Research map / Graph learning / Knowledge and temporal graphs
Edge-aware Graph Representation Learning and Reasoning for Face Parsing
Gusi Te, Yinglu Liu, Wei Hu, Hailin Shi, Tao Mei
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 edge-aware pooling projection (ablation) | baseline · Helen | mIoU: 76.5 -> 77.5 | Improved |
| add graph projection and reasoning (ablation) | baseline · Helen | mIoU: 76.5 -> 77.3 | Improved |
| add graph projection and reasoning (ablation) | baseline · Helen | F1-score: 91.4 -> 92.3 | Improved |
| add edge attention to graph model (ablation) | Graph+Reasoning · Helen | mIoU: 77.3 -> 77.8 | Improved |
| add edge attention to graph model (ablation) | Graph+Reasoning · Helen | pixel accuracy: 85.8 -> 84.6 | Hurt |
| add boundary-attention loss (ablation) | Edge+Graph+Reasoning · Helen | mIoU: 77.8 -> 78.2 | Improved |
| add boundary-attention loss (ablation) | Edge+Graph+Reasoning · Helen | pixel accuracy: 84.6 -> 87.3 | Improved |
| add graph reasoning (GCN) on top of edge+graph with BA-loss (ablation) | EAGRNet · Helen | mIoU: 77.3 -> 78.2 | Improved |
| apply EAGR to low and high-level features separately (2 modules) versus none (ablation) | EAGRNet · Helen | mIoU: 77.6 -> 78.2 | Improved |
| add a third EAGR module (3 vs 2 modules) (ablation) | EAGRNet · Helen | mIoU: 78.2 -> 77.4 | Hurt |
| use single EAGR module on concatenated features versus 0 modules (ablation) | EAGRNet · Helen | mIoU: 77.6 -> 77.6 | No clear effect |
| increase pooling size from 6x6 to 8x8 (ablation) | EAGRNet · Helen | mIoU: 78.2 -> 78.0 | Hurt |
Closest papers by meaning
- A New Dataset and Boundary-Attention Semantic Segmentation for Face Parsing (AAAI 2020 · similarity 0.89)
- Face Parsing With RoI Tanh-Warping (CVPR 2019 · similarity 0.89)
- Decoupled Multi-Task Learning With Cyclical Self-Regulation for Face Parsing (CVPR 2022 · similarity 0.88)
- Structure-Aware Face Clustering on a Large-Scale Graph With 107 Nodes (CVPR 2021 · similarity 0.88)
- SegFace: Face Segmentation of Long-Tail Classes (AAAI 2025 · similarity 0.86)
- Self-Supervised Facial Representation Learning with Facial Region Awareness (CVPR 2024 · similarity 0.85)
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.
