Research map / 3D vision / Human pose estimation
RSGNet: Relation based Skeleton Graph Network for Crowded Scenes Pose Estimation
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 |
|---|---|---|---|
| Replace HRNet-W32 with RSGNet at 384 × 288 | CrowdPose pose estimation · CrowdPose test | AP: 73.5 -> 74.3 (+0.8) | Helped |
| Replace HRNet-W32 with RSGNet at 256 × 192 | CrowdPose pose estimation · CrowdPose test | AP: 71.7 -> 73.6 (+1.9) | Helped |
| Replace HRNet-W48 with RSGNet at 384 × 288 | COCO pose estimation · COCO val | AP: 76.3 -> 77.0 (+0.7) | Helped |
| Replace HRNet-W48 with RSGNet at 384 × 288 | COCO pose estimation · COCO test-dev | AP: 75.5 -> 76.0 (+0.5) | Helped |
| Add Target-aware Relation Parser (TRP) (ablation) | HRNet-W32 · CrowdPose test | AP: 71.7 -> 73.1 | Helped |
| Add Skeleton Graph Machine (SGM) to HRNet-W32 + TRP (ablation) | CrowdPose test · CrowdPose test | AP: 73.1 -> 73.6 | Helped |
| Add Target-aware Relation Parser (TRP) (ablation) | HRNet-W32 · COCO minival | AP: 74.4 -> 74.9 | Helped |
| Add Skeleton Graph Machine (SGM) to HRNet-W32 + TRP (ablation) | COCO minival · COCO minival | AP: 74.9 -> 75.7 | Helped |
Closest papers by meaning
- CrowdPose: Efficient Crowded Scenes Pose Estimation and a New Benchmark (CVPR 2019 · similarity 0.89)
- Robust Pose Estimation in Crowded Scenes with Direct Pose-Level Inference (NeurIPS 2021 · similarity 0.88)
- Graph and Temporal Convolutional Networks for 3D Multi-person Pose Estimation in Monocular Videos (AAAI 2021 · similarity 0.87)
- ArtTrack: Articulated Multi-Person Tracking in the Wild (CVPR 2017 · similarity 0.87)
- COMPOSE: Hypergraph Cover Optimization for Multi-view 3D Human Pose Estimation (NeurIPS 2026 · similarity 0.87)
- GRPose: Learning Graph Relations for Human Image Generation with Pose Priors (AAAI 2025 · similarity 0.86)
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.