Exploring the Structure of the Similarity Score Space

Similarity scores in face recognition represent the proximity between pairs of images as computed by a matching algorithm. Given a large set of images and the proximities between all pairs, a similarity score space is defined. Cluster analysis was applied to the similarity score space to develop various taxonomies. Given the number of subjects in the dataset, we used hierarchical methods to aggregate images of the same subject. We also explored the hierarchies occurring above and below this level, including clustering by gender and ethnicity. Furthermore, image specific error rates (a tool for measuring the usefulness of iris images within a dataset) were extended to face images. Observations on the effectiveness of this adaption are presented.