To Impute or Not: Recommendations for Multibiometric Fusion

TL;DR

This paper evaluates different score imputation methods for multibiometric fusion, demonstrating that imputation generally improves recognition accuracy and highlighting the importance of class balancing and score correlation considerations.

Contribution

It provides a comprehensive assessment of score imputation techniques across multiple datasets and identifies key factors influencing their effectiveness.

Findings

Imputation outperforms no imputation even with incomplete data.

Balancing training classes reduces bias in imputation.

Multivariate methods work better with correlated scores.

Abstract

Combining match scores from different biometric systems via fusion is a well-established approach to improving recognition accuracy. However, missing scores can degrade performance as well as limit the possible fusion techniques that can be applied. Imputation is a promising technique in multibiometric systems for replacing missing data. In this paper, we evaluate various score imputation approaches on three multimodal biometric score datasets, viz. NIST BSSR1, BIOCOP2008, and MIT LL Trimodal, and investigate the factors which might influence the effectiveness of imputation. Our studies reveal three key observations: (1) Imputation is preferable over not imputing missing scores, even when the fusion rule does not require complete score data. (2) Balancing the classes in the training data is crucial to mitigate negative biases in the imputation technique towards the under-represented class, even if it involves dropping a substantial number of score vectors. (3) Multivariate imputation approaches seem to be beneficial when scores between modalities are correlated, while univariate approaches seem to benefit scenarios where scores between modalities are less correlated.