The Community for Technology Leaders



Pages: pp. 3-4

I first encountered the imposingly-named IEEE Transactions on Pattern Analysis and Machine Intelligence ( TPAMI) when I was a graduate student. It was the publication venue for a paper by Geman and Geman which every article I read seemed to cite, but which no researcher I queried claimed to understand. My efforts in tracking down Geman and Geman’s paper were well rewarded, as it influenced me enormously; to my surprise, the journal that published it is now poised to play a central role in my professional life.

It is an honor indeed toa be selected as the Editor-in-Chief of this journal, which has published so many fundamental papers. TPAMI is rightfully a source of pride for the research community it represents, not only for its history of excellence but also for its nonprofit status. Yet, the journal’s quality does not result from top-down decisions; rather, it emerges from an enormous investment on the part of the community as a whole, in submitting the best work to TPAMI and in reviewing an ever-growing number of submissions. It is striking that this entire effort is completely voluntary and uncompensated; no one involved in the decision-making process is paid for their work and the subscription income from TPAMI flows back to IEEE and thus benefits the entire community. This impressive ecosystem is a testament to our willingness to cheerfully shoulder short-term burdens that produce long-term rewards.

TPAMI has fared exceedingly well under the wise stewardship of David Kriegman, who has been involved in running the journal since he joined the editorial board in 1997. In my experience as a reviewer, author, and editorial board member, the review and publication process runs fairly smoothly (my occasional involvement with other publications has generally left me with an even greater appreciation for TPAMI). The most important aspect of any journal, of course, is the quality of the research that it publishes, and here TPAMI has continued to excel.

Yet, even the best-run journal has room for improvement and it is clear that TPAMI faces several challenges. Here are a few issues that I feel are important for TPAMI to address. Of course, this list is not exhaustive and feedback of any kind is more than welcome.

One issue involves the growing “Open Access” movement, which is particularly focused on free public access to government-funded research. This movement has been embraced by several top-flight journals and has also found substantial support at leading universities. Like most academics, I am extremely sympathetic to the goals of the open access movement. Yet it would be irresponsible to ignore the challenges that open access poses to the financial model that supports IEEE journals such as TPAMI. I believe that TPAMI can play a leadership role as this issue plays out, due to its large and technically sophisticated audience and its long-standing reputation for excellence, as well as IEEE’s nonprofit nature.

Another issue involves the main conferences in TPAMI’s research area. While the number of submissions to TPAMI is quite substantial, there has been an unfortunate tendency for authors of quality work to forego journal publication and to leave the research purely in its conference form. This phenomenon testifies to the strength of the top conferences, yet it deprives both the authors and the wider community of the benefits of the journal publication process. In particular, conferences do not generally permit submissions to be revised and subjected to further review; instead, a conference paper is accepted based on an initial submission and there is typically little review of the final publication. In contrast, a paper appears in TPAMI only after it is has been sufficiently polished to allay any significant concerns of the reviewers, as judged by an editorial board member. In addition, of course, TPAMI papers can be longer than conference papers and the review process is less time-compressed than for conferences, and thus (hopefully!) of uniformly high quality. One of my priorities is to smooth the way for conference papers to be submitted to TPAMI, while still ensuring that they undergo the same rigorous reviewing process.

In terms of the day-to-day operations of TPAMI, two main issues spring to mind. First, there continues to be an appreciable lag between the date when a paper is accepted and the date it actually appears in print. TPAMI has introduced relatively rapid online publication and, hopefully, over time, the date of online publication will become more widely recognized. Second, the sheer number of submissions poses an ongoing challenge and we need to investigate ways to reward reviewers for their efforts (perhaps by instituting some annual reviewing awards). This is especially necessary as TPAMI attracts more papers in areas that were once viewed as peripheral, such as machine learning, computer graphics, or optimization. While it is extremely gratifying that so much exciting work crosses into these areas, this exacerbates the difficulty of attracting the appropriate editors and reviewers.

I want to publicly thank and acknowledge David Kriegman and David Fleet for the enormous amount of dedicated work that they have put into this journal. David Kriegman first appointed me to the TPAMI editorial board and has been a source of tremendous help and encouragement in all aspects of the process. If pressed to single out one instance, it would be the flurry of after-midnight e-mails that we regularly exchanged when I was cochairing CVPR in 2007. While David Fleet encouraged me to apply for this position, outside of this one lapse, he has shown remarkable taste and judgment over the years.

It is clear that the many difficult decisions involved in running a journal of TPAMI’s magnitude require ongoing advice from a small group of senior researchers. In the past, TPAMI had an advisory board as well as an editorial board and, effective with this issue, the advisory board is being reconstituted. The initial members of the advisory board are Chris Bishop, Andrew Blake, Eric Grimson, Dan Huttenlocher, and Gerard Medioni. In addition, I have the pleasure to announce that both David Fleet and David Kriegman have also agreed to join (thus extending David Kriegman’s record-setting tenure on the inside cover and ensuring the continued involvement of two researchers who have helped make TPAMI what it is today). While the TPAMI advisory board does not have a supervisory role, it will serve as a valuable source of institutional wisdom.

Finally, I am delighted that Zoubin Ghahramani has agreed to continue on as an Associate Editor-in-Chief. His expertise, especially in the area of machine learning, will continue to be of great value to TPAMI. While David Fleet’s storehouse of knowledge and experience is irreplaceable, he has now completed his term and the workload clearly requires that his vacancy be filled. I am therefore happy to announce that Jiri Matas has been selected by the IEEE Computer Society Publications Board to serve as Associate Editor-in-Chief of TPAMI for the next two years.

Jiri’s brief biography appears below. I believe that Jiri, Zoubin, and I will work well together as a team, and we will do our best to live up to the confidence that the community has vested in us and to uphold the traditional excellence of this journal.

Ramin Zabih


About the Authors

Jiri Matas received the MSc degree in cybernetics (with honours) from the Czech Technical University, Prague, Czech Republic, in 1987 and the PhD degree from the University of Surrey, United Kingdom, in 1995. From 1991 to 1997, he was a research fellow at the Centre for Vision, Speech and Signal Processing at the University of Surrey. In 1997, he joined the Center for Machine Perception at the Czech Technical University in Prague. Since 1997, he has held various positions at these two institutions. He has published more than 100 papers in refereed journals and conferences. His publications have more than 1,100 citations in the Science Citation Index. He received the best paper prize at the British Machine Vision Conferences in 2002 and 2005 and at the Asian Conference on Computer Vision in 2007. Dr. Matas has served in various roles at major international conferences (e.g., ICCV, CVPR, ICPR, NIPS, ECCV), cochairing ECCV ’04 and CVPR ’07. He is on the editorial board of the International Journal of Computer Vision, Pattern Recognition, and the IEEE Transactions on Pattern Analysis and Machine Intelligence. His research interests include object recognition, image retrieval, sequential pattern recognition, ensemble methods, invariant feature detection, and Hough Transform and RANSAC-type optimization. He has been the lead scientist in industrial projects funded by a number of high-tech companies, e.g., Toyota, Hitachi, Samsung, Xerox, Boeing, and 2d3. He is one of the founders of Eyedea Recognition, a spin-off company of CTU Prague focusing on applications of computer vision and pattern recognition.
66 ms
(Ver 3.x)