Institute of Information Science, Academia Sinica

In this talk, we will talk about the problem of grouping a collection of unconstrained face images in which the number of subjects is not known. We propose unsupervised clustering algorithms which are based on measuring the affinities between local neighborhoods in the feature space. By learning the local information for each neighborhood, information about the underlying structure is encapsulated. The encapsulation aids in measuring the neighborhood similarity. Extensive experiments show that the proposed methods are superior candidates for clustering unconstrained faces when the number of subjects is unknown. Unlike conventional linkage and density-based methods that are sensitive to the selection operating points, the proposed approach attains more consistent and improved performance. In addition, we will also discuss an extension using the same similarity learning approach for single subject tracking.

Jun-Cheng Chen currently is an assistant research fellow at the research center of information technology innovation, Academia Sinica. He received his bachelor’s and master’s degrees in 2004 and 2006, respectively, both from Department of Computer Science and Information Engineering, National Taiwan University, Taipei. He received his Ph.D. degree from the University of Maryland, College Park, in 2016. He is a postdoctoral research fellow at the University of Maryland Institute for Advanced Computer Studies from 2017 to 2019. His current research interests include computer vision and machine learning with applications to face recognition and facial analysis. He was a recipient of the 2006 Association for Computing Machinery Multimedia Best Technical Full Paper Award.