SHAMUS
UFAL Search and Hyperlinking Multimedia System
SHAMUS
SHAMUS (UFAL Search and Hyperlinking Multimedia System) is a system for an easy search and navigation in multimedia archives. The system consists of a text-based Search, an automatic selection of Anchoring segments and video-based retrieval of Hyperlinks.
The segments of the videos in a collection relevant to the user-typed query are provided by the Search component. The Anchoring component determines the most important segments of the videos. The beginnings of the most informative segments are marked in each video and the transcription of the beginning of each Anchoring segment is displayed. The segments topically related to each Anchoring segments (the Hyperlinks) are then retrieved. The list of the segments is generated on the fly for each Anchoring segment. All three system components use Terrier IR Platform.
Demo
A demo of the system available online. Demo currently works on a collection of 1219 TED talks.
The source codes are available under the GNU General Public License and could be downloaded from the git repository:
git clone https://redmine.ms.mff.cuni.cz/shamus-proj/shamus.git
Reference
Demo:
Galuščáková Petra, Saleh Shadi, Pecina Pavel: SHAMUS: UFAL Search and Hyperlinking Multimedia System. In: Proceedings of the 38th European Conference on IR Research, ECIR 2016, ISBN 978-3-319-30670-4, pp 853-856, 2016.
Search component:
Galuščáková Petra, Pecina Pavel: CUNI at MediaEval 2014 Search and Hyperlinking Task: Search Task Experiments. In: Working Notes Proceedings of the MediaEval 2014 Workshop, Copyright © CEUR-WS.org, Aachen, Germany, ISSN 1613-0073, 2014 [biblio]
Anchoring segment selection:
Galuščáková Petra, Pecina Pavel: CUNI at MediaEval 2015 Search and Anchoring in Video Archives: Anchoring via Information Retrieval. In: Working Notes Proceedings of the MediaEval 2015 Workshop, Copyright © CEUR-WS.org, Aachen, Germany, ISSN 1613-0073, 2015 [biblio]
Hyperlinking component:
Galuščáková Petra, Kruliš Martin, Lokoč Jakub, Pecina Pavel: CUNI at MediaEval 2014 Search and Hyperlinking Task: Visual and Prosodic Features in Hyperlinking. In: Working Notes Proceedings of the MediaEval 2014 Workshop, Copyright © CEUR-WS.org, Aachen, Germany, ISSN 1613-0073, 2014 [biblio]
Galuščáková Petra, Batko Michal, Kruliš Martin, Lokoč Jakub, Novák David, and Pecina Pavel: CUNI at TRECVID 2015 Video Hyperlinking Task. In: TRECVID 2015 Workshop Notebook, Gaithersburg, MD, USA, 2015. [poster]
Terrier IR Platform:
Ounis Iadh, Amati Gianni, Plachouras Vassilis, He Ben, Macdonald Craig, and Lioma Christina. Terrier: A High Performance and Scalable Information Retrieval Platform. In: Proceedings of ACM SIGIR'06 Workshop on Open Source Information Retrieval (OSIR 2006). 10th August, 2006. Seattle, Washington, USA.
Contact
Acknowledgement
The work on the project is supported by the Czech Science Foundation, grant number P103/12/G084, Charles University Grant Agency GA UK, grant number 920913, and by SVV project number 260 224.