Keynote and invited practitioner speakers - CIVR2006

Keynote Speaker

Exploratory Search Interfaces to Support Image Discovery

Ben Shneiderman

Ben Shneiderman
University of Maryland
Founding Director,
Human Computer Interaction Laboratory

Abstract:

Enabling searchers to find desirable images or videos remains a challenge to interface designers because of at least these three challenges:

  1. the users' knowledge of the data is incomplete (they don't know what is in the data);
  2. the users' tasks are vaguely specified (they don't know exactly what they want);
  3. the system indexes are not in harmony with the users' search request (they know what they want, but the system has no indexing for it).
A growing research area, called 'exporatory search interfaces', is emerging to enable users to get what they need. This talk will present some current interface designs, describe novel search strategies, and propose metadata recommendations. Meaningful and stable hierarchies will be offered as one way to help users find what they need.

Biography:

Ben Shneiderman is a Professor in the Department of Computer Science, Founding Director (1983-2000) of the Human-Computer Interaction Laboratory (http://www.cs.umd.edu/hcil/), and Member of the Institutes for Advanced Computer Studies and for Systems Research, all at the University of Maryland at College Park. He was elected as a Fellow of the Association for Computing (ACM) in 1997, and as a Fellow of the American Association for the Advancement of Science (AAAS) in 2001. He received the ACM SIGCHI Lifetime Achievement Award in 2001. Ben is the author of Software Psychology: Human Factors in Computer and Information Systems (1980) and Designing the User Interface: Strategies for Effective Human-Computer Interaction (4th ed. 2004) (http://www.awl.com/DTUI/). He pioneered the highlighted textual link in 1983, and it became part of Hyperties, a precursor to the web. His move into information visualization helped spawn the successful company Spotfire (http://www.spotfire.com/). He is a technical advisor for the HiveGroup and ILOG. With S. Card and J. Mackinlay, he co-authored Readings in Information Visualization: Using Vision to Think (1999). His recent books include Leonardo's Laptop: Human Needs and the New Computing Technologies (MIT Press) which won the IEEE book award in 2004.

Invited Practitioner Speakers

Embrace and Tame the Digital Content

Gulrukh Ahanger

Gulrukh Ahanger
Turner Broadcasting Systems

Abstract

Typically, on a daily basis, large amounts of video content are processed by broad-casting stations. This includes from ingest to cutting packages and eventual transmission and storage. New digital broadcast systems are being put in place and these sys-tems are enabling the transition from tape-based to file-based workflow. In addition, news production systems with varying and changing workflows are increasingly becoming distributed across the bureaus and pushed out into the field. The expectations of news producers and journalists have changed; they want easy access to media for broadcast as well as for package production anywhere in the world, and at anytime. Providing this access to content, when and where needed, significantly impacts the quality of the broadcast product.

Broadcast stations are significantly gaining the ability to move content quickly and efficiently along the digital supply chain throughout the entire production and distribution process. It is being made possible largely due to a file-based environment as the digital file acquisition begins with camera. We need to maximize efficiencies gained from this change in news gathering paradigm; to be of any use, the digital content river coming our way needs not only to be embraced but also to be tamed. Solutions are needed to deliver business value and return on investment through organizing, accessing, distributing, and tracking the flow of vast amounts of the digital media across multiple channels. Technologies and tools are needed that will provide the means of searching, accessing, and sharing content across different location transparently and efficiently.

This talk, unlike many of the others you will hear today, is not about the wonderful things that I am currently doing, or the amazing projects that are just around the corner. I was invited today to speak to you about a project that we began in 1996 and that continues in various forms well into 2006.

Biography

Gulrukh Ahanger has a background in multimedia technologies involving digital asset management applications and broadcast production systems. Dr. Ahanger is currently a technical program director for Broadcast Systems at Turner Broadcast Systems. She is spearheading the Integration Production Environment program, an initiative to put a file-based production workflow in place at CNN. It includes creation, search, and access of digital media to enable CNN's on-air production; in addition, it includes integration of CNN and its various bureaus worldwide to access media globally. She also has been instrumental in directing product development and strategy related to enhancing the scope of some of the asset management products in market today. Dr. Ahanger received her B.E. in electronics and communication engineering from the Regional Engineering College, Kashmir, India in 1998; M.S. in 1993 and Ph.D. in 1999 in systems engineering at the Department of Electrical and Computer Engineering at Boston University. While at Boston University, Dr. Ahanger performed various research activities with an emphasis on dynamic information generation and she proposed new techniques for automatic digital video production systems.

Discovering a fish in a forest of trees. False positives and user expectations in visual retrieval.
Experiments in CBIR and the Visual Arts

Marty Harris
Adobe Systems Incorporated

Abstract

Museums, Libraries, and Archives are the keepers of huge numbers of images including photographic prints, images of artworks, photos of sculpture, and many other image types. In the fall of 1995 the Getty Art History Information Program, a division of the J.Paul Getty Trust which focused on technologies which could be used to facilitate access to the visual arts, began a project to investigate and demonstrate to its community, content based image-retrieval technologies over the Web. Partnered with NEC and based on their Amore image-matching engine, The "Arthur" system mixed CBIR models and enhanced text-searching tools. The resulting project allowed web users to search visual materials collected from over 1000 Web Museum, Image Archive and Digital Library sites.

Between 1996 and 1999, users of the Getty/Arthur CBIR system logged several million queries from a broad range of users including art scholars and researchers, archivists, educators, students, k-12 to graduate, librarians, and information scientists. Each group had their own expectations regarding what they were asking for, and the relevance of the resulting retrieval set. Questionnaires and internal usability studies identified several consistent and important questions, which guided the development and growth of the Arthur project. This talk examines some of the issues and lessons learned and describes some steps that where taken to improve user satisfaction.

Biography

Marty Harris is a Computer Scientist currently working for the Digital Imaging Group at Adobe Systems Incorporated. From 1986 until 1999 he was Manager and then Director of Research and Technology for the Getty Art History Information Program. In late 1999 he formed the Center for Cultural Technology at the University of Southern California's, Information Sciences Institute. In 2001 he and several partners started QuickDog, a company focused on using Text Retrieval Technologies, Information Clustering, and CBIR to enhance online shopping. In 2003 he and a partner formed Albathion, a company focused on providing Venture Capital technology evaluations. He currently is involved in evaluating image retrieval and tagging models.