A Bayesian algorithm for detecting identity matches and fraud in image databases

TL;DR

This paper introduces a Bayesian statistical algorithm designed to identify identity matches and detect fraud in image databases by modeling relationships between images and identities.

Contribution

The paper presents a novel Bayesian generative model that improves detection of matches and fraud in image databases over previous methods.

Findings

Effective detection of identity matches demonstrated.

Improved fraud detection accuracy shown.

Model outperforms existing algorithms in experiments.

Abstract

A statistical algorithm for categorizing different types of matches and fraud in image databases is presented. The approach is based on a generative model of a graph representing images and connections between pairs of identities, trained using properties of a matching algorithm between images.