Research map / 3D vision / Human pose estimation
Uncertainty-Aware Adaptation for Self-Supervised 3D Human Pose Estimation
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 |
|---|---|---|---|
| Add target joint-uncertainty adaptation to the joint-level baseline (ablation) | MRP-Net joint-level adaptation · 3DPW truncated/occluded test split | MPJPE (in-view joints): 191.2 -> 130.7 | Helped |
| Add target pseudo-label self-training to joint-level adaptation (ablation) | MRP-Net joint-level adaptation · 3DPW truncated/occluded test split | MPJPE (in-view joints): 130.7 -> 98.0 | Helped |
| Add target pose-uncertainty minimization (ablation) | MRP-Net pose-level adaptation · Human3.6M (SURREAL→H3.6M) | MPJPE: 122.4 -> 113.4 | Helped |
| Add target pseudo-label self-training (ablation) | MRP-Net pose-level adaptation · Human3.6M (SURREAL→H3.6M) | MPJPE: 113.4 -> 106.3 | Helped |
| Add the fusion network at inference (ablation) | MRP-Net pose-level adaptation · Human3.6M (SURREAL→H3.6M) | MPJPE: 106.3 -> 103.2 | Helped |
| Fuse regression pose, localization pose, and joint confidences rather than relying on an individual input (ablation) | MRP-Net fusion network · 3DPW | MPJPE: 100, 122, 115 -> 91 | Helped |
| Use MRP-Net unsupervised adaptation instead of the strongest listed general-adaptation baseline | Human3.6M-source adaptation · 3DPW | MPJPE: 94.7 -> 91.9 | Helped |
| Use MRP-Net unsupervised adaptation instead of the strongest listed general-adaptation baseline | Human3.6M-source adaptation · 3DHP | MPJPE: 99.3 -> 96.2 | Helped |
| Use MRP-Net unsupervised adaptation instead of the strongest listed general-adaptation baseline | Human3.6M-source adaptation · SURREAL | MPJPE: 103.3 -> 99.6 | Helped |
| Use MRP-Net unsupervised adaptation instead of the strongest listed general-adaptation baseline | Human3.6M-source adaptation · HumanEva | MPJPE: 69.2 -> 66.8 | Helped |
| Add ISO test-time optimization | MRP-Net · 3DPW | MPJPE: 91.9 -> 89.6 | Helped |
Closest papers by meaning
- Sparseness Meets Deepness: 3D Human Pose Estimation From Monocular Video (CVPR 2016 · similarity 0.90)
- Self-Supervised 3D Human Mesh Recovery from a Single Image with Uncertainty-Aware Learning (AAAI 2024 · similarity 0.89)
- PoseIRM: Enhance 3D Human Pose Estimation on Unseen Camera Settings via Invariant Risk Minimization (CVPR 2024 · similarity 0.88)
- ManiPose: Manifold-Constrained Multi-Hypothesis 3D Human Pose Estimation (NeurIPS 2024 · similarity 0.88)
- Source-Free and Image-Only Unsupervised Domain Adaptation for Category Level Object Pose Estimation (ICLR 2024 · similarity 0.88)
- Semantics-aware Test-time Adaptation for 3D Human Pose Estimation (ICML 2025 · similarity 0.88)
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.