Face Identification and Clustering

TL;DR

This thesis explores how visual attributes and clustering strategies can enhance face identification and template aggregation, demonstrating performance improvements on challenging face recognition datasets.

Contribution

It introduces a clustering-based approach leveraging visual attributes and analyzes the impact of cluster quantity on face recognition performance.

Findings

Adding more visual attributes improves clustering accuracy.

Optimal number of clusters maximizes performance before degradation.

Experiments validate methods on IJB-A, CS2, and CS3 datasets.

Abstract

In this thesis, we study two problems based on clustering algorithms. In the first problem, we study the role of visual attributes using an agglomerative clustering algorithm to whittle down the search area where the number of classes is high to improve the performance of clustering. We observe that as we add more attributes, the clustering performance increases overall. In the second problem, we study the role of clustering in aggregating templates in a 1:N open set protocol using multi-shot video as a probe. We observe that by increasing the number of clusters, the performance increases with respect to the baseline and reaches a peak, after which increasing the number of clusters causes the performance to degrade. Experiments are conducted using recently introduced unconstrained IARPA Janus IJB-A, CS2, and CS3 face recognition datasets.