Welcome to Portland, Oregon and the 26th IEEE Conference on Computer Vision and Pattern Recognition (CVPR). In addition to the main three-day program of oral and poster presentations (in two parallel tracks), CVPR 2013 has a number of co-located events, including 22 workshops, 9 tutorials, and on-site demos and exhibits. In order to allow for one-minute spotlight presentation of each poster, oral presentations have been shortened to 15 minutes each, but oral presenters now optionally get to present a poster as well.
For this year’s main conference, we received 1816 completed submissions to the conference, of which 1798 were fully reviewed. (The other papers were either rejected for technical reasons or withdrawn before review.) To select papers from these submissions, we invited 52 well-known vision researchers to act as Areas Chairs (ACs) and recruited an expert team of 932 reviewers from the broader computer vision community, with a maximum of 11 papers per reviewer and an average/median load of 5 papers.
Recognizing the crucial importance of qualified reviewers to the review and decision process, the initially compiled reviewer pool was first vetted by the Program Chairs through cross-checking a reviewer's recent publications in a number of major computer vision related conferences and journals, and then augmented by additional reviewers recommended by the ACs. We again used the CMT conference management service sponsored by Microsoft Research to manage the submission and selection of papers from beginning to end.
After the submission deadline, the Program Chairs distributed the papers to the ACs with help from the automated Toronto Paper Matching System (TPMS) developed by Charlin et al. [UAI 2011]. TPMS suggests matches between papers and reviewers (ACs, in our case) based on bag-of-words descriptors extracted from the PDF files of submitted manuscripts and representative publications by each potential reviewer; for CVPR 2013 we had a Program Coordination Chair who was in charge of the interface with TPMS. The ACs in turn used the results of a TPMS matching of papers to reviewers to help them determine the potential reviewers for each of their assigned papers, from which the CMT system automatically selected three non-conflicted reviewers per paper. Finally, extensive manual adjustments were made by the ACs and Program Chairs to achieve better matches between the papers and reviewers under the workload constraints. In summary, the critical task of matching papers to ACs and reviewers were made by the Program Chairs and ACs, with support from the CMT and TPMS software.
Reviewers were given five weeks to complete their reviews, at which time the ACs stepped back in to vet the reviews for quality (initiating discussions, where necessary) before they were released to the authors. After the author rebuttals were collected, the area chairs finished their pre-meeting work, i.e., consolidating the reviews and author rebuttals, initiating discussions for clarification, and making recommendations for decisions on papers. The Program Chairs and the ACs strove to ensure that every paper eligible for full review received at least three good quality reviews.
Every paper, its reviews and author rebuttal were looked at by at least two ACs. To further support a thorough review process, at the AC Meeting at the University of Southern California, the ACs were divided among six panels, with no conflicts between the ACs and papers associated with each panel. The Program Chairs served as the panel chairs and worked hard to maintain consistency between the panels. All decisions were made by at least two ACs working together and, as needed, by the whole panel. A consensus of the entire panel was sought on the most difficult cases. By the end of the meeting, the ACs were asked to produce detailed consolidation reports to justify all their decisions.
The Program Chairs and General Chairs did not submit any papers to CVPR 2013, allowing them to work without any direct conflicts throughout the review process. Additionally, the respective panel chairs were excluded from any decisions associated with papers from their affiliated institutions. The double-blind nature of the CVPR review process was thus strictly maintained throughout the review process.
At the final program committee meeting, the ACs accepted 60 papers as orals (3.3% of submissions) and 412 papers as posters, giving an overall acceptance rate of 26.2% of submissions. There was no quota for the number of orals or posters.
The proceedings of CVPR 2013 are being published in USB drive form. All papers in the main conference and associated workshops will be indexed by the IEEE, and available through the IEEE Computer Society Digital Library and under IEEE Xplore.
While the most important aspect of CVPR 2013 is the high degree of care that the Program Chairs exercised in the paper selection process, the conference is also introducing two important organizational changes. This is the first CVPR where the winning bid was put together by the PAMI-TC Conference Committee, after no bids were received by the deadline. This new process, which is now part of the PAMI-TC’s charter for CVPR, is designed to avoid the last minute scramble to put together a bid which has occasionally been seen in the past.
In addition, CVPR 2013 introduces a new sponsorship model, under another new provision of the charter. Until recently, CVPR was 100% sponsored by the IEEE Computer Society. Following CVPR 2011, some senior computer vision researchers created our own non-profit, with the self-explanatory name of “The Computer Vision Foundation” (CVF). After extensive discussions with the IEEE Computer Society, a mutually satisfactory co-sponsorship arrangement was created where CVF and IEEE serve as equal partners. The intent is to continue to provide CVPR with its longstanding IEEE affiliation, while also ensuring that the vision community’s interests and concerns are given the appropriate degree of priority. The sponsorship model that CVPR 2013 is pioneering has been provisionally adopted by ICCV 2013, CVPR 2014 and CVPR 2015.
We wish to thank the other members of the Organizing Committee, the Area Chairs, Reviewers, Authors, and the CMT team for the immense amount of hard work and professionalism that has gone in to making CVPR 2013. Our thanks also go to the organizers of previous CVPRs for their helpful advice and support. We are grateful to the sponsors as well, and are happy to report that CVPR has set a record with over $120,000 of industrial support. Finally, we wish all the delegates a highly stimulating, informative, and enjoyable conference.
Gérard Medioni and Ramin Zabih, General Co-Chairs
Martial Hebert, Bill Freeman, Greg Hager, and Richard Szeliski, Program Co-Chairs