Research map / Graph learning / Graph self-supervised learning
Self-Learning With Rectification Strategy for Human Parsing
Check 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 |
|---|---|---|---|
| Use graph-rectified pseudo-labels for retraining with GSM and LCM | CE2P · LIP, 1/8 labeled | mIoU increase: +3.00% | Helped |
| Use graph-rectified pseudo-labels for retraining with GSM and LCM | CE2P · LIP, 1/4 labeled | mIoU increase: +5.54% | Helped |
| Use graph-rectified pseudo-labels for retraining with GSM and LCM | CE2P · LIP, 1/2 labeled | mIoU increase: +3.99% | Helped |
| Retrain using unrectified pseudo-labels (ablation) | CE2P · LIP, 1/8, 1/4, and 1/2 labeled | mIoU increase: {0.85%, 1.60%, 0.84%} | No clear effect |
| Add GSM to retraining (ablation) | CE2P · LIP, 1/8, 1/4, and 1/2 labeled | mIoU increase: {2.53%, 5.28%, 3.56%} | Helped |
| Add LCM to retraining (ablation) | CE2P · LIP, 1/8, 1/4, and 1/2 labeled | mIoU increase: {2.39%, 5.18%, 2.85%} | Helped |
| Add GSM and LCM to retraining | CE2P · ATR, 1/2 labeled | mIoU increase: 4.42% | Helped |
| Add GSM to retraining (ablation) | CE2P · ATR, 1/2 labeled | mIoU increase: 3.67% | Helped |
| Add LCM to retraining (ablation) | CE2P · ATR, 1/2 labeled | mIoU increase: 3.23% | Helped |
| Add GSM to the supervised segmentation model (ablation) | CE2P · LIP validation | mIoU: 3.94% | Helped |
| Add LCM to the supervised segmentation model (ablation) | CE2P · LIP validation | mIoU: 3.85% | Helped |
| Add GSM and LCM to the supervised segmentation model (ablation) | PSPNet · LIP validation | mIoU: 2.51% and 2.63% | Helped |
Closest papers by meaning
- Graphonomy: Universal Human Parsing via Graph Transfer Learning (CVPR 2019 · similarity 0.91)
- Grapy-ML: Graph Pyramid Mutual Learning for Cross-Dataset Human Parsing (AAAI 2020 · similarity 0.88)
- CDGNet: Class Distribution Guided Network for Human Parsing (CVPR 2022 · similarity 0.88)
- Self Iterative Label Refinement via Robust Unlabeled Learning (NeurIPS 2025 · similarity 0.84)
- SSTAG: Structure-Aware Self-Supervised Learning Method for Text-Attributed Graphs (NeurIPS 2025 · similarity 0.84)
- Graph Parsing Networks (ICLR 2024 · similarity 0.83)
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.