Incorporating Graph-Based Models in a Deep Learning Framework for Operational Face Recognition

In classical face recognition, an input probe image is compared against a gallery of labeled face images in order to determine its identity. In most applications, the gallery images (identities) are assumed to be independent of each other, i.e., the relationship between gallery images is not exploited during the face recognition process. In this work, we propose a graph-based approach in which gallery images are used to generate a powerful network structure where the nodes correspond to individual identities (and consist of face images as well as biographic attributes such as gender, ethnicity, name, etc.) and the edge weights define the degree of similarity between two such nodes.

Learn More

Auto-Tuned Models (ATM)

Many machine learning problems follow a familiar pattern: a set of features are selected and then many classifiers with varying parameters are manually tried using the feature set. We attempt to develop a system that automates the classifier discovery process. This work is a collaboration with the Data to AI (DAI) Lab at MIT.

Learn More

Spoofing Faces Using Makeup

Makeup can be used to alter the facial appearance of a person. We analyze the potential of using makeup for spoofing an identity, where an individual attempts to impersonate another person’s facial appearance.

Learn more