Research map / Label-efficient and robust learning / Recognition and long-tail
Density-Aware Feature Embedding for Face Clustering
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 |
|---|---|---|---|
| Add local clique feature aggregation to the original features (ablation) | DA-Net · MS-Celeb-1M | F-score: 72.04 -> 86.51 | Helped |
| Add long-range density-chain feature aggregation to the original features (ablation) | DA-Net · MS-Celeb-1M | F-score: 72.04 -> 81.72 | Helped |
| Combine local clique aggregation with long-range density-chain aggregation (ablation) | DA-Net · MS-Celeb-1M | F-score: 86.51 -> 90.60 | Helped |
| Add the long-range LSTM to a two-layer GCN (ablation) | DA-Net · MS-Celeb-1M | F-score: 86.51 -> 90.60 | Helped |
| Increase GCN depth from two to four layers without LSTM (ablation) | DA-Net · MS-Celeb-1M | F-score: 86.51 -> 86.80 | No clear effect |
| Increase both GCN and LSTM depth from two to four layers (ablation) | DA-Net · MS-Celeb-1M | F-score: 90.60 -> 91.10 | Helped |
| Use DA-Net features instead of original features | Face clustering · YouTube Faces DB | F-score: 92.04 -> 95.10 | Helped |
| Use DA-Net features instead of original features | Face clustering · IJB-B (F1845) | F-score: 0.758 -> 0.828 | Helped |
| Use DA-Net features rather than original features with K-Means | K-Means · MS-Celeb-1M | F-score: 60.18 -> 89.53 | Helped |
| Use DA-Net features rather than original features with DBSCAN | DBSCAN · MS-Celeb-1M | F-score: 53.50 -> 87.63 | Helped |
Closest papers by meaning
- Linkage Based Face Clustering via Graph Convolution Network (CVPR 2019 · similarity 0.89)
- Learning to Cluster Faces via Confidence and Connectivity Estimation (CVPR 2020 · similarity 0.89)
- Deep Density Clustering of Unconstrained Faces (CVPR 2018 · similarity 0.89)
- Learning to Cluster Faces on an Affinity Graph (CVPR 2019 · similarity 0.88)
- Local Connectivity-Based Density Estimation for Face Clustering (CVPR 2023 · similarity 0.87)
- FC-DGCN: Deep Graph Convolutional Network for Face Clustering and Recognition (NeurIPS 2026 · similarity 0.87)
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.