AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Stylized-Face: A Million-level Stylized Face Dataset for Face Recognition

Zhengyuan Peng, Jianqing Xu, Yuge Huang, Jinkun Hao, Shouhong Ding, Zhizhong Zhang, Xin Tan, Lizhuang Ma

ICCV 2025 · Label-efficient and robust learning › Recognition and long-tail · approach: Fair and quality-aware face recognition · 1 citations (OpenAlex)

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 changedWhereMeasuredEffect
train on Stylized-Face instead of MS1MV3IResNet-50 face recognition · Standard stylized face recognitionTAR@FAR=1e-4: 49.7 -> 90.6Improved
train on Stylized-Face instead of Glint360kIResNet-50 face recognition · Standard stylized face recognitionTAR@FAR=1e-4: 74.7 -> 90.6Improved
train on Stylized-Face instead of Glint360kIResNet-50 face recognition · Cross-prompt stylized face recognitionTAR@FAR=1e-4: 73.9 -> 90.3Improved
train on Stylized-Face instead of Glint360kIResNet-50 face recognition · Cross-method (all generators)TAR@FAR=1e-3: 58.6 -> 71.9Improved
train on Stylized-Face instead of Glint360kIResNet-50 face recognition · Cross-method Flashface hard promptsTAR@FAR=1e-2: 84.9 -> 87.1Improved
train on Stylized-Face instead of Glint360kIResNet-50 face recognition · Cross-method Flashface easy promptsTAR@FAR=1e-2: 96.9 -> 96.9No clear effect
scale training data from MS1MV3 to larger Glint360kIResNet-50 face recognition · Standard stylized face recognitionTAR@FAR=1e-4: 49.7 -> 74.7Improved
use IResNet-101 instead of IResNet-50 backbone (ablation)Stylized-Face training · Standard stylized face recognitionTAR@FAR=1e-4: 90.6 -> 92.0Improved
use Arcface instead of Cosface (ablation)Stylized-Face training · Standard stylized face recognitionTAR@FAR=1e-4: 90.6 -> 89.9No clear effect
use CurricularFace instead of Cosface (ablation)Stylized-Face training · Standard stylized face recognitionTAR@FAR=1e-4: 90.6 -> 90.5No clear effect
train on Stylized-Face instead of Glint360kIResNet-50 face recognition · Real face recognition averageaccuracy: 99.1 -> 98.7Hurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy