The ACM Multimedia Grand Challenge 2011 in a Nutshell

By Gerald Friedland and Yohan Jin

Gerald Friedland is a senior research scientist at the International Computer Science Institute and the 2011 chair of the ACM MM Grand Challenge. Contact him at fractor@icsi.berkeley.edu.

Yohan Jin is a Senior Data Scientist at Tapjoy and the 2011 chair of the ACM MM Grand Challenge. Contact him at yohanjin@gmail.com.

The 2011 ACM Multimedia Grand Challenge proved to be the most competitive so far. This article provides a quick summary of the competition. More comprehensive coverage can be found in an IEEE Multimedia Magazine report.

When the ACM Multimedia Grand Challenge was started in 2009 it was a radical idea: instead of submitting conference papers on solutions to self-imposed problems, scientists from the multimedia community were encouraged to engage in problems formulated by industry sponsors, called the Grand Challenges. In continuation of what now has become a tradition, the 2011 Grand Challenge consisted of six challenges sponsored by five industry sponsors: HP, Yahoo, Technicolor, Nokia, and Huawei/3DLife. The 2011 challenges, of which some are likely to be continued in 2012 can be found at the original website.

The initial submissions were pared down to 12 finalists in a two-round peer review with two industry and two academic reviewers. Each finalist submitted a four-page short paper, published in the ACM Multimedia 2011 proceedings.

The 12 finalists then presented a 180-seconds elevator pitch at ACM Multimedia in Scottsdale, Arizona, in a quick succession followed by a 120-minute question grilling by the audience and a jury. The jury consisted of representatives of the companies sponsoring the challenges, namely Qian Lin (HP), Christophe Diot (Technicolor), and Timo Pekka Pylvanainen (Nokia). After the audience had enjoyed the fast-paced conference session with the 12 innovative presentations, they sat down with the chairs to reach a decision based on concrete selection criteria formulated as part of the call. A major focus of the criteria was on the solution’s workability and generalizability as well as on the reproducibility of the results by other researchers.

The first prize was the only grading that the jury was guaranteed to award. It was nominated with a cash award of $500 and was reserved for the finalists who stood out in addressing the selection criteria. The second prize, for the runner up, was optional. Two special prizes optionally recognized the projects that showed the Best Integration of Multiple Media and the Best Application of a Theory Framework.

Here are the winners:

  • The first prize was awarded to “Photo Search by Face Positions and Facial Attributes on Touch Devices” by Yu-Heng Lei, Yan-Ying Chen, Lime Iida, Bor-Chun Chen, Hsiao-Hang Su, and Winston H. Hsu from National Taiwan University, Taipei, Taiwan.
  • The second prize was given to “Towards Textually Describing Complex Video Contents with Audio-Visual Concept Classifiers” by Chunk Chet Tan, Yu-Gang Jiang, Chong-Wah Ngo. (the first and third authors are with the Department of Computer Science, City University of Hong Kong, Kowloon, Hong Kong, and the second author is with the School of Computer Science, Fudan University, Shanghai, China).
  • The Best Application of a Theory Framework special award was given to “Understanding Images with Natural Sentences” by Yoshitaka Ushiku, Tatsuya Harada, and Yasuo Kuniyoshi, who are all at the University of Tokyo’s Graduate School of Information Science and Technology. Tatsuya Harade is also at JST PRESTO.
  • The Best Integration of Multiple Media special award was given to “News Browsing System: Multimodal Analysis” by Bruno do Nascimento Teixeira, Júlia Epischina Engrácia de Oliveira, Fillipe Dias Moreira de Souza, Tiago Oliveira Cunha, Igor Calil Loures de Oliveira, Chrisitane Okamoto, Lucas Figueiredo Gonçalves, Vinícius de Oliveira Silva, and Arnaldo de Albuquerque Araújo from the Department of Computer Science at the Federal University of Minas Gerais Belo Horizonte, Brazil.

 

The ACM Multimedia Grand Challenge is a continuing event at ACM Multimedia, and the 2012 conference is carrying the tradition into the fourth year. The new call can be found in this issue of the SIGMM Records and on the ACM Multimedia 2012 web pages.

We encourage everybody to consider participation as the competition has not only become prestigious but participants also report to have had a lot of fun!

References
  1. Y.-H. Lei et al., “Photo Search by Face Positions and Facial Attributes on Touch Devices,” Proc. ACM Multimedia, ACM Press, 2011, pp. 651-654.
  2. C. Chet Tan et al., “Towards Textually Describing Complex Video Contents with Audio-Visual Concept Classifiers by C,” Proc. ACM Multimedia, ACM Press, 2011, pp. 655-658.
  3. Y. Ushiku, T. Harada, and Y. Kuniyoshi, “Understanding Images with Natural Sentences,” Proc. ACM Multimedia, ACM Press, 2011, pp. 679-682.
  4. B. do Nascimento Teixeira et al., “News Browsing System: Multimodal Analysis,” Proc. ACM Multimedia, ACM Press, 2011, pp. 671-674.
  5. G. Friedland, Y. Jin, Industry Dares You: The ACM Multimedia Grand Challenge 2011,” IEEE Multimedia Magazine, 19(1), Jan 2012.
Bookmark the permalink.