Research map / 3D vision / Human pose estimation
TRB: A Novel Triplet Representation for Understanding 2D Human Body
Haodong Duan, Kwan-Yee Lin, Sheng Jin, Wentao Liu, Chen Qian, Wanli Ouyang
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 |
|---|---|---|---|
| jointly learn skeleton and contour keypoints (multitask) (ablation) | TRB-Net · MPII_trb val | skeleton PCKh: 88.0 -> 88.6 | Improved |
| jointly learn skeleton and contour keypoints (multitask) (ablation) | TRB-Net · MPII_trb val | contour PCKh: 84.1 -> 84.8 | Improved |
| add X-structured message passing (ablation) | TRB-Net · MPII_trb val | TRB mean PCKh: 86.2 -> 86.8 | Improved |
| replace Xs with directional convolution (ablation) | TRB-Net · MPII_trb val | TRB mean PCKh: 86.8 -> 87.4 | Improved |
| use directional convolution instead of normal convolution with same parameters (ablation) | TRB-Net · MPII_trb val | TRB mean PCKh: 86.5 -> 87.4 | Improved |
| add adaptive weights to multi-path directional convolution fusion (ablation) | TRB-Net · MPII_trb val | TRB mean PCKh: 87.1 -> 87.4 | Improved |
| add directional convolution to multitask baseline (ablation) | TRB-Net · MPII_trb val | contour PCKh: 84.8 -> 86.2 | Improved |
| add pairwise mapping to the multitask baseline (ablation) | TRB-Net · MPII_trb val | TRB mean PCKh: 86.2 -> 87.2 | Improved |
| fuse warped heatmaps (pairwise mapping) at stack 1 (ablation) | TRB-Net · MPII_trb val | TRB mean PCKh: 84.3 -> 85.3 | Improved |
| fuse warped heatmaps (pairwise mapping) at stack 2 (ablation) | TRB-Net · MPII_trb val | TRB mean PCKh: 86.6 -> 87.2 | Improved |
| combine directional convolution with pairwise mapping (ablation) | TRB-Net · MPII_trb val | TRB mean PCKh: 87.2 -> 87.6 | Improved |
| use contour keypoints with half of COCO data (hg2-ours_sub) versus baseline with all data | hourglass-2 baseline · COCO val | AP | No clear effect |
Closest papers by meaning
- Learning 3D Human Shape and Pose from Dense Body Parts (TPAMI 2020 · similarity 0.89)
- Unite the People: Closing the Loop Between 3D and 2D Human Representations (CVPR 2017 · similarity 0.88)
- Capturing the Motion of Every Joint: 3D Human Pose and Shape Estimation with Independent Tokens (ICLR 2023 · similarity 0.88)
- Lifting 2D Human Pose to 3D with Domain Adapted 3D Body Concept (IJCV 2023 · similarity 0.88)
- Probabilistic 3D Human Shape and Pose Estimation From Multiple Unconstrained Images in the Wild (CVPR 2021 · similarity 0.88)
- PoseBH: Prototypical Multi-Dataset Training Beyond Human Pose Estimation (CVPR 2025 · 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.
