Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Progressive One-shot Human Parsing
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 |
|---|---|---|---|
| Use POPNet instead of the strongest baseline | ATR-OS k-way · ATR-OS k-way | Novel Class MIoU (Mean): 13.0 -> 21.8 | Helped |
| Use POPNet instead of SG-One | ATR-OS k-way · ATR-OS k-way | Human MIoU (Mean): 44.4 -> 60.5 | Helped |
| Use POPNet instead of AMP | ATR-OS one-way · ATR-OS one-way | Novel Class MIoU (Mean): 8.8 -> 27.7 | Helped |
| Use POPNet instead of SG-One | ATR-OS one-way · ATR-OS one-way | Human MIoU (Mean): 39.8 -> 52.8 | Helped |
| Use AGM and NCM together instead of either module alone (ablation) | POPNet on ATR-OS Fold 1 · ATR-OS Fold 1 | Novel Class MIoU: 1.0, 3.5 -> 15.1 | Helped |
| Use AGM and NCM together instead of either module alone (ablation) | POPNet on ATR-OS Fold 1 · ATR-OS Fold 1 | Human MIoU: 40.2, 40.4 -> 47.9 | Helped |
| Add weight shifting to DML (ablation) | POPNet on ATR-OS Fold 1 · ATR-OS Fold 1 | Novel Class MIoU: 15.1 -> 18.6 | Helped |
| Add weight shifting to DML (ablation) | POPNet on ATR-OS Fold 1 · ATR-OS Fold 1 | Human MIoU: 47.9 -> 47.1 | Hurt |
| Add progressive parsing with KIM (ablation) | POPNet on ATR-OS Fold 1 · ATR-OS Fold 1 | Novel Class MIoU: 18.6 -> 20.0 | Helped |
| Add progressive parsing with KIM (ablation) | POPNet on ATR-OS Fold 1 · ATR-OS Fold 1 | Human MIoU: 47.1 -> 52.5 | Helped |
| Add dynamic prototype generation (ablation) | POPNet on ATR-OS Fold 1 · ATR-OS Fold 1 | Novel Class MIoU: 20.0 -> 24.1 | Helped |
| Add dynamic prototype generation (ablation) | POPNet on ATR-OS Fold 1 · ATR-OS Fold 1 | Human MIoU: 52.5 -> 60.6 | Helped |
Closest papers by meaning
- Part-Aware Context Network for Human Parsing (CVPR 2020 · similarity 0.89)
- Look Into Person: Self-Supervised Structure-Sensitive Learning and a New Benchmark for Human Parsing (CVPR 2017 · similarity 0.88)
- CDGNet: Class Distribution Guided Network for Human Parsing (CVPR 2022 · similarity 0.87)
- Progressive Cognitive Human Parsing (AAAI 2018 · similarity 0.87)
- A Large-Scale Human-Centric Benchmark for Referring Expression Comprehension in the LMM Era (NeurIPS 2024 · similarity 0.83)
- OSLoPrompt: Bridging Low-Supervision Challenges and Open-Set Domain Generalization in CLIP (CVPR 2025 · similarity 0.82)
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.