AI Paper Atlas

Research map / Image generation / Editing and style transfer

Anti-Makeup: Learning A Bi-Level Adversarial Network for Makeup-Invariant Face Verification

Yi Li, Lingxiao Song, Xiang Wu, Ran He, Tieniu Tan

AAAI 2018 · Image generation › Editing and style transfer · approach: Identity-preserving face manipulation · 52 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
Use BLAN to generate non-makeup faces before Light CNN feature extraction, instead of matching Light CNN features from the original imagesLight CNN · Dataset 1rank-1 accuracy (%): 92.4 -> 94.8Improved
Use BLAN to generate non-makeup faces before Light CNN feature extraction, instead of matching Light CNN features from the original imagesLight CNN · Dataset 2rank-1 accuracy (%): 91.5 -> 92.3Improved
Use BLAN to generate non-makeup faces before Light CNN feature extraction, instead of matching Light CNN features from the original imagesLight CNN · FAMrank-1 accuracy (%): 86.3 -> 88.1Improved
Add edge loss (ablation)BLAN · Dataset 1rank-1 accuracy (%): 92.9 -> 94.8Improved
Add symmetry loss (ablation)BLAN · Dataset 1rank-1 accuracy (%): 91.8 -> 94.8Improved
Add feature-level adversarial loss (ablation)BLAN · Dataset 1rank-1 accuracy (%): 89.6 -> 94.8Improved
Add feature reconstruction loss (ablation)BLAN · Dataset 1rank-1 accuracy (%): 76.5 -> 94.8Improved

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