Research map / 3D vision / Human pose estimation
3D Hand Shape and Pose Estimation From a Single RGB Image
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 Graph CNN mesh reconstruction instead of MANO-based reconstruction | 3D hand mesh estimation · Our synthetic dataset validation | mesh error (mm): 12.12 -> 8.01 | Helped |
| use Graph CNN mesh reconstruction instead of direct LBS | 3D hand mesh estimation · Our real-world dataset | mesh error (mm): 13.33 -> 12.72 | Helped |
| add normal loss (ablation) | Full-loss mesh and pose estimation · Synthetic dataset validation | mesh error (mm): 8.34 -> 7.95 | Helped |
| add normal loss (ablation) | Full-loss mesh and pose estimation · Synthetic dataset validation | pose error (mm): 8.30 -> 8.03 | Helped |
| add edge loss (ablation) | Full-loss mesh and pose estimation · Synthetic dataset validation | mesh error (mm): 9.09 -> 7.95 | Helped |
| add edge loss (ablation) | Full-loss mesh and pose estimation · Synthetic dataset validation | pose error (mm): 9.06 -> 8.03 | Helped |
| add Laplacian loss (ablation) | Full-loss mesh and pose estimation · Synthetic dataset validation | mesh error (mm): 8.63 -> 7.95 | Helped |
| add Laplacian loss (ablation) | Full-loss mesh and pose estimation · Synthetic dataset validation | pose error (mm): 8.55 -> 8.03 | Helped |
| add 3D pose loss (ablation) | Full-loss mesh and pose estimation · Synthetic dataset validation | mesh error (mm): 9.04 -> 7.95 | Helped |
| add 3D pose loss (ablation) | Full-loss mesh and pose estimation · Synthetic dataset validation | pose error (mm): 9.24 -> 8.03 | Helped |
| fine-tune with depth-map loss (ablation) | 3D mesh-based hand pose estimation on STB · STB | 3D PCK | Helped |
| add pseudo-ground-truth mesh loss (ablation) | Weakly supervised mesh fine-tuning · Real-world training data | mesh quality | Helped |
Closest papers by meaning
- 3D Hand Shape and Pose From Images in the Wild (CVPR 2019 · similarity 0.93)
- 3D Convolutional Neural Networks for Efficient and Robust Hand Pose Estimation From Single Depth Images (CVPR 2017 · similarity 0.93)
- Pushing the Envelope for RGB-Based Dense 3D Hand Pose Estimation via Neural Rendering (CVPR 2019 · similarity 0.93)
- Model-Based 3D Hand Reconstruction via Self-Supervised Learning (CVPR 2021 · similarity 0.92)
- HandVoxNet: Deep Voxel-Based Network for 3D Hand Shape and Pose Estimation From a Single Depth Map (CVPR 2020 · similarity 0.92)
- Keypoint Fusion for RGB-D Based 3D Hand Pose Estimation (AAAI 2024 · similarity 0.90)
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.