新闻资讯
网站首页 >  新闻资讯
李锡荣博士获2012年度IEEE Transactions on Multimedia最佳论文奖

发布日期:2012-09-20  访问量:

  数据工程与知识工程教育部重点实验室教师李锡荣博士获2012年度IEEE Transactions on Multimedia最佳论文奖。作为多媒体领域国际顶级期刊,IEEE Transactions on Multimedia每年从其过去三年发表的文章中根据论文的原创性、实用性、时效性和表述清晰度等评选出一篇最佳论文。李博士的论文 "Learning Social Tag Relevance by Neighbor Voting" 从14篇提名论文中最终胜出。据了解,李博士作为论文第一作者,也是首位获得该奖项的中国大陆高校教师。该文是李博士在荷兰阿姆斯特丹大学攻读博士学位期间,与其导师的合作成果。自2009年11月发表以来,该论文已被SCI他引26次。  

  随着人类社会进入"社会化媒体"时代,我们已由传统的信息消费者变成了信息产消者。图像信息作为人类最直观和表达最为丰富的信息表现形态,每天都有数以亿计的图像被世界各地的互联网用户上传到网上。用户赋予的社会化标签是这些图像的主要描述信息,但却是主观和不准确的。李博士在论文中提出了一种基于近邻投票的计算机算法,能够自动判断标签所描述的物体、场景等是否真正在图像内容中出现。论文形式化地分析和证明了算法在图像排序和标签排序上的有效性。

论文信息
 
Xirong Li, Cees G. M. Snoek, and Marcel Worring, "Learning Social Tag Relevance by Neighbor Voting," IEEE Transactions on Multimedia, volume 11, issue 7, pp. 1310-1322, November 2009
 
Dr. Xirong Li wins
IEEE Transactions on Multimedia Prize Paper Award
Dr. Xirong Li of the MOE Key Lab of Data Engineering and Knowledge Engineering has been awarded the 2012 IEEE Transactions on Multimedia Prize Paper Award. Li received the prize for the publication "Learning Social Tag Relevance by Neighbor Voting".   The Multimedia Prize Paper Award is an annual award for an original paper in the field of multimedia published in the IEEE Transactions on Multimedia in the previous three calendar years. The paper of Xirong Li was selected out of 14 nominations. The basis for judging is the composite of: originality, utility, timeliness, and clarity of presentation.   About the research   In a world where the amount of digital images is ever-growing, it is important to be able to search based on the visual content. Xirong Li was inspired by social media and investigated the value of images with social tags for visual search. He developed an algorithm that automatically determines whether the tag people assign to a photo matches what is actually visible in the image. Moreover, the paper provides a formal analysis on the proposed algorithm, theoretically showing its effectiveness for both image ranking and tag ranking.   Paper   Xirong Li, Cees G. M. Snoek, and Marcel Worring, "Learning Social Tag Relevance by Neighbor Voting," IEEE Transactions on Multimedia, volume 11, issue 7, pp. 1310-1322, November 2009.