Research map / 3D vision / Human pose estimation
HEMlets PoSh: Learning Part-Centric Heatmap Triplets for 3D Human Pose and Shape Estimation
Kun Zhou, Xiaoguang Han, Nianjuan Jiang, Kui Jia, Jiangbo Lu
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 2D heatmap supervision (ablation) | 3D pose baseline · Human3.6M Protocol #1 | MPJPE: 47.1 -> 44.2 | Improved |
| add HEMlets supervision (ablation) | 3D pose baseline · Human3.6M Protocol #1 | MPJPE: 47.1 -> 42.6 | Improved |
| add HEMlets supervision to 2D heatmaps (ablation) | 3D pose with 2D heatmaps · Human3.6M Protocol #1 | MPJPE: 44.2 -> 39.9 | Improved |
| add 2D heatmaps to HEMlets (no MPII) (ablation) | HEMlets model · Human3.6M Protocol #1 (no MPII) | MPJPE: 46.0 -> 45.1 | Improved |
| add extra MPII 2D dataset | HEMlets model · Human3.6M Protocol #1 | MPJPE: 45.1 -> 39.9 | Improved |
| add FBI dataset to training | HEMlets model · Human3.6M Protocol #1 | MPJPE: 39.9 -> 36.5 | Improved |
| add Ordinal depth-order data (ablation) | HEMlets model · MPI-INF-3DHP | 3DPCK: 75.3 -> 76.1 | Improved |
| add FBI data (ablation) | HEMlets model · MPI-INF-3DHP | 3DPCK: 75.3 -> 76.9 | Improved |
| add Ordinal on top of FBI (ablation) | HEMlets model · MPI-INF-3DHP | 3DPCK: 76.9 -> 76.5 | Hurt |
| add 3D joints branch to high-level feature branch (ablation) | HEMlets PoSh · 3DPW | Rec. error: 63.1 -> 58.8 | Improved |
| add high-level feature branch to 3D joints branch (ablation) | HEMlets PoSh · 3DPW | Rec. error: 60.3 -> 58.8 | Improved |
| use groundtruth pose instead of predicted (ablation) | HEMlets PoSh · 3DPW | Rec. error: 58.8 -> 9.4 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- SPIN* — SPIN: Structure-Preserving Inner Offset Network for Scene Text Recognition (AAAI 2021) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- HEMlets Pose: Learning Part-Centric Heatmap Triplets for Accurate 3D Human Pose Estimation (ICCV 2019 · similarity 0.98)
- A Simple yet Effective Baseline for 3D Human Pose Estimation (ICCV 2017 · similarity 0.89)
- Keep It SMPL: Automatic Estimation of 3D Human Pose and Shape from a Single Image (ECCV 2016 · similarity 0.88)
- Learning to Reconstruct 3D Human Pose and Shape via Model-Fitting in the Loop (ICCV 2019 · similarity 0.88)
- HybrIK: A Hybrid Analytical-Neural Inverse Kinematics Solution for 3D Human Pose and Shape Estimation (CVPR 2021 · similarity 0.88)
- Neural Localizer Fields for Continuous 3D Human Pose and Shape Estimation (NeurIPS 2024 · similarity 0.87)
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.
