Learning from a single labeled face and a stream of unlabeled data

2026-04-30

AI Summary

This work addresses the challenging scenario of single-sample-per-person face recognition—common in personal device authentication—where only one labeled image per individual is available and no negative samples from other identities exist. The problem is formulated as a one-class classification task, and the study introduces a novel approach that leverages a continuous stream of unlabeled data to enhance model performance under this extreme data scarcity. By adopting a non-parametric modeling strategy, the method enables effective learning without requiring negative examples and provides practical guidelines for parameter selection. Evaluated on a dataset of 43 subjects, the proposed approach achieves a 90% identification rate with near-zero false positives, yielding a recall improvement of over 25% compared to the strongest baseline.

Abstract

Face recognition from a single image per person is a challenging problem because the training sample is extremely small. We consider a variation of this problem. In our problem, we recognize only one person, and there are no labeled data for any other person. This setting naturally arises in authentication on personal computers and mobile devices, and poses additional challenges because it lacks negative examples. We formalize our problem as one-class classification, and propose and analyze an algorithm that learns a non-parametric model of the face from a single labeled image and a stream of unlabeled data. In many domains, for instance when a person interacts with a computer with a camera, unlabeled data are abundant and easy to utilize. This is the first paper that investigates how these data can help in learning better models in the single-image-per-person setting. Our method is evaluated on a dataset of 43 people and we show that these people can be recognized 90% of time at nearly zero false positives. This recall is 25+% higher than the recall of our best performing baseline. Finally, we conduct a comprehensive sensitivity analysis of our algorithm and provide a guideline for setting its parameters in practice.

Problem

• one-class classification
• face recognition
• single image per person
• unlabeled data
• authentication

Innovation

• one-class classification
• single-sample face recognition
• unlabeled data stream
• non-parametric modeling
• personal authentication