Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
End-to-End One-Shot Human Parsing
Haoyu He, Jing Zhang, Bohan Zhuang, Jianfei Cai, Dacheng Tao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 NPM to AGM with equal fixed weight (DML without weight shifting) (ablation) | AGM-only EOP-Net · ATR-OS Fold 1 | C_novel mIoU: 0.4 -> 13.1 | Improved |
| add weight-shifting strategy to DML (ablation) | DML w/o WS · ATR-OS Fold 1 | C_novel mIoU: 13.1 -> 21.6 | Improved |
| add weight-shifting strategy to DML (ablation) | DML w/o WS · ATR-OS Fold 1 | C_human mIoU: 33.3 -> 58.6 | Improved |
| add coarse-grained metric space (end-to-end framework) (ablation) | DML · ATR-OS Fold 1 | C_novel mIoU: 21.6 -> 30.2 | Improved |
| add prototype-level contrastive loss (ablation) | DML + CG & FG · ATR-OS Fold 1 | C_novel mIoU: 30.2 -> 31.1 | Improved |
| use end-to-end instead of progressive framework (ablation) | POPNet-style framework · ATR-OS Fold 1 | C_human mIoU: 55.3 -> 61.1 | Improved |
| use NPM only instead of AGM only (ablation) | EOP-Net · ATR-OS Fold 1 | C_human mIoU: 51.6 -> 16.5 | Hurt |
| raise momentum coefficient alpha from 0.8 to 0.999 (ablation) | momentum-updated prototypes · ATR-OS Fold 1 | C_novel mIoU: 25.7 -> 31.1 | Improved |
| raise momentum coefficient alpha from 0.999 to 0.9999 (ablation) | momentum-updated prototypes · ATR-OS Fold 1 | C_novel mIoU: 31.1 -> 31.0 | No clear effect |
| use 62-layer instead of 74-layer embedding network (ablation) | EOP-Net · ATR-OS Fold 1 | C_novel mIoU: 30.4 -> 31.1 | Improved |
| replace progressive POPNet with end-to-end EOP-Net | POPNet · CIHP-OS k-way | C_novel mIoU: 17.7 -> 22.8 | Improved |
| replace POPNet with EOP-Net in 1-way setting | POPNet · CIHP-OS 1-way | C_novel mIoU: 35.7 -> 30.9 | Hurt |
Closest papers by meaning
- Progressive One-shot Human Parsing (AAAI 2021 · similarity 0.95)
- Deep Learning Technique for Human Parsing: A Survey and Outlook (IJCV 2024 · similarity 0.90)
- Look Into Person: Self-Supervised Structure-Sensitive Learning and a New Benchmark for Human Parsing (CVPR 2017 · similarity 0.90)
- Part-Aware Context Network for Human Parsing (CVPR 2020 · similarity 0.89)
- Devil in the Details: Towards Accurate Single and Multiple Human Parsing (AAAI 2019 · similarity 0.89)
- Instance-level Human Parsing via Part Grouping Network (ECCV 2018 · similarity 0.88)
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.
