A non-parametric algorithm recognizes one person from a single labeled face image and unlabeled data stream, achieving 90% recall at near-zero false positives on 43 people, 25% above baselines.
Semi-supervised learning using Gaussian fields and harmonic functions,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Learning from a single labeled face and a stream of unlabeled data
A non-parametric algorithm recognizes one person from a single labeled face image and unlabeled data stream, achieving 90% recall at near-zero false positives on 43 people, 25% above baselines.