Research map / 3D vision / Face and shape
HUMBI: A Large Multiview Dataset of Human Body Expressions and Benchmark Challenge
Jae Shin Yoon, Zhixuan Yu, Jaesik Park, Hyun Soo Park
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 HUMBI to MPII training data | gaze network · 3D gaze, test on MPII | mean error (degrees): 6.1 -> 7.4 | Hurt |
| add HUMBI to UTMV training data | gaze network · 3D gaze, test on HUMBI | mean error (degrees): 14.6 -> 8.0 | Improved |
| train on HUMBI instead of MPII | gaze network · 3D gaze, test on UTMV | mean error (degrees): 23.3 -> 8.2 | Improved |
| add HUMBI to 3DDFA training data | face mesh network · 3D face mesh, test on 3DDFA | mean error (pixels): 7.1 -> 4.3 | Improved |
| add HUMBI to 3DDFA training data | face mesh network · 3D face mesh, test on HUMBI | mean error (pixels): 13.3 -> 8.4 | Improved |
| add HUMBI Hand to RHD training data | hand pose detector · 3D hand keypoints, average | AUC of PCK: 0.44 -> 0.48 | Improved |
| add HUMBI Hand to ObMan training data | hand mesh network · 3D hand mesh, test on HUMBI | mean error (pixels): 10.6 -> 4.8 | Improved |
| add HUMBI Body to H36M training data | body pose detector · 3D body keypoints, average | AUC of PCK: 0.376 -> 0.433 | Improved |
| add HUMBI Body to UP-3D training data | body mesh network · 3D body mesh, test on UP-3D | mean error (pixels): 22.7 -> 18.4 | Improved |
| train on HUMBI alone instead of UP-3D | body mesh network · 3D body mesh, test on UP-3D | mean error (pixels): 22.7 -> 49.4 | Hurt |
| train on HUMBI Hand instead of SHPTB | hand pose detector · 3D hand keypoints, average | AUC of PCK: 0.30 -> 0.46 | Improved |
Closest papers by meaning
- HUMBI: A Large Multiview Dataset of Human Body Expressions (CVPR 2020 · similarity 0.98)
- MUC: Mixture of Uncalibrated Cameras for Robust 3D Human Body Reconstruction (AAAI 2025 · similarity 0.86)
- Expressive Gaussian Human Avatars from Monocular RGB Video (NeurIPS 2024 · similarity 0.86)
- MVHumanNet: A Large-scale Dataset of Multi-view Daily Dressing Human Captures (CVPR 2024 · similarity 0.86)
- Visual Persona: Foundation Model for Full-Body Human Customization (CVPR 2025 · similarity 0.86)
- Reconstructing Close Human Interaction with Appearance and Proxemics Reasoning (CVPR 2025 · similarity 0.85)
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.
