Welcome to the second Internet Imaging Conference at IS&T/SPIE's International Symposium on Electronic Imaging. In a time when IPOs lure Internet specialists to start-up companies and corporate research laboratories are no longer of strategic importance, why organize a conference related to an Internet technology? The old business strategy of locking in customers and locking out competitors -- requiring captive research labs -- has been replaced by a strategy based on mergers and acquisitions. In science and technology, this paradigm shift in business strategy is reflected in a shift from proprietary technologies to an open source model that can evolve much more rapidly.
The Electronic Imaging Symposium is the stalwart of the new research paradigm. Rather than a stuffy restricted club, the symposium has been more of an uplifting bazaar, where the emphasis is on the rapid communication of new ideas. The conference chairs have been courageous visionaries, who have been willing to take the risk of encouraging unknown researchers with brilliant ideas, a task that is much harder than accepting papers based on the author's fame.
In this new research paradigm, career is no longer an escalator where one moves up as long as one works hard. Today there are many different -- often intersecting -- paths up the hill to successful careers; therefore, it is essential to build networks of colleagues with similar interests. New breakthroughs have to be detected immediately and must be assimilated in a very short time. In this situation the Internet Imaging Conference fulfills two roles.
The symposium's conferences are organized in programs that group similar research areas; the system of synchronized presentations encourages you to hop from conference to conference, maximizing the likelihood of serendipity. Last but not least, the informal atmosphere is meant to encourage you to interact with the speakers; take advantage of discussion sessions, panel sessions, and receptions to network and build a strong community.
Internet imaging differs from other forms of electronic imaging in that it employs an internet (network of networks) as a transmission vehicle. However, the Internet is only one component (albeit a major one) in the total imaging system. The total system comprises client applications internetworked with server applications, as well as off-line authoring tools.
The Internet is an evolving communication system. Its functionality, reliability, scaling properties, and performance limits are largely unknown. The transmission of images over the Internet pushes on its engineering envelope more than most applications. Consequently, the issues we are interested in exploring pertain to all aspects of the total system; not just images or imaging algorithms.
This emphasis on systems is what sets this conference apart from other electronic imaging conferences. For a local imaging application, even when it is split between a client and a server linked by an ethernet, a system can be designed by stringing algorithms in a pipeline. If performance is an issue, it is easy to identify the weak link and replace it with a better performing component.
On the Internet, the servers are unknown, the clients are unknown, and the network is unknown. The system is not easily predictable and the result is that the most common problem today is scalability. To be successful one has to follow a top-down design strategy, where the first step is a detailed analysis of the problems to be solved. When a solution is invented, algorithms are selected to produce a balanced system, instead of choosing algorithms of best absolute performance as is done in bottom-up approaches.
Appropriately, the conference opens with a plenary speaker on digital cinema and a session on systems and architecture.
The second session is on visual organization. As detailed by Tuesday's plenary speaker, users are shedding wires very fast and adopting wireless technologies. Concomitantly, the next generation of adults was born using video games before learning to write; proficiency in video games is acquired by rapid trial and error, not by reading manuals. Game machines have higher computational power than desktop computers, and multimedia effects are highly integrated. Reference information is not looked up in encyclopedias, but searched on the Web. The next generation will expect to have a system with these performance characteristics.
We are thus confronted with the dichotomy of huge amounts of visual material that must be browsed over a slow wireless communications channel. This requires understanding the principles of visual organization so that intelligent image processing algorithms can be designed. The third session, on video summarization, continues the conversation for image sequences.
After a joint session on image retrieval, the conference continues with sessions on metadata, which allows to encode in a very compact form the semantic information about images, and on visualization, which follows the goals of compact data representation and computation in a fast client, rather than an overburdened server on a congested link.
The World Wide Web is an Internet application that was invented to foster collaborative research. It has encouraged the creation of geographically distributed communities of practice, also known as extended knowledge networks (EKN). These EKNs have been essential for the rapid adoption of the Internet by individuals, businesses, and institutions. Because location is irrelevant and like in a village organization is low and participation is high, Marshall McLuhan coined the term of global village.
Two sessions are dedicated to these global village aspects of Internet imaging. The first covers telepresence and collaborative design, while the second is devoted more specifically to the rapidly evolving field of telemedicine.
After a joint session on video retrieval, the conference concludes with a session on algorithmic issues in Internet imaging.
At the first Internet Imaging conference in 2000, Theo Gevers suggested the creation of an event where the performance of image retrieval algorithms over the Internet can be benchmarked. The motivation was that although this is an algorithm class of high commercial importance, there is currently no benchmark available to compare the various proposed algorithms. Despite the cheering enthusiasm in the crowd, we have been less successful than expected in building an EKN willing to implement it.
The event is called Benchathlon and Thursday evening there will be a paper describing the event as well as a proposal for a benchmark, called BIRDS-I for Benchmark for Image Retrieval using Distributed Systems over the Internet. We have a server for an image database and Perl scripts for benchmarking; we are seeking researchers interested in playing an active role as part of an EKN.
This conference is for you and with you. Please network with the speakers and the other participants, and interact with us. Tell us how we can improve the conference and make it more worthwhile. We hope we are providing a stimulating experience that will lead to a paper of yours in the 2002 conference.
Giordano Beretta, Hewlett-Packard Co.
Raimondo Schettini, CNR (Italy)
Conference Chairs