| Popis: |
Investigations that require the exploitation of large volumes of face imagery are increasingly common in current forensic scenarios (e.g., Boston Marathon bombing), but effective solutions for triaging such imagery (i.e., low importance, moderate importance, and of critical interest) are not available in the literature. General issues for investigators in these scenarios are a lack of systems that can scale to volumes of images of the order of a few million, and a lack of established methods for clustering the face images into the unknown number of persons of interest contained in the collection. As such, we explore best practices for clustering large sets of face images (up to 1 million here) into large numbers of clusters (approximately 200 thousand) as a method of reducing the volume of data to be investigated by forensic analysts. Our analysis involves a performance comparison of several clustering algorithms in terms of the accuracy of grouping face images by identity, run-time, and efficiency in representing large datasets of face images in terms of compact and isolated clusters. For two different face datasets, a mugshot database (PCSO) and the well known unconstrained dataset, LFW, we find the rank-order clustering method to be effective in clustering accuracy, and relatively efficient in terms of run-time. |
