Research map / Label-efficient and robust learning / Recognition and long-tail
Pairwise Relational Networks for Face Recognition
Bong-Nam Kang, Yonghyun Kim, Daijin Kim
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 face identity state feature to the PRN (ablation) | PRN · VGGFace2 validation set | classification accuracy: 94.2 -> 96.7 | Improved |
| add PRN (no identity state) to the global CNN feature (ablation) | model A · LFW | accuracy (%): 99.6 -> 99.65 | Improved |
| add identity state feature to the PRN branch (ablation) | model B · LFW | accuracy (%): 99.65 -> 99.76 | Improved |
| add PRN+ to the global CNN feature (ablation) | model A · LFW | accuracy (%): 99.6 -> 99.76 | Improved |
| add PRN (no identity state) to the global CNN feature (ablation) | model A · YTF | accuracy (%): 95.1 -> 95.7 | Improved |
| add identity state feature to the PRN branch (ablation) | model B · YTF | accuracy (%): 95.7 -> 96.3 | Improved |
| add PRN+ to the global CNN feature (ablation) | model A · YTF | accuracy (%): 95.1 -> 96.3 | Improved |
| add identity state feature to the PRN branch (ablation) | model B · IJB-A 1:1 verification FAR=0.001 | TAR: 0.901 -> 0.919 | Improved |
| add PRN+ to the global CNN feature (ablation) | model A · IJB-A 1:1 verification FAR=0.001 | TAR: 0.895 -> 0.919 | Improved |
| add PRN (no identity state) to the global CNN feature (ablation) | model A · IJB-A 1:N identification | Rank-1 TPIR: 0.975 -> 0.976 | No clear effect |
| add identity state feature to the PRN branch (ablation) | model B · IJB-B 1:1 verification FAR=0.001 | TAR: 0.910 -> 0.923 | Improved |
| add PRN+ to the global CNN feature (ablation) | model A · IJB-B 1:1 verification FAR=0.001 | TAR: 0.892 -> 0.923 | Improved |
Closest papers by meaning
- Attentional Feature-Pair Relation Networks for Accurate Face Recognition (ICCV 2019 · similarity 0.89)
- CoupleFace: Relation Matters for Face Recognition Distillation (ECCV 2022 · similarity 0.86)
- Hybrid Deep Learning for Face Verification (TPAMI 2016 · similarity 0.85)
- Pairwise Similarity Learning is SimPLE (ICCV 2023 · similarity 0.85)
- Occlusion Robust Face Recognition Based on Mask Learning With Pairwise Differential Siamese Network (ICCV 2019 · similarity 0.85)
- Learn To Cluster Faces via Pairwise Classification (ICCV 2021 · 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.
