Research map / 3D vision / Human pose estimation
DeepCut: Joint Subset Partition and Labeling for Multi Person 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 partition and label body-part candidates instead of estimating poses independently in detected ROIs | DeepCut MP Dense-CNN · WAF | mPCP: 55.3 -> 84.7 | Helped |
| Jointly partition and label body-part candidates instead of estimating poses independently in detected ROIs | DeepCut MP Dense-CNN · WAF | AOP: 73.8 -> 86.5 | Helped |
| Add joint subset partitioning and labeling | AFR-CNN · MPII Multi-Person | FBody AP: 47.1 -> 51.4 | Helped |
| Increase proposal region scale from 1 to 4 to include more context (ablation) | AFR-CNN AlexNet · LSP (PC) | PCK: 56.9 -> 65.1 | Helped |
| Tune learning rate, schedule, and training duration (ablation) | AFR-CNN AlexNet · LSP (PC) | PCK: 65.1 -> 72.4 | Helped |
| Fine-tune the MPII-trained AFR-CNN on LSP (ablation) | AFR-CNN AlexNet · LSP (PC) | PCK: 72.4 -> 77.9 | Helped |
| Use VGG instead of AlexNet (ablation) | AFR-CNN · LSP (PC) | PCK: 72.4 -> 77.9 | Helped |
| Add LSPET training data (ablation) | Dense-CNN · LSP (PC) | PCK: 80.8 -> 83.0 | Helped |
| Replace softmax with sigmoid multi-label classification (ablation) | Dense-CNN · LSP (PC) | PCK: 83.0 -> 83.8 | Helped |
| Add location refinement (ablation) | Dense-CNN · LSP (PC) | PCK: 83.8 -> 84.8 | Helped |
| Add regression from each part to other part locations (ablation) | Dense-CNN · LSP (PC) | PCK: 84.8 -> 86.1 | Helped |
| Add joint subset partitioning and labeling (ablation) | AFR-CNN · LSP (PC) | PCK: 82.8 -> 83.0 | Helped |
Closest papers by meaning
- Simple Pose: Rethinking and Improving a Bottom-up Approach for Multi-Person Pose Estimation (AAAI 2020 · similarity 0.91)
- Detangling People: Individuating Multiple Close People and Their Body Parts via Region Assembly (CVPR 2017 · similarity 0.91)
- ArtTrack: Articulated Multi-Person Tracking in the Wild (CVPR 2017 · similarity 0.90)
- Joint Multi-Person Pose Estimation and Semantic Part Segmentation (CVPR 2017 · similarity 0.90)
- PoseTrack: Joint Multi-Person Pose Estimation and Tracking (CVPR 2017 · similarity 0.89)
- Monocular 3D Pose and Shape Estimation of Multiple People in Natural Scenes - The Importance of Multiple Scene Constraints (CVPR 2018 · similarity 0.89)
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.