Research map / 3D vision / Human pose estimation
Human Pose Estimation Using Global and Local Normalization
Ke Sun, Cuiling Lan, Junliang Xing, Wenjun Zeng, Dong Liu, Jingdong Wang
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 body normalization in refinement stage 1 (ablation) | FCN refinement · LSP OC (train LSP) | PCK@0.2 total: 82.9 -> 83.2 | Improved |
| add limb normalization in refinement stage 2 (ablation) | FCN refinement · LSP OC (train LSP) | PCK@0.2 total: 83 -> 83.5 | Improved |
| add body normalization in stage 1 (ablation) | FCN refinement · LSP OC (train LSP+LSPET) | PCK@0.2 total: 86 -> 86.8 | Improved |
| add limb normalization in stage 2 (ablation) | FCN refinement · LSP OC (train LSP+LSPET) | PCK@0.2 total: 85.3 -> 87.4 | Improved |
| add multi-scale extra supervision (ablation) | FCN_8s · LSP OC (train LSP) | PCK@0.2 total: 82 -> 82.6 | Improved |
| add multi-scale score map fusion (ablation) | FCN_8s (Extra) · LSP OC (train LSP) | PCK@0.2 total: 82.6 -> 82.8 | Improved |
| add global normalization (type-agnostic) versus type supervision or multi-branch (ablation) | FCN · LSP OC (train LSP) | PCK@0.2 total: 81.5 -> 83.3; type supervision 82.3; multi-branch 82.5 | Improved |
| add refinement to ResNet-152 detector | ResNet-152 · LSP PC (train MPII+LSPET+LSP) | PCK@0.2 total: 89.2 -> 90.1 | Improved |
| add refinement to Hourglass detector | Hourglass · LSP PC (train MPII+LSPET+LSP) | PCK@0.2 total: 90.9 -> 91.6 | Improved |
| add refinement to ResNet-152 detector | ResNet-152 · FLIC OC | AUC: 75.3 -> 76.9 | Improved |
| add refinement to Hourglass detector | Hourglass · MPII test | PCKh@0.5 total: 90.9 -> 91.0 | No clear effect |
| add refinement to FCN detector | FCN · LSP OC (train LSP) | PCK@0.2 total: 82.8 -> 83.7 | Improved |
Closest papers by meaning
- SRNet: Improving Generalization in 3D Human Pose Estimation with a Split-and-Recombine Approach (ECCV 2020 · similarity 0.90)
- Deep Globally Constrained MRFs for Human Pose Estimation (ICCV 2017 · similarity 0.89)
- 3D Human Pose Estimation From a Single Image via Distance Matrix Regression (CVPR 2017 · similarity 0.89)
- A Simple yet Effective Baseline for 3D Human Pose Estimation (ICCV 2017 · similarity 0.89)
- Optimizing Human Pose Estimation Through Focused Human and Joint Regions (AAAI 2025 · similarity 0.88)
- Sequential Joint Dependency Aware Human Pose Estimation with State Space Model (AAAI 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.
