Web-Scale Near-Duplicate Search: Techniques and Applications

Submission Deadline: 29 June 2012
Publication Issue: July–September 2013

As bandwidth accessible to average users is increasing, audiovisual material has become the fastestgrowing data type on the Internet. The impressive growth of the social Web where userscan exchange user-generatedcontent contributes to the overwhelming number of multimedia filesavailable. Among these huge volumes of data, there exist large numbers of near-duplicatesand copies. File copies are easy to detect using hashes. Near-duplicates are based on thesame original content, but have been edited and postprocessed, resulting in different files. Another typeof near-duplicate relates to footage about the same event or scene. The detection of near-duplicatesposes a challenge to multimedia content analysis, especially when speed, scale, and copied fragmentlength are pushed to operational levels. Near-duplicates carry both informative and redundantsignals, for example, providing rich visual clues for indexing and summarizing videos fromdifferent sources. On the other hand, the excessive amount of near-duplicates makes browsing Web videosstreamed over Internet an extremely time-consuming task. As a result, there is strong interest fromindustry, academia, and governmental agencies in Web-scale search, elimination, detection, anduse of near-duplicates for various multimedia applications.


This special issue seeks innovative contributions dedicated to the theme of Web-scale near-duplicatesearch. Topics of interests include, but are not limited to, the following:

Techniques and algorithms

  • near-duplicate and/or partial near-duplicate detection;
  • cross-media search of near-duplicates;
  • semantic-based detection of near-duplicates;
  • framework and algorithm for real-time, near-duplicate detection;
  • semantic indexing and hashing techniques;
  • similarity and perception learning;
  • multimedia fingerprint extraction; and
  • instance search, matching, and localization.


  • search results ranking and diversification;
  • novelty detection;
  • topic detection, tracking, and threading;
  • data-driven applications;
  • Internet media management and service;
  • Web-scale multimedia mining;
  • Web-scale summarization and browsing of multimedia data; and
  • multimedia archaeology mining.


For more information, please contact the guest editors:

Submission Procedures

Submit your paper at https://mc.manuscriptcentral.com/cs-ieee. When uploading your paper, please select the appropriate special issue title under the category "Manuscript Type." If you have any questions regarding the submission system, please contact Andy Morton at mm-ma@computer.org. All submissions will undergo a blind peer review by at least two expert reviewers to ensure a high standard of quality. Referees will consider originality, significance, technical soundness, clarity of exposition, and relevance to the special issue topics. All submissions must contain original, previously unpublished research or engineering work.Papers must stay within the following limits: 6,500 words maximum, 12 total combined figures and tables with each figure counting as 200 words toward the total word count, and 18 references.


To submit a paper to the July-September 2013 special issue, please observe the following deadlines:

  • 29 June 2012: Full paper must be submittedusing our online manuscript submission serviceand prepared according to the instructions for authors (please see the Author Resources page at http://www.computer.org/multimedia/author.htm).
  • 15 January 2013: Authors notified of acceptance,rejection, or needed revisions.
  • 5 April 2013: Final versions due.


Submit an article »

Back to IEEE Multimedia homepage »

Back to IEEE Multimedia call for papers »