Research map / Graph learning / Graph self-supervised learning
Graphonomy: Universal Human Parsing via Graph Transfer Learning
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 |
|---|---|---|---|
| Transfer the CIHP semantic graph to PASCAL-Person-Part | DeepLab v3+ · PASCAL-Person-Part test | Mean IoU(%): 67.84 -> 71.14 | Helped |
| Transfer the PASCAL-Person-Part semantic graph to ATR | DeepLab v3+ · ATR test | Overall accuracy (%): 97.30 -> 98.32 | Helped |
| Transfer the PASCAL-Person-Part semantic graph to ATR | DeepLab v3+ · ATR test | F-1 score (%): 84.50 -> 90.89 | Helped |
| Transfer the PASCAL-Person-Part semantic graph to CIHP | DeepLab v3+ · CIHP test | Mean IoU(%): 57.13 -> 58.58 | Helped |
| Add Intra-Graph Reasoning without the adjacency matrix (ablation) | DeepLab v3+ on PASCAL-Person-Part · PASCAL-Person-Part test | Mean IoU(%): 67.84 -> 67.89 | No clear effect |
| Add the human-structure adjacency matrix to Intra-Graph Reasoning (ablation) | DeepLab v3+ on PASCAL-Person-Part · PASCAL-Person-Part test | Mean IoU(%): 67.89 -> 68.34 | Helped |
| Add Intra-Graph Reasoning and adjacency after CIHP pretraining (ablation) | DeepLab v3+ on PASCAL-Person-Part · PASCAL-Person-Part test | Mean IoU(%): 70.33 -> 70.47 | Helped |
| Use handcraft relation for Inter-Graph Transfer (ablation) | Graphonomy on PASCAL-Person-Part · PASCAL-Person-Part test | Mean IoU(%): 70.47 -> 70.22 | Hurt |
| Use feature similarity for Inter-Graph Transfer (ablation) | Graphonomy on PASCAL-Person-Part · PASCAL-Person-Part test | Mean IoU(%): 70.47 -> 71.05 | Helped |
| Use semantic similarity for Inter-Graph Transfer (ablation) | Graphonomy on PASCAL-Person-Part · PASCAL-Person-Part test | Mean IoU(%): 70.47 -> 70.95 | Helped |
| Combine feature similarity and semantic similarity for Inter-Graph Transfer (ablation) | Graphonomy on PASCAL-Person-Part · PASCAL-Person-Part test | Mean IoU(%): 71.05 -> 71.14 | Helped |
| Add learnable-matrix transfer to feature and semantic similarity (ablation) | Graphonomy on PASCAL-Person-Part · PASCAL-Person-Part test | Mean IoU(%): 71.14 -> 70.87 | Hurt |
Closest papers by meaning
- Semantic Human Parsing via Scalable Semantic Transfer Over Multiple Label Domains (CVPR 2023 · similarity 0.91)
- Self-Learning With Rectification Strategy for Human Parsing (CVPR 2020 · similarity 0.91)
- Hierarchical Human Parsing With Typed Part-Relation Reasoning (CVPR 2020 · similarity 0.89)
- Grapy-ML: Graph Pyramid Mutual Learning for Cross-Dataset Human Parsing (AAAI 2020 · similarity 0.89)
- SSTAG: Structure-Aware Self-Supervised Learning Method for Text-Attributed Graphs (NeurIPS 2025 · similarity 0.84)
- UniGTE: Unified Graph–Text Encoding for Zero-Shot Generalization across Graph Tasks and Domains (NeurIPS 2025 · 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.