InclusiveFaceNet: Improving Face Attribute Detection with Race and Gender Diversity
This addresses fairness and privacy issues in face recognition for diverse populations, but is incremental as it builds on existing methods.
The paper tackled face attribute detection by learning demographic representations to improve accuracy across gender and race subgroups, achieving some of the best reported numbers on Faces of the World and CelebA datasets.
We demonstrate an approach to face attribute detection that retains or improves attribute detection accuracy across gender and race subgroups by learning demographic information prior to learning the attribute detection task. The system, which we call InclusiveFaceNet, detects face attributes by transferring race and gender representations learned from a held-out dataset of public race and gender identities. Leveraging learned demographic representations while withholding demographic inference from the downstream face attribute detection task preserves potential users' demographic privacy while resulting in some of the best reported numbers to date on attribute detection in the Faces of the World and CelebA datasets.