Research map / 3D vision / Human pose estimation
End-To-End Learning of Deformable Mixture of Parts and Deep Convolutional Neural Networks for 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 |
|---|---|---|---|
| jointly train the DCNN and deformable mixture-of-parts model (ablation) | ChenNet with a loopy graph · LSP test | strict PCP (Mean): 74.2 -> 78.6 | Helped |
| add an edge between knees to form a loopy graph (ablation) | ChenNet tree-structured model · LSP test | strict PCP (Mean): 78.1 -> 78.6 | Helped |
| use VGG instead of ChenNet (ablation) | tree-structured model · LSP test | strict PCP (Mean): 78.1 -> 80.7 | Helped |
| add spatial constraints through message passing (ablation) | VGG part detectors · LSP test | strict PCP (Mean): 60.1 -> 80.7 | Helped |
| add spatial constraints through message passing (ablation) | ChenNet part detectors · LSP test | strict PCP (Mean): 40.6 -> 78.1 | Helped |
| use a loopy graph instead of a tree-structured graph (ablation) | VGG pose model · LSP test | strict PCP (Mean): 80.7 -> 81.1 | Helped |
| increase message passing from one to two layers (ablation) | VGG loopy model · LSP test | strict PCP (Mean): 80.7 -> 80.9 | Helped |
| increase message passing from two to three layers (ablation) | VGG loopy model · LSP test | strict PCP (Mean): 80.9 -> 81.1 | Helped |
| increase mixture types from 11 to 13 (ablation) | pose model · LSP test | strict PCP: +1% | Helped |
| change patch size to 0.8 times the original scale (ablation) | VGG pose model · LSP | strict PCP: -6.2% | Hurt |
| change patch size to 1.2 times the original scale (ablation) | VGG pose model · LSP | strict PCP: -0.5% | Hurt |
| increase the ground-truth heat-map threshold δ from 1/5 to 1/3 of patch size (ablation) | pose model · LSP | strict PCP: less than 1% | No clear effect |
Closest papers by meaning
- Joint Multi-Person Pose Estimation and Semantic Part Segmentation (CVPR 2017 · similarity 0.89)
- 3D Human Pose Estimation From a Single Image via Distance Matrix Regression (CVPR 2017 · similarity 0.88)
- Thin-Slicing Network: A Deep Structured Model for Pose Estimation in Videos (CVPR 2017 · similarity 0.88)
- DeepCut: Joint Subset Partition and Labeling for Multi Person Pose Estimation (CVPR 2016 · similarity 0.87)
- Deep Dual Consecutive Network for Human Pose Estimation (CVPR 2021 · similarity 0.87)
- Structured Feature Learning for Pose Estimation (CVPR 2016 · 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.