AI Paper Atlas

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

Jointly De-biasing Face Recognition and Demographic Attribute Estimation

Sixue Gong, Xiaoming Liu, Anil K. Jain

ECCV 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Fair and quality-aware face recognition · 144 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
add adversarial disentanglement (DebFace) to the baseline face encoderface verification · Verification biasness - Allbiasness: 6.83 -> 5.07Improved
add adversarial disentanglement (DebFace) to the baselineface verification · Verification biasness - Genderbiasness: 0.50 -> 0.15Improved
add adversarial disentanglement (DebFace) to the baselineface verification · Verification biasness - Agebiasness: 3.13 -> 1.83Improved
add adversarial disentanglement (DebFace) to the baselineface verification · Verification biasness - Racebiasness: 5.49 -> 3.70Improved
add joint adversarial de-biasing to the gender estimatordemographic estimation · Gender estimation biasnessbiasness: 12.38 -> 10.22Improved
add joint adversarial de-biasing to the age estimatordemographic estimation · Age estimation biasnessbiasness: 10.83 -> 7.61Improved
add joint adversarial de-biasing to the race estimatordemographic estimation · Race estimation biasnessbiasness: 14.58 -> 10.00Improved
use de-biased DebFace-ID features instead of BaseFace featuresface verification · LFWaccuracy (%): 99.38 -> 98.97Hurt
use de-biased DebFace-ID features instead of BaseFace featuresface verification · IJB-ATAR @ 0.1% FAR (%): 90.2 ± 1.1 -> 87.6 ± 0.9Hurt
use DebFace-ID instead of BaseFaceface verification · IJB-CTAR @ 0.001% FAR (%): 80.2 -> 82.0Mixed
aggregate gender and race features with DebFace-ID (DemoID) instead of DebFace-ID aloneface verification · LFWaccuracy (%): 98.97 -> 99.50Improved
aggregate gender and race features with DebFace-ID (DemoID) instead of DebFace-ID aloneface verification · IJB-ATAR @ 0.1% FAR (%): 87.6 ± 0.9 -> 92.2 ± 0.8Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 2 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