Research map / Learning theory and algorithms / Clustering
Deep Density Clustering of Unconstrained Faces
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 |
|---|---|---|---|
| add negative set mining (ablation) | DDC · YTF | BCubed F-measure: 0.906 -> 0.919 | Helped |
| add negative set mining (ablation) | DDC · LFW | BCubed F-measure: 0.943 -> 0.955 | Helped |
| add negative set mining (ablation) | DDC · IJB-B-128 | BCubed F-measure: 0.810 -> 0.829 | Helped |
| add negative set mining (ablation) | DDC · IJB-B-256 | BCubed F-measure: 0.788 -> 0.816 | Helped |
| add negative set mining (ablation) | DDC · IJB-B-1024 | BCubed F-measure: 0.723 -> 0.751 | Helped |
| add negative set mining (ablation) | DDC · IJB-B-1845 | BCubed F-measure: 0.725 -> 0.746 | Helped |
| use density-based neighborhood similarity instead of pointwise cosine linkage | DDC · IJB-B-1024 | BCubed F-measure: 0.736 -> 0.723 | Hurt |
| use density-based neighborhood similarity instead of pointwise cosine linkage | DDC · IJB-B-1845 | BCubed F-measure: 0.729 -> 0.725 | Hurt |
| use density-based neighborhood similarity instead of pointwise cosine linkage | DDC · YTF | BCubed F-measure: 0.923 -> 0.906 | Hurt |
| use density-based neighborhood similarity instead of pointwise cosine linkage | DDC · LFW | BCubed F-measure: 0.940 -> 0.943 | Helped |
| use density-based neighborhood similarity instead of pointwise cosine linkage | DDC · IJB-B-128 | BCubed F-measure: 0.824 -> 0.810 | Hurt |
| use density-based neighborhood similarity instead of pointwise cosine linkage | DDC · IJB-B-256 | BCubed F-measure: 0.805 -> 0.788 | Hurt |
Closest papers by meaning
- Density-Aware Feature Embedding for Face Clustering (CVPR 2020 · similarity 0.89)
- Linkage Based Face Clustering via Graph Convolution Network (CVPR 2019 · similarity 0.87)
- Learning to Cluster Faces on an Affinity Graph (CVPR 2019 · similarity 0.87)
- Local Connectivity-Based Density Estimation for Face Clustering (CVPR 2023 · similarity 0.87)
- Structure-Aware Face Clustering on a Large-Scale Graph With 107 Nodes (CVPR 2021 · similarity 0.86)
- Learning to Cluster Faces via Confidence and Connectivity Estimation (CVPR 2020 · similarity 0.86)
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.