Google researchers achieved nearly 91% smile-detection accuracy on the Faces of the World (FotW) benchmark by incorporating racial and gender classifiers into their model.
Notes on verification
Confirmed directly by the original Google Research arXiv paper (1712.00193) with precise figures (90.96%), corroborated by independent tech press coverage (The Register, Tech Monitor). [tier=gold indep_score=0.838 clusters=4 claim_tier=notable]
Sources
- Google’s New AI Smile Detector Shows How Embracing Race and Gender Can Reduce Bias (seed:technology_and_ai)
- https://www.theregister.com/software/2017/12/06/google-learns-to-smile-because-ais-bad-at-it/798222 (corroboration)
- https://www.techmonitor.ai/digital-economy/ai-and-automation/google-ai-diversity (corroboration)
- https://arxiv.org/pdf/1712.00193 (corroboration)
- https://research.google/pubs/pub46513/ (corroboration)