Research map / Label-efficient and robust learning / Semi-supervised learning
MONET: Multiview Semi-Supervised Keypoint Detection via Epipolar Divergence
Yuan Yao, Yasamin Jafarian, 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 3D-tracked spatial augmentation to supervised training (ablation) | CPM supervised · Multiview Human | reprojection error: 77.8 -> 69.0 | Improved |
| add 3D-tracked spatial augmentation to supervised training (ablation) | CPM supervised · Multiview Monkey | reprojection error: 31.1 -> 12.9 | Improved |
| add temporal tracking (spatiotemporal augmentation) to spatial augmentation (ablation) | CPM · Multiview Monkey | reprojection error: 12.9 -> 8.10 | Improved |
| add temporal tracking (spatiotemporal augmentation) to spatial augmentation (ablation) | CPM · Multiview Dog | reprojection error: 37.5 -> 24.0 | Improved |
| add re-triangulation and retraining (Bootstrapping II) to Bootstrapping I | CPM · Multiview Human | reprojection error: 28.5 -> 35.4 | Hurt |
| add re-triangulation and retraining (Bootstrapping II) to Bootstrapping I | CPM · Multiview Dog | reprojection error: 18.9 -> 17.1 | Improved |
| replace Bootstrapping II by epipolar divergence (MONET) | CPM · Multiview Human | reprojection error: 35.4 -> 15.0 | Improved |
| replace Bootstrapping II by epipolar divergence (MONET) | CPM · Multiview Monkey | reprojection error: 9.97 -> 5.45 | Improved |
| replace Bootstrapping II by epipolar divergence (MONET) | CPM · CMU Panoptic | reprojection error: 13.7 -> 12.8 | Improved |
| add a second labeled subject (S1,5) to S1 (ablation) | MONET · Human3.6M | mean pixel error: 5.5 -> 5.2 | Improved |
| add a third labeled subject (S1,5,6) to S1,5 (ablation) | MONET · Human3.6M | mean pixel error: 5.2 -> 4.3 | Improved |
| add more labeled subjects (S1,5,6 instead of S1) (ablation) | MONET · Human3.6M foot (L.Foot) | mean pixel error: 6.2 -> 4.7 | Mixed |
Closest papers by meaning
- Dense Keypoints via Multiview Supervision (NeurIPS 2021 · similarity 0.91)
- Semi-supervised Keypoint Localization (ICLR 2021 · similarity 0.87)
- ED-Pose++: Enhanced Explicit Box Detection for Conventional and Interactive Multi-Object Keypoint Detection (TPAMI 2025 · similarity 0.87)
- Towards Complementary Keypoint Detection via Mixture of Detectors (NeurIPS 2026 · similarity 0.86)
- Back to 3D: Few-Shot 3D Keypoint Detection with Back-Projected 2D Features (CVPR 2024 · similarity 0.85)
- Detect Any Keypoints: An Efficient Light-Weight Few-Shot Keypoint Detector (AAAI 2024 · similarity 0.84)
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.
