TUTORIAL SESSION

Advances in Image Search and Retrieval

Dr. Oge Marques
Florida Atlantic University, USA
omarques@fau.edu

Abstract

Topic: Content-based Multimedia Retrieval
Visual information retrieval (VIR) is an active and vibrant research area, which attempts at providing means for organizing, indexing, annotating, and retrieving visual information (images and videos) form large, unstructured repositories. In its early years (1995-2000) the research efforts were dominated by content-based approaches contributed primarily by the image and video processing community. During the past decade, it was widely recognized that the challenges imposed by the semantic gap (the lack of coincidence between an image’s visual contents and its semantic interpretation) required a clever use of textual metadata (in addition to information extracted from the image’s pixel contents) to make image and video retrieval solutions efficient and effective. The need to bridge (or at least narrow) the semantic gap has been one of the driving forces behind current VIR research. Additionally, other related research problems and market opportunities have started to emerge, offering a broad range of exciting problems for computer scientists and engineers to work on.
In this tutorial we revisit the field of content-based image retrieval (CBIR) 10 years after "the end of the early years" (as announced in a seminal paper in the field) and highlights the most relevant advances, pending challenges, and promising opportunities in CBIR and related areas.

Objectives

At the end of this tutorial, participants should be able to:
• Understand the fundamentals of visual information retrieval systems and their visual, textual, and computational aspects.
• Understand and appreciate the challenge involved in designing visual information retrieval systems.
• Identify directions for potential research in this field.

Timeline

1. Motivation: "What is it that we're trying to do and why is it so difficult?" (10 min)
2. Principles of Visual Information Retrieval (VIR) (40 min)
   • Selected concepts and principles from Information Retrieval (IR) )
   • Visual features)
   • Machine learning techniques)
   • Semantic properties of visual data)
3. Examples of contemporary VIR systems (research prototypes and commercial solutions) (10 min) )
4. Designing and evaluating a VIR system (25 min) )
5. Datasets, challenges, and benchmarks for VIR (20 min) )
6. Medical image retrieval (15 min) )
7. Mobile visual search (15 min) )
8. mage sharing, tagging, and annotation (15 min) )
9. User intentions in multimedia search (10 min) )
10. Case studies and success stories (10 min) )
11. Future directions (10 min)

Target Audience

Researchers and practitioners with familiarity with image/video processing, multimedia, information retrieval, and/or database systems.

Qualifications of the Instructor(s)

Dr. Oge Marques is Professor of Engineering and Computer Science at Florida Atlantic University. He is Tau Beta Pi Eminent Engineer, ACM Distinguished Speaker, and the author of more than 100 publications in the area of intelligent processing of visual information. Professor Marques is Senior Member of both the IEEE and the ACM and member of the honor societies of Sigma Xi, Phi Kappa Phi and Upsilon Pi Epsilon. He has more than 30 years of teaching and research experience in different countries (USA, Austria, Brazil, India, Spain, France, and the Netherlands).