Research map / 3D vision / Human pose estimation
Human Pose Estimation With Iterative Error Feedback
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 IEF instead of direct pose prediction (ablation) | GoogLeNet pose estimator · MPII validation set | FBody PCKh-0.5: 74.8 -> 81.0 | Helped |
| predict bounded corrections rather than target keypoint locations at each iteration (ablation) | iterative GoogLeNet pose estimator · MPII validation set | FBody PCKh-0.5: 73.4 -> 81.0 | Helped |
| train all correction steps together instead of using Fixed Path Consolidation (ablation) | IEF · MPII validation set | PCKh: almost 10 PCKh points | Hurt |
| use feedback for the left knee instead of direct prediction (ablation) | left-knee pose prediction · MPII validation set, left knee | PCKh-0.5: 64.6 -> 69.2 | Helped |
| add the left-hip channel to left-knee feedback (ablation) | IEF Left Knee · MPII validation set, left knee | PCKh-0.5: 69.2 -> 72.8 | Helped |
| model all joints together instead of only the left knee and left hip (ablation) | IEF Left Knee + Left Hip · MPII validation set, left knee | PCKh-0.5: 72.8 -> 73.8 | Helped |
| use a fourth correction step (ablation) | IEF · MPII pose estimation | accuracy | No clear effect |
Closest papers by meaning
- Self-Correctable and Adaptable Inference for Generalizable Human Pose Estimation (CVPR 2023 · similarity 0.87)
- Convolutional Pose Machines (CVPR 2016 · similarity 0.87)
- Articulated Pose Estimation Using Hierarchical Exemplar-Based Models (AAAI 2016 · similarity 0.86)
- Deep High-Resolution Representation Learning for Human Pose Estimation (CVPR 2019 · similarity 0.86)
- Personalizing Human Video Pose Estimation (CVPR 2016 · similarity 0.86)
- HiPoser: 3D Human Pose Estimation with Hierarchical Shared Learning at Parts-Level Using Inertial Measurement Units (AAAI 2025 · 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.