Research map / Detection and segmentation / CNN features
Decoupled Multi-Task Learning With Cyclical Self-Regulation for Face Parsing
Qingping Zheng, Jiankang Deng, Zheng Zhu, Ying Li, Stefanos Zafeiriou
Check 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 DDGCN context embedding without pooling (ablation) | Baseline face parsing model · Helen | Overall F1: 92.11 -> 92.56 | Improved |
| Add DDGCN context embedding without pooling (ablation) | Baseline face parsing model · CelebAMask-HQ | Mean F1: 84.34 -> 85.33 | Improved |
| Add DDGCN context embedding without pooling (ablation) | Baseline face parsing model · LaPa | Mean F1: 89.84 -> 92.10 | Improved |
| Add decoupled multi-task edge learning (ablation) | DDGCN face parsing model · Helen | Overall F1: 92.56 -> 93.75 | Improved |
| Add decoupled multi-task edge learning (ablation) | DDGCN face parsing model · CelebAMask-HQ | Mean F1: 85.33 -> 85.98 | Improved |
| Add decoupled multi-task edge learning (ablation) | DDGCN face parsing model · LaPa | Mean F1: 92.10 -> 92.32 | Improved |
| Add cyclical self-regulation (ablation) | DDGCN + DML face parsing model · Helen | Overall F1: 93.75 -> 93.78 | Improved |
| Add cyclical self-regulation (ablation) | DDGCN + DML face parsing model · CelebAMask-HQ | Mean F1: 85.98 -> 86.07 | Improved |
| Add cyclical self-regulation (ablation) | DDGCN + DML face parsing model · LaPa | Mean F1: 92.32 -> 92.38 | Improved |
| Replace pooled context with pooling-removed PSP (ablation) | Baseline face parsing model · Helen | Overall F1: 92.20 -> 92.37 | Improved |
| Replace pooled context with pooling-removed DGCNet (ablation) | Baseline face parsing model · Helen | Overall F1: 92.41 -> 92.45 | Improved |
| Add binary edge branch (ablation) | DDGCN face parsing model · CelebAMask-HQ | Mean F1: 84.34 -> 85.58 | Improved |
Closest papers by meaning
- Edge-aware Graph Representation Learning and Reasoning for Face Parsing (ECCV 2020 · similarity 0.88)
- A New Dataset and Boundary-Attention Semantic Segmentation for Face Parsing (AAAI 2020 · similarity 0.86)
- Face Parsing With RoI Tanh-Warping (CVPR 2019 · similarity 0.86)
- SegFace: Face Segmentation of Long-Tail Classes (AAAI 2025 · similarity 0.85)
- Distribution Matching for Multi-Task Learning of Classification Tasks: A Large-Scale Study on Faces & Beyond (AAAI 2024 · similarity 0.84)
- GroundingFace: Fine-grained Face Understanding via Pixel Grounding Multimodal Large Language Model (CVPR 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.
