核心概念
PrivatEyes introduces a novel approach for privacy-enhancing gaze estimation using federated learning and secure multi-party computation.
統計資料
PrivatEyes는 개인의 시선 데이터를 보호하면서 시선 추정 모델을 훈련하는 혁신적인 방법을 소개합니다.
引述
"PrivatEyes guarantees that individual gaze data remains private even if a majority of the aggregating servers is malicious."
"Evaluations show that the improved privacy of PrivatEyes does not lead to a lower gaze estimation accuracy or substantially higher computational costs."