Research

My interest in image processing, computer vision, and related topics dates back from 1988 when I took my first Digital Image Processing class as a Master’s student and had the opportunity to do research work on Optical Character Recognition (OCR) at Philips Research Labs (Eindhoven, the Netherlands) which eventually led to my Master’s thesis and a number of associated publications (not listed in my Publications page). Ever since then, those topics have been present in most of my publications, courses I created and taught, and graduate students I have advised.

From 2001 until 2004, my research had built upon the topic of my PhD dissertation (content-based image retrieval, CBIR) and expanded to other relevant topics under the umbrella of image and video processing, analysis, annotation, search, and retrieval.

From 2004 until now, my research efforts have benefited tremendously from external funding coming from two federal government research contracts, one with the Office of Naval Research (ONR) under the “Center for Coastline Security Technologies (CCST)”, another with the Department of Defense (DoD) under the topic of “Secure Multimedia Communications”. The former provided the opportunity to leverage my image and video processing and analysis expertise, extending it to the marine surveillance domain. The latter opened the doors to new research efforts in promising new directions.

I have started working on the third topic in my list of research interests (human and computer vision) in 2005. The motivation for learning more about the psychological, physiological, and psychophysical aspects of human vision research stemmed from recent progress in the field of computational neuroscience of vision. I have created and taught a new graduate course (Foundations of Vision) on the topic, which has motivated several students to pursue research work in related areas.

My particular focus of interest so far has been the use of computational models of visual attention in the context of content-based image retrieval, a topic that combines the latest developments in computational models of vision with my expertise in an application domain that still has many unresolved issues. I've had the privilege of working with a team that includes one of my PhD students at FAU (Liam M. Mayron), a PhD student at UTFPR (Curitiba, Brazil) (Gustavo B. Borba) and his main advisor (Prof. Humberto R. Gamba), and a former FAU student, currently pursuing his PhD at University of Amsterdam (Vladimir Nedovic). This ongoing research effort has led to the development of an innovative prototype for image retrieval, organization and annotation (PRISM), a patent application, and several papers (see Publications page).