Research map / 3D vision / Human pose estimation
HMOR: Hierarchical Multi-Person Ordinal Relations for Monocular Multi-Person 3D Pose Estimation
Can Wang, Jiefeng Li, Wentao Liu, Chen Qian, Cewu 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 HMOR supervision (joint, part, instance relations) over vanilla baseline (ablation) | HMOR integrated model · 3DPW | MPJPE: 95.7 -> 88.3 | Improved |
| add HMOR supervision over baseline (ablation) | HMOR integrated model · 3DPW | ABS-MPJPE: 169.3 -> 119.6 | Improved |
| add absolute-pose L1 loss to baseline (ablation) | HMOR integrated model · 3DPW | ABS-MPJPE: 169.3 -> 158.2 | Improved |
| use HMOR instead of L_abs supervision (ablation) | HMOR integrated model · 3DPW | ABS-MPJPE: 158.2 -> 119.6 | Improved |
| add instance relation (on top of joint+part) (ablation) | HMOR integrated model · 3DPW | ABS-MPJPE: 125.9 -> 119.6 | Improved |
| add part relation to joint+instance (ablation) | HMOR integrated model · 3DPW | PA-MPJPE: 58.5 -> 57.8 | Improved |
| add joint relation to part+instance (ablation) | HMOR integrated model · 3DPW | MPJPE: 89.5 -> 88.3 | Improved |
| use vector part relations instead of particle representation (ablation) | HMOR integrated model · 3DPW | MPJPE: 89.0 -> 88.3 | Improved |
| add random virtual view sampling during training (ablation) | HMOR integrated model · 3DPW | MPJPE: 88.3 -> 87.7 | Improved |
| add coarse-to-fine depth refinement (ablation) | HMOR integrated model · 3DPW | ABS-MPJPE: 133.6 -> 118.5 | Improved |
| compare with Moon et al. on MuPoTS-3D absolute pose | HMOR integrated model · MuPoTS-3D | 3DPCK_abs: 31.5 -> 43.8 | Improved |
| compare with Zanfir on CMU Panoptic | HMOR integrated model · CMU Panoptic | MPJPE: 72.1 -> 51.6 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- LCRNet — LCR-Net++: Multi-person 2D and 3D Pose Detection in Natural Images (TPAMI 2019) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Dual Networks Based 3D Multi-Person Pose Estimation From Monocular Video (TPAMI 2022 · similarity 0.91)
- Ordinal Depth Supervision for 3D Human Pose Estimation (CVPR 2018 · similarity 0.90)
- Monocular 3D Human Pose Estimation by Generation and Ordinal Ranking (ICCV 2019 · similarity 0.90)
- Monocular 3D Multi-Person Pose Estimation by Integrating Top-Down and Bottom-Up Networks (CVPR 2021 · similarity 0.90)
- MV-SSM: Multi-View State Space Modeling for 3D Human Pose Estimation (CVPR 2025 · similarity 0.87)
- COMPOSE: Hypergraph Cover Optimization for Multi-view 3D Human Pose Estimation (NeurIPS 2026 · 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.
