Demonstration
Real-Time Adaptive Face Recognition
Branislav Kveton
Abstract:
High-accuracy face recognition with a sufficiently high recall is a challenging problem. This demo shows how to achieve this efficiently with just a few labeled examples, streams of unlabeled data, and state-of-the-art machine learning techniques. In particular, we show an online semi-supervised learning algorithm that tracks the manifold of human faces in real time and based on it predicts whether the observed face belongs to one of the labeled individuals or not.
Chat is not available.
Successful Page Load