Research map / 3D vision / Human pose estimation
Pose-Guided Human Parsing by an AND/OR Graph Using Pose-Context Features
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 pose-context feature to appearance features (ablation) | SVR part selection · Penn-Fudan top-1 AOI | mean AOI (%): 45.0 -> 48.1 | Helped |
| add non-linearly coded pose-context feature (ablation) | SVR part selection · Penn-Fudan top-1 AOI | mean AOI (%): 48.1 -> 53.8 | Helped |
| add FCN deep semantic feature (ablation) | SVR part selection · Penn-Fudan top-1 AOI | mean AOI (%): 53.8 -> 56.9 | Helped |
| add pose features and FCN feature (ablation) | SVR part selection · Penn-Fudan top-10 AOI | mean AOI (%): 64.6 -> 66.7 | Helped |
| pose-guided proposals instead of standard RIGOR (ablation) | part segment proposal generation · Penn-Fudan | APR and AOI: over 10% on average | Helped |
| use basic AOG with parent-child terms instead of naive assembling (ablation) | part assembling · Penn-Fudan | average per-pixel accuracy (%): 56.2 -> 57.5 | Helped |
| add side-way pairwise terms to AOG (ablation) | basic AOG · Penn-Fudan | average per-pixel accuracy (%): 57.5 -> 60.5 | Helped |
| add state pruning in inference (ablation) | AOG · Penn-Fudan | average per-pixel accuracy (%): 60.5 -> 60.5 | No clear effect |
| pose-guided AOG parsing instead of FCN | human parsing · Penn-Fudan | per-pixel accuracy: over 10% | Helped |
Closest papers by meaning
- Look Into Person: Self-Supervised Structure-Sensitive Learning and a New Benchmark for Human Parsing (CVPR 2017 · similarity 0.91)
- Weakly and Semi Supervised Human Body Part Parsing via Pose-Guided Knowledge Transfer (CVPR 2018 · similarity 0.90)
- Correlating Edge, Pose With Parsing (CVPR 2020 · similarity 0.89)
- CDGNet: Class Distribution Guided Network for Human Parsing (CVPR 2022 · similarity 0.88)
- A Pedestrian is Worth One Prompt: Towards Language Guidance Person Re-Identification (CVPR 2024 · similarity 0.85)
- Referring Human Pose and Mask Estimation In the Wild (NeurIPS 2024 · similarity 0.84)
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.