ACM SIGGRAPH 2009 papers

Welcome to the SIGGRAPH 2009 Technical Papers Program! Here you will find an exciting collection of technical papers on topics in computer graphics, human-computer interaction, and related research areas. These papers push the state-of-the-art in wide variety of research areas, including image processing, surface modeling, fluid simulation, character animation, shape analysis, reflectance modeling, image synthesis, real-time interaction, and much more. There were 439 submissions, of which 78 were accepted, which is an 18% acceptance rate. In addition, 8 papers were conditionally accepted with major revisions to ACM Transactions on Graphics. Though decisions were made based purely on merit without a specific target number in mind, the acceptance rate was in line with previous years (2008=17%, 2007=24%, 2006=18%). The number of submissions was slightly down from last year (2008=520, 2007=455, 2006=474). This decrease may be due to the creation of SIGGRAPH Asia, which published 59 papers, some of which might otherwise have been submitted to SIGGRAPH 2009. Overall, 160 papers were published in TOG during the last year, which is 24% more than any previous year. Several changes were made to the Technical Papers review process this year. First, there were no uploads of images or videos with rebuttals. Over the past few years, authors could ask the committee for permission to post images, audio, and/or videos on the public BBS. While this feature was sometimes helpful for providing examples that answer specific questions posed by referees, it was used very differently by different authors and regulated differently by different referees. In some cases, an author would be allowed to upload entirely new examples, while nothing was allowed in others. The instructions clearly stated that rebuttals are only for "addressing factual errors in reviews." Yet, some authors would push the limits (for example: "The review said my method doesn't work, and so here are several new results to show that it does work."), and some referees were more lenient than others in allowing such uploads. To improve the uniformity of the review process, rebuttals were limited only to 2,000 words of text in 2009 (no images and no videos were uploaded with the rebuttal for any paper). This change improved the uniformity of the rebuttal process. Second, there was no discussion back and forth between authors and referees on the public BBS during the review process. Last year, referees could ask questions of authors on the public BBS at any time prior to the committee meeting, and authors could provide extended answers, sometimes with new visual results in response to specific questions. Thus, the review process was different for different papers, and unnecessarily stressful for all. This year, posts by referees to the public BBS were disabled, and thus the public BBS was used only by authors to upload a single, text-only rebuttal. This change improved the uniformity and reduced the stress of the rebuttal process. Third, there was an extra week in the review process, inserted after the rebuttal period and before the committee meeting. In recent years, the rebuttal period ended six days before the committee meeting, and thus there was little time for referees to discuss rebuttals and to solicit the opinions of other referees prior to the meeting. This year, an extra week was available, which gave plenty of time for the committee to find extra referees for papers where consensus could not be reached by the original five referees. The extra referees had enough time to read the papers in detail, complete full reviews, and discuss issues on the private BBS prior to the committee meeting. This change improved the thoughtfulness and transparency of the review process. Finally, there were no designated area coordinators this year. For the last two years, there were eight area coordinators (two each in modeling, rendering, animation, and imaging), which have helped sort papers amongst committee members, provided oversight during the review process, and helped run the committee meeting (during parallel sessions). This year, oversight and reviewing roles were distributed equally amongst all committee members. For every paper, a committee member from outside the research area was chosen to oversee the review process and arbitrate discussions. The goals of this change were to ensure that the review process for every paper was fair and timely and to diffuse information about how decisions are made across different subareas of graphics. These changes were minor adjustments to an already incredible review process. As papers chair this year, I was able to observe the tremendous amount of work that goes into selecting the highest quality papers for acceptance to SIGGRAPH. It's awe-inspiring. At least five reviews were written for every paper, and seven or more were written for most borderline cases. In all, 2204 reviews were written for 432 papers. If you consider all the text written during the review process (reviews, rebuttals, and BBS posts), it adds up to approximately 2.2 million words (15.5MB), which is approximately 70-80% as much as the text in all the submitted papers combined, including references (3.1 million words, 19MB)! It is an honor to be involved with such a committed and dedicated community of reviewers.