ACM SIGMM Award for Outstanding Technical Contributions to Multimedia Computing, Communications and Applications

The 2020 winner of the prestigious ACM Special Interest Group on Multimedia (SIGMM) award for Outstanding Technical Contributions to Multimedia Computing, Communications and Applications is Dr. John R. Smith. The award is given in recognition of his outstanding, pioneering and continued research contributions in the areas of multimedia content analysis and retrieval and for outstanding and continued service to the multimedia community.
Dr. Smith is a pioneer in the field of multimedia content analysis and retrieval and is a prominent leader in the ACM Multimedia community. Dr. Smith conducted some of the earliest and most influential work in the multimedia field in the late-1990’s that established the field of content-based retrieval. His work subsequently achieved high impact broadly across multimedia as evidenced through highly-cited and influential papers (h-index = 76, i-index = 387, # citations = 29,903) and leadership on many fronts of multimedia research and development including directing research on multimedia and vision at one of the world’s top industrial research labs, contributing to the establishment of significant public evaluations, benchmarks, and standards in multimedia, organizing conferences, workshops, and publications as well as fostering collaborations and guiding development of numerous prominent scientists and members of the multimedia community.
Dr. Smith's early work was highly influential in defining the field of content-based retrieval pioneering a novel and highly effective approach for visually querying contents of image databases. As evidence of his impact, Dr. Smith’s seminal paper on VisualSEEk, published at ACM Multimedia in 1997, is the second most cited paper in the history of ACM Multimedia with 2,949 citations. This work also resulted in WebSEEk, one of the first image and video search engines for the Web in 1997, which cataloged, analyzed, and enabled searching of more than 300K images and videos on the Web.
Dr. Smith was also an early champion and prominent leader in establishing public evaluations for multimedia tasks. Today, leaderboards have become fundamental to how research is done in the broader AI field and are essential for continuing scientific and technical advancement. Dr. Smith first argued for formation of an image retrieval evaluation in 1998 and subsequently contributed to the development of the NIST TRECVID video retrieval evaluation. Dr. Smith led the definition of key aspects of the TRECVID’s benchmark for semantic concept detection, which was later replicated by the AI community in ImageNet. At the same time, Dr. Smith led his extended IBM Research team in achieving top performance on NIST TRECVID for multiple years.
Dr. Smith also contributed to the establishment of the new paradigm called "Universal Multimedia Access," for automatically adapting multimedia content (images, video, text) for client devices. His paper in IEEE Trans. Multimedia won the "Multimedia Prize" award for most influential paper in the journal over a four-year period from 1999-2003. This seminal work formed the basis for MPEG-7 standard’s definition of Universal Multimedia Access and was commercialized in IBM’s Websphere Transcoding Publisher. Dr. Smith also contributed significantly to the broader development of standards for multimedia. He served as Editor of MPEG-7 Multimedia Description Schemes standard and led MPEG's development of the MPEG-21 Digital Item Adaptation standard by serving as Chair of MPEG Multimedia Description Schemes Group from 2001-2004.
Dr. Smith provided important leadership for the multimedia field through organizing conferences, workshops, journals, and by giving tutorials and keynote talks. Dr. Smith was co-General of ACM Multimedia in 2012, co-Technical Chair of ACM Multimedia in 2003, and co-General Chair of ACM Intl. Conf. on Multimedia Retrieval in 2016. Dr. Smith served as Editor-in-Chief of IEEE MultiMedia from 2010-2014 and Associate Editor-in-Chief of IEEE MultiMedia from 2006-2009. Dr. Smith organized numerous special issues in the top journals including IEEE Trans. on Multimedia, IEEE Trans. Circuits and Systems for Video Technology, and Proceedings of IEEE. Dr. Smith gave keynote talks at prominent conferences including ACM Special Interest Group on Information Retrieval Intl. Conf (SIGIR), IEEE Intl. Symposium on Biomedical Imaging (ISBI), ACM Conference on Embedded Networked Sensor Systems (SenSys), and Int. Joint Conf. on Artificial Intelligence (IJCAI). He recently gave two invited talks at workshops at the Computer Vision and Pattern Recognition (CVPR) in June 2020.
In parallel to his many efforts contributing to the broader multimedia field, Dr. Smith has led commercial research in multimedia and vision at IBM T. J. Watson Research Center, one of the top research institutions in the world, for the last two decades. During this tenure, Dr. Smith has led IBM’s research and development efforts behind numerous successful commercial products including: IBM’s Watson Visual Recognition, IBM Video Analytics, Watson Video Enrichment, Watson Video Highlights, Power AI Vision, and Visual Insights. Some recent prominent results from Dr. Smith’s work at IBM include the first AI-created movie trailer (>3M views on YouTube) and development of the first AI-powered system for generating highlights from sports events, which has been fielded since 2018 at both Wimbledon and US Open tennis events and US Master golf.

 Many congratulations to John

- Nicu Sebe
- SIGMM Vice-Chair