AI Paper Atlas

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

Mitigating Face Recognition Bias via Group Adaptive Classifier

Sixue Gong, Xiaoming Liu, Anil K. Jain

CVPR 2021 · Label-efficient and robust learning › Recognition and long-tail · approach: Fair and quality-aware face recognition · 88 citations (OpenAlex)

Check 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 group-adaptive kernels and attention with automatic layer selectionArcFace-50 · RFWSTD across groups: 0.96 -> 0.58Improved
Use group-adaptive kernels and attention with automatic layer selectionArcFace-50 · RFWAvg verification accuracy (%): 94.64 -> 95.21Improved
Use group-adaptive kernels and attention with automatic layer selectionArcFace-50 · LFWVerification accuracy (%): 99.75 -> 99.78Improved
Use group-adaptive kernels and attention with automatic layer selectionArcFace-50 · IJB-A (0.1% FAR)Verification accuracy (%): 90.2 ± 1.1 -> 91.3 ± 1.2Improved
Use group-adaptive kernels and attention with automatic layer selectionArcFace-50 · IJB-C (0.001% FAR)Verification accuracy (%): 80.2 -> 83.5Improved
Use group-adaptive kernels and attention with automatic layer selectionArcFace-50 · IJB-C (0.01% FAR)Verification accuracy (%): 88.0 -> 89.2Improved
Use group-adaptive kernels and attention with automatic layer selectionArcFace-50 · IJB-C (0.1% FAR)Verification accuracy (%): 92.9 -> 93.7Improved
Use channel-wise attention only (ablation)GAC · RFWSTD across groups: 1.11 -> 0.83Improved
Use adaptive kernels only (ablation)GAC · RFWSTD across groups: 1.11 -> 0.78Improved
Use spatial attention only (ablation)GAC · RFWAvg verification accuracy (%): 94.64 -> 94.19Hurt
Add spatial attention to the adaptive kernels and channel-wise attention (ablation)GAC · RFWAvg verification accuracy (%): 95.21 -> 94.72Hurt
Set the adaptation threshold to τ=-0.1 instead of τ=-0.2 (ablation)GAC · RFWAvg verification accuracy (%): 95.21 -> 95.01Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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