Research map / 3D vision / Human pose estimation
HEMlets Pose: Learning Part-Centric Heatmap Triplets for Accurate 3D Human Pose 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 intermediate supervision (ablation) | 3D pose network (no intermediate supervision) · Human3.6M Protocol #1 | MPJPE (mm): 47.1 -> 44.2 | Improved |
| add HEMlets supervision instead of 2D heatmaps only (ablation) | 3D pose network · Human3.6M Protocol #1 | MPJPE (mm): 44.2 -> 42.6 | Improved |
| add both HEMlets and 2D heatmap supervision (ablation) | 3D pose network · Human3.6M Protocol #1 | MPJPE (mm): 47.1 -> 39.9 | Improved |
| add 2D heatmap supervision on top of HEMlets (no MPII data) (ablation) | HEMlets · Human3.6M Protocol #1 (Human3.6M only) | MPJPE (mm): 46.0 -> 45.1 | Improved |
| add HEMlets supervision (no MPII data) (ablation) | 3D pose network · Human3.6M Protocol #1 (Human3.6M only) | MPJPE (mm): 55.3 -> 46.0 | Improved |
| use three-state HEMlets instead of 5s-HEM or 2s-HEM variants (ablation) | HEMlets · Human3.6M validation loss | validation loss | Improved |
| add FBI relative-depth annotations (ablation) | HEMlets · MPI-INF-3DHP | 3DPCK: 75.3 -> 76.9 | Improved |
| add Ordinal relative-depth annotations (ablation) | HEMlets · MPI-INF-3DHP | 3DPCK: 75.3 -> 76.1 | Improved |
| add both FBI and Ordinal annotations rather than FBI alone (ablation) | HEMlets · MPI-INF-3DHP | 3DPCK: 76.9 -> 76.5 | No clear effect |
| use HEMlets pose instead of Sun et al. | HEMlets · Human3.6M Protocol #1 | MPJPE (mm): 49.6 -> 39.9 | Improved |
| use HEMlets pose instead of Pavlakos et al. | HEMlets · MPI-INF-3DHP | 3DPCK All: 71.9 -> 75.3 | Improved |
| use HEMlets pose instead of Pavlakos et al. on Studio GS | HEMlets · MPI-INF-3DHP | 3DPCK Studio GS: 76.5 -> 75.6 | Hurt |
Closest papers by meaning
- HEMlets PoSh: Learning Part-Centric Heatmap Triplets for 3D Human Pose and Shape Estimation (TPAMI 2021 · similarity 0.98)
- A Simple yet Effective Baseline for 3D Human Pose Estimation (ICCV 2017 · similarity 0.89)
- Human Pose Estimation via Convolutional Part Heatmap Regression (ECCV 2016 · similarity 0.89)
- Coarse-To-Fine Volumetric Prediction for Single-Image 3D Human Pose (CVPR 2017 · similarity 0.88)
- Attention-Propagation Network for Egocentric Heatmap to 3D Pose Lifting (CVPR 2024 · 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.
