Research map / Label-efficient and robust learning / Recognition and long-tail
RegularFace: Deep Face Recognition via Exclusive Regularization
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 exclusive regularization to SphereFace | SphereFace · LFW (WebFace) | accuracy: 99.26 -> 99.33 | Helped |
| Add exclusive regularization to SphereFace | SphereFace · YTF (WebFace) | accuracy: 94.1 -> 94.4 | Helped |
| Add exclusive regularization to SphereFace | SphereFace · LFW (VGGFace2) | accuracy: 99.50 -> 99.61 | Helped |
| Add exclusive regularization to SphereFace | SphereFace · YTF (VGGFace2) | accuracy: 95.9 -> 96.7 | Helped |
| Add exclusive regularization to SphereFace | SphereFace · MegaFace Challenge 1 FaceScrub (Small) | Rank1 Acc: 69.62 -> 70.23 | Helped |
| Add exclusive regularization to SphereFace | SphereFace · MegaFace Challenge 1 FaceScrub (Small) | Ver.: 83.16 -> 84.07 | Helped |
| Add exclusive regularization to SphereFace | SphereFace · MegaFace Challenge 1 FaceScrub (Large) | Rank1 Acc: 74.82 -> 75.61 | Helped |
| Add exclusive regularization to SphereFace | SphereFace · MegaFace Challenge 1 FaceScrub (Large) | Ver.: 89.01 -> 91.13 | Helped |
| Use an annealing warm-up for the exclusive-regularization weight (ablation) | RegularFace · Training loss stability | exclusive loss stability | Helped |
| Set the exclusive-regularization weight to λ = 6 (ablation) | RegularFace · LFW | accuracy | Helped |
Closest papers by meaning
- Towards Universal Representation Learning for Deep Face Recognition (CVPR 2020 · similarity 0.88)
- Vec2Face: Scaling Face Dataset Generation with Loosely Constrained Vectors (ICLR 2025 · similarity 0.87)
- Masked Face Recognition with Generative-to-Discriminative Representations (ICML 2024 · similarity 0.87)
- UIFace: Unleashing Inherent Model Capabilities to Enhance Intra-Class Diversity in Synthetic Face Recognition (ICLR 2025 · similarity 0.87)
- HyperFace: Generating Synthetic Face Recognition Datasets by Exploring Face Embedding Hypersphere (ICLR 2025 · similarity 0.86)
- $\text{ID}^3$: Identity-Preserving-yet-Diversified Diffusion Models for Synthetic Face Recognition (NeurIPS 2024 · 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.