Dressing Human Captures (CVPR 2024 · similarity 0.85)
- Expressive Gaussian Human Avatars from Monocular RGB Video (NeurIPS 2024 · similarity 0.85)
- Visual Persona: Foundation Model for Full-Body Human Customization (CVPR 2025 · similarity 0.84)
- Codec Avatar Studio: Paired Human Captures for Complete, Driveable, and Generalizable Avatars (NeurIPS 2024 · similarity 0.84)
- FRESA: Feedforward Reconstruction of Personalized Skinned Avatars from Few Images (CVPR 2025 · similarity 0.84)
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.