Research map / Image generation / Editing and style transfer
Controllable and Guided Face Synthesis for Unconstrained Face Recognition
Feng Liu, Minchul Kim, Anil Jain, Xiaoming Liu
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 FR-model-guided CFSM augmentation | ArcFace (cleaned MS1MV2) · IJB-B | TAR@FAR=1e-5: 87.26 -> 90.95 | Improved |
| add FR-model-guided CFSM augmentation | ArcFace (cleaned MS1MV2) · IJB-C | TAR@FAR=1e-6: 87.24 -> 89.34 | Improved |
| add guided CFSM augmentation | ArcFace · IJB-S V2V | Rank-1: 14.81 -> 21.38 | Improved |
| add guided CFSM augmentation | ArcFace · IJB-S V2S | Rank-1: 58.78 -> 63.86 | Improved |
| add guided CFSM augmentation | ArcFace · TinyFace | Rank-1: 62.21 -> 63.01 | Improved |
| add guided CFSM augmentation on large-scale WebFace12M | AdaFace IResNet-100 · IJB-S V2V | Rank-1: 36.71 -> 39.14 | Improved |
| add guided CFSM augmentation on large-scale WebFace12M | AdaFace IResNet-100 · TinyFace | Rank-1: 72.29 -> 73.87 | Improved |
| add FR-model guidance (adversarial FGSM on style coefficient) instead of random coefficients (ablation) | CFSM augmentation · IJB-S V2V | Rank-1: 18.14 -> 21.38 | Improved |
| add FR-model guidance instead of random coefficients (ablation) | CFSM augmentation · IJB-S V2B | Rank-1: 62.20 -> 65.95 | Improved |
| add FR-model guidance instead of random coefficients (ablation) | CFSM augmentation · TinyFace | Rank-1: 62.39 -> 63.01 | Improved |
| use random-coefficient CFSM instead of no augmentation (ablation) | ArcFace · IJB-S V2S | TAR@FAR=1%: 40.99 -> 43.99 | Mixed |
| use more diverse unlabeled target dataset (IJB-S > WiderFace > LFW) (ablation) | CFSM · IJB-S | accuracy | Improved |
Closest papers by meaning
- Vec2Face: Scaling Face Dataset Generation with Loosely Constrained Vectors (ICLR 2025 · similarity 0.87)
- Data Synthesis with Diverse Styles for Face Recognition via 3DMM-Guided Diffusion (CVPR 2025 · similarity 0.86)
- HyperFace: Generating Synthetic Face Recognition Datasets by Exploring Face Embedding Hypersphere (ICLR 2025 · similarity 0.86)
- UIFace: Unleashing Inherent Model Capabilities to Enhance Intra-Class Diversity in Synthetic Face Recognition (ICLR 2025 · similarity 0.85)
- CemiFace: Center-based Semi-hard Synthetic Face Generation for Face Recognition (NeurIPS 2024 · similarity 0.85)
- GIF: Generative Inspiration for Face Recognition at Scale (CVPR 2025 · 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.
