Research map / 3D vision / Human pose estimation
Peeking into occluded joints: A novel framework for crowd pose estimation
Lingteng Qiu, Xuanye Zhang, Yanran Li, Guanbin Li, Xiaojun Wu, Zixiang Xiong, Xiaoguang Han, Shuguang Cui
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 OPEC-Net GCN correction stage | AlphaPose+ · OCPose | mAP@0.5:0.95: 30.8 -> 32.8 | Improved |
| add OPEC-Net GCN correction stage | AlphaPose+ · CrowdPose | mAP@0.5:0.95: 68.5 -> 70.6 | Improved |
| add OPEC-Net GCN correction stage | AlphaPose+ · OCHuman | mAP@0.5:0.95: 27.5 -> 29.1 | Improved |
| add OPEC-Net GCN correction stage | SimplePose · MSCOCO test-dev | mAP@0.5:0.95: 73.7 -> 73.9 | No clear effect |
| add CoupleGraph on top of OPEC-Net | OPEC-Net · OCPose | mAP@0.5:0.95: 32.8 -> 33.6 | Improved |
| add Image-Guided strategy in GCN (ablation) | OPEC-Net · OCPose | mAP@0.5:0.95: 30.8 -> 32.8 | Improved |
| add Progressive design in GCN (ablation) | OPEC-Net · OCPose | mAP@0.5:0.95: 32.2 -> 32.8 | Improved |
| use multi-scale features instead of single F3 (ablation) | OPEC-Net · OCPose | mAP@0.5:0.95: 32.5 -> 32.8 | Improved |
| add Cascaded Feature Adaption module (ablation) | OPEC-Net · OCPose | mAP@0.5:0.95: 32.1 -> 32.8 | Improved |
| add Fusion Block in CFA module (ablation) | OPEC-Net · CrowdPose | mAP@0.5:0.95: 69.6 -> 70.6 | Improved |
| add OPEC-Net on invisible joints | AlphaPose+ · OCPose Inv@75 | Inv@75: 50.7% -> 55.6% | Improved |
| add OPEC-Net on visible joints | AlphaPose+ · OCPose V@90 | V@90: 55.3% -> 55.1% | No clear effect |
Closest papers by meaning
- Human Pose Recognition via Occlusion-Preserving Abstract Images (ECCV 2024 · similarity 0.88)
- Explicit Occlusion Reasoning for Multi-Person 3D Human Pose Estimation (ECCV 2022 · similarity 0.88)
- Occlusion-Aware Networks for 3D Human Pose Estimation in Video (ICCV 2019 · similarity 0.88)
- Occ^2Net: Robust Image Matching Based on 3D Occupancy Estimation for Occluded Regions (ICCV 2023 · similarity 0.88)
- $\text{Di}^2\text{Pose}$: Discrete Diffusion Model for Occluded 3D Human Pose Estimation (NeurIPS 2024 · similarity 0.86)
- ODHSR: Online Dense 3D Reconstruction of Humans and Scenes from Monocular Videos (CVPR 2025 · similarity 0.85)
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.
