Research map / 3D vision / Human pose estimation
HandMap: Robust Hand Pose Estimation via Intermediate Dense Guidance Map Supervision
Xiaokun Wu, Daniel Finnegan, Eamonn O'Neill, Yong-Liang Yang
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 weighted approximate geodesic distance guidance map supervision to 2D coordinate regression | 2D CR · HANDS 2017 | mean error (mm): 19.63 -> 6.68 | Improved |
| add weighted surface distance supervision to 2D coordinate regression | 2D CR · HANDS 2017 | max-per-joint error (mm): 32.98 -> 8.73 | Improved |
| swap Inception-ResNet-v2 into 2D CR baseline (ablation) | 2D CR · HANDS 2017 | mean error (mm): 19.63 -> 17.94 | Improved |
| add heatmap guidance supervision to 2D CR (ablation) | 2D CR · HANDS 2017 | mean error (mm): 19.63 -> 7.62 | Improved |
| add plain Euclidean distance map supervision to 2D CR (ablation) | 2D CR · HANDS 2017 | mean error (mm): 19.63 -> 10.90 | Improved |
| use normalized offset map instead of plain Euclidean distance map (ablation) | 2D CR · HANDS 2017 | mean error (mm): 10.90 -> 9.34 | Improved |
| weight the surface distance map (ablation) | 2D CR · HANDS 2017 | mean error (mm): 7.89 -> 6.68 | Improved |
| add offset-map supervision to 3D coordinate regression (ablation) | 3D CR · HANDS 2017 | mean error (mm): 11.39 -> 6.76 | Improved |
| add detection supervision to 3D coordinate regression (ablation) | 3D CR · HANDS 2017 | mean error (mm): 11.39 -> 7.00 | Improved |
| use truncated Euclidean distance instead of Euclidean distance in 3D (ablation) | 3D CR · HANDS 2017 | mean error (mm): 7.92 -> 8.98 | Hurt |
| add final coordinate regression to offset maps (combine detection with regression) (ablation) | 2D offset map · HANDS 2017 | mean error (mm): 49.83 -> 9.34 | Improved |
| add final coordinate regression to offset maps in 3D (ablation) | 3D offset map · HANDS 2017 | mean error (mm): 51.17 -> 6.76 | Improved |
Closest papers by meaning
- Dense 3D Regression for Hand Pose Estimation (CVPR 2018 · similarity 0.91)
- Robust 3D Hand Pose Estimation in Single Depth Images: From Single-View CNN to Multi-View CNNs (CVPR 2016 · similarity 0.89)
- BigHand2.2M Benchmark: Hand Pose Dataset and State of the Art Analysis (CVPR 2017 · similarity 0.89)
- Self-Supervised 3D Hand Pose Estimation Through Training by Fitting (CVPR 2019 · similarity 0.89)
- DeepHand: Robust Hand Pose Estimation by Completing a Matrix Imputed With Deep Features (CVPR 2016 · similarity 0.88)
- Keypoint Fusion for RGB-D Based 3D Hand Pose Estimation (AAAI 2024 · similarity 0.86)
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.
