Blip10000: A social Video Dataset containing SPUG Content for Tagging and Retrieval

Author: TU Berlin

Partner: No





The increasing amount of digital multimedia content available is inspiring potential new types of user interaction with video data. Users want to easily find the content by searching and browsing. For this reason, techniques are needed that allow automatic categorisation, searching the content and linking to related information. The presented dataset contains comprehensive semi-professional user-generated (SPUG) content, including audiovisual content, user-contributed metadata, automatic speech recognition transcripts, automatic shot boundary files, and social information for multiple ‘social levels’. The principal characteristics of this dataset are described and results are presented that have been achieved on different tasks.


The files are available for download via HTTP. Link: Direct link to the files: Link:

References and Citation

Use of the datasets in published work should be acknowledged by a full citation to the paper [SXF13] at the MMSys conference (Proceedings of ACM MMSys 13, February 27 - March 1, 2013, Oslo, Norway).


  • SXF13: S. Schmiedeke, P. Xu, I. Ferrané, M. Eskevich, C. Kofler, M. Larson, Y. Estève, L. Lamel, G. Jones, T. Sikora, Blip10000: a social video dataset containing SPUG content for tagging and retrieval, Proceedings of the 4th ACM Multimedia Systems Conferen (MMSys), Oslo, Norway, USA, February 27 - March 1, 2013.