Research map / Label-efficient and robust learning / Domain adaptation and generalization
Domain Adaptive Hand Keypoint and Pixel Localization in the Wild
Takehiko Ohkawa, Yu-Jhe Li, Qichen Fu, Ryosuke Furuta, Kris M. Kitani, Yoichi Sato
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 multi-task adaptation (joint keypoint and mask) instead of single-task (ablation) | C-GAC · DexYCB to HO3D | IoU: improved by 2.7 / 2.6 over the single-task adaption in hand segmentation | Improved |
| add multi-task adaptation instead of single-task, keypoint branch (ablation) | C-GAC · DexYCB to HO3D | PCK: 50.1/51.0 in the PCK single-task; marginal gain | No clear effect |
| add confidence weighting from two-teacher divergence to GAC-Distill (ablation) | GAC-Distill · DexYCB to HO3D, HanCo, FPHA | average PCK/IoU | Improved |
| replace EMA teacher update with distillation-based teacher update (ablation) | GAC + MT · DexYCB to HO3D | hand segmentation IoU | Improved |
| add geometric augmentation consistency (GAC) self-training | Source only · DexYCB to HanCo | average score: up to 47.4/47.9 in the average score | Improved |
| use adversarial adaptation (DANN, RegDA) instead of consistency training | GAC · DexYCB to HO3D, HanCo, FPHA | average score | Hurt |
| add UMA dropout-based confidence to GAC | GAC · DexYCB to HO3D, HanCo, FPHA | average score | Mixed |
| add CPL confident instance selection to GAC | GAC · DexYCB to HO3D, HanCo, FPHA | keypoint regression | Mixed |
| add C-GAC adaptation versus source only | Source only · DexYCB to HO3D | average score: improved by 5.3/14.4 in the average task score | Improved |
Closest papers by meaning
- Self-Supervised 3D Hand Pose Estimation Through Training by Fitting (CVPR 2019 · similarity 0.87)
- Detecting Hands and Recognizing Physical Contact in the Wild (NeurIPS 2020 · similarity 0.87)
- Dual Grid Net: Hand Mesh Vertex Regression from Single Depth Maps (ECCV 2020 · similarity 0.86)
- FoundHand: Large-Scale Domain-Specific Learning for Controllable Hand Image Generation (CVPR 2025 · similarity 0.86)
- SiMHand: Mining Similar Hands for Large-Scale 3D Hand Pose Pre-training (ICLR 2025 · similarity 0.85)
- 3D Visibility-Aware Generalizable Neural Radiance Fields for Interacting Hands (AAAI 2024 · 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.
