Research map / 3D vision / Human pose estimation
Weakly-Supervised Mesh-Convolutional Hand Reconstruction in the Wild
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 |
|---|---|---|---|
| Use spatial Spiral rather than Spectral mesh decoding (ablation) | Mesh reconstruction network · YouTube | Pose Error [mm]: 13.612 -> 10.698 | Helped |
| Use spatial Spiral rather than Spectral mesh decoding (ablation) | Mesh reconstruction network · YouTube | Mesh Error [mm]: 19.317 -> 17.096 | Helped |
| Train end-to-end with mesh loss in image coordinates instead of pretrained canonical-frame decoding and camera estimation (ablation) | Mesh reconstruction network · YouTube | Pose Error [mm]: 11.799 -> 10.698 | Helped |
| Add YouTube training data (ablation) | Network trained on MPII+RHD+MTC · YouTube | Pose Error [mm]: 16.214 -> 10.698 | Helped |
| Add YouTube training data to MPII+RHD+MTC+FreiHAND (ablation) | Network trained on MPII+RHD+MTC+FreiHAND · YouTube | Pose Error [mm]: 12.637 -> 10.536 | Helped |
| Train with MPII+RHD instead of adding MTC and YouTube (ablation) | Hand reconstruction network · MPII | Pose Error [px]: 11.547 -> 9.434 | Hurt |
| Add an explicit pose estimation loss (ablation) | Mesh reconstruction network · Training | Pose estimation performance | No clear effect |
| Add edge-length preservation loss (ablation) | MANO decoder · YouTube | Mesh shape stability | Helped |
| Use the network rather than iterative fitting | Hand reconstruction system · MPII | Pose Error [px]: 10.295 -> 9.265 | Helped |
| Use the network rather than iterative fitting | Hand reconstruction system · MPII Filtered | Pose Error [px]: 5.111 -> 5.555 | Hurt |
Closest papers by meaning
- 3D Hand Shape and Pose From Images in the Wild (CVPR 2019 · similarity 0.91)
- Weakly-Supervised Domain Adaptation via GAN and Mesh Model for Estimating 3D Hand Poses Interacting Objects (CVPR 2020 · similarity 0.89)
- Reconstructing Hands in 3D with Transformers (CVPR 2024 · similarity 0.89)
- HandBooster: Boosting 3D Hand-Mesh Reconstruction by Conditional Synthesis and Sampling of Hand-Object Interactions (CVPR 2024 · similarity 0.89)
- HOLD: Category-agnostic 3D Reconstruction of Interacting Hands and Objects from Video (CVPR 2024 · similarity 0.88)
- SiMHand: Mining Similar Hands for Large-Scale 3D Hand Pose Pre-training (ICLR 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.