Research map / 3D vision / Human pose estimation
Monocular 3D Pose and Shape Estimation of Multiple People in Natural Scenes - The Importance of Multiple Scene Constraints
Read the paperCheck 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 |
|---|---|---|---|
| Add the semantic projection term to the 2d projection loss (ablation) | Monocular multi-person reconstruction · Panoptic | translation error: 390.3 -> 377.6 | Helped |
| Add the semantic projection term to the 2d projection loss (ablation) | Monocular multi-person reconstruction · Panoptic | pose error: 157.7 -> 165.6 | Hurt |
| Add temporal smoothing (ablation) | Semantic-feedback monocular reconstruction · Panoptic | translation error: 377.6 -> 371.7 | Helped |
| Add temporal smoothing (ablation) | Semantic-feedback monocular reconstruction · Panoptic | pose error: 165.6 -> 160.8 | Helped |
| Add the ground-plane constraint (ablation) | Semantic-feedback reconstruction with temporal smoothing · Panoptic | translation error: 371.7 -> 315.5; -15% | Helped |
| Add the ground-plane constraint (ablation) | Semantic-feedback reconstruction with temporal smoothing · Panoptic | pose error: 160.8 -> 153.4 | Helped |
| Add semantic 3d pose and shape feedback (ablation) | DMHS · Human3.6M test set: WalkingDog, Sitting, Sitting Down | mean per joint 3d position error: 78 -> 75; 119 -> 109; 106 -> 101 | Helped |
| Use multiple camera views (ablation) | Semantic-feedback pose estimation · Human3.6M test set: WalkingDog, Sitting, Sitting Down | mean per joint 3d position error: 75 -> 51; 109 -> 71; 101 -> 65 | Helped |
| Add temporal smoothing (ablation) | Multi-view pose estimation · Human3.6M test set: WalkingDog, Sitting, Sitting Down | mean per joint 3d position error: 51 -> 48; 71 -> 68; 65 -> 64 | Helped |
| Use the complete method rather than DMHS initialization alone | Monocular multi-person reconstruction · Panoptic | pose error: 203.4 -> 153.4; -25% | Helped |
Closest papers by meaning
- Deep Network for the Integrated 3D Sensing of Multiple People in Natural Images (NeurIPS 2018 · similarity 0.91)
- Monocular 3D Multi-Person Pose Estimation by Integrating Top-Down and Bottom-Up Networks (CVPR 2021 · similarity 0.91)
- MultiPly: Reconstruction of Multiple People from Monocular Video in the Wild (CVPR 2024 · similarity 0.90)
- Graph and Temporal Convolutional Networks for 3D Multi-person Pose Estimation in Monocular Videos (AAAI 2021 · similarity 0.90)
- CoMotion: Concurrent Multi-person 3D Motion (ICLR 2025 · similarity 0.89)
- ODHSR: Online Dense 3D Reconstruction of Humans and Scenes from Monocular Videos (CVPR 2025 · similarity 0.88)
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.