Filters








5 Hits in 3.0 sec

Telefonica Research at TRECVID 2010 Content-Based Copy Detection

Ehsan Younessian, Xavier Anguera, Tomasz Adamek, Nuria Oliver
2010 TREC Video Retrieval Evaluation  
This notebook paper presents the participation of Telefonica Research in the task of Video Copy Detection in TRECVID 2010.  ...  and with a temporal consistency postprocessing. • TID.m.BALANCED.audioonly: This submission is based on the monomodal audio-based system using frequency-based audio local features.  ...  ACKNOWLEDGMENT Telefonica I+D participates in Torres Quevedo subprogram (MICINN), cofinanced by the European Social Fund, for Researchers recruitment. X. Anguera and T.  ... 
dblp:conf/trecvid/YounessianAAO10 fatcat:fjahcf3so5aw3elv4ftedrqur4

Telefonica Research at TRECVID 2011 Content-Based Copy Detection

Xavier Anguera, Tomasz Adamek, Daru Xu, Juan Manuel Barrios
2011 TREC Video Retrieval Evaluation  
This notebook paper summarizes the algorithms behind Telefonica Research participation in the NIST-TRECVID 2011 evaluation on the Video Copy Detection task.  ...  This year we have focused on 1) Improving the image-based matching system to better process video files; 2) implemented and tested a novel audio local fingerprint; and 3) improved the multimodality fusion  ...  TELEFONICA RESEARCH MULTIMODAL VIDEO COPY DETECTION The system presented by Telefonica in the video copy detection task at Trecvid is based on the fusion at decision level of several monomodal systems,  ... 
dblp:conf/trecvid/AngueraAXB11 fatcat:evwqqyk2urhvhhsdz3pthr55cm

Multimodal fusion for video copy detection

Xavier Anguera, Juan Manuel Barrios, Tomasz Adamek, Nuria Oliver
2011 Proceedings of the 19th ACM international conference on Multimedia - MM '11  
Content-based video copy detection algorithms (CBCD) focus on detecting video segments that are identical or transformed versions of segments in a known video.  ...  We performed several tests using 2010 TRECVID VCD datasets and obtain up to 46% relative improvement in min-NDCR while also improving the F1 metric on the fused results in comparison to just using the  ...  EXPERIMENTS We tested the suitability of the proposed algorithm using the content-based copy detection dataset used in the 2010 NIST TRECVID benchmarking campaign [12] .  ... 
doi:10.1145/2072298.2071979 dblp:conf/mm/AngueraBAO11 fatcat:j2nodh7md5f4dfixiiutiorilq

Content-based video copy detection

Juan Manuel Barrios
2009 Proceedings of the seventeen ACM international conference on Multimedia - MM '09  
In particular, Content-Based Video Copy Detection (CBVCD) addresses the need to retrieve videos in a collection that are copies of some original document.  ...  The copy detection process relies exclusively on the audiovisual content, and must be robust to common audiovisual transformations. This thesis details a novel CBVCD system, called P-VCD.  ...  The system participated on the Content-Based Copy Detection task at TRECVID 2010 and 2011.  ... 
doi:10.1145/1631272.1631539 dblp:conf/mm/Barrios09 fatcat:kpjhi2p3orcsrfi444ewana33i

Voice Modeling Methods for Automatic Speaker Recognition

Thilo Stadelmann, Mathematik Und Informatik, Freisleben, Bernd (Prof. Dr.)
2011
TRECVid offers a task for semantic concept detection called "high-level feature detection" [Smeaton et al. 2009 ] (since 2010: "semantic indexing") in which the Videana-based multimodal video content  ...  at TRECVid'07, grouped by Videana's ability to detect them.  ...  They assess the performance of each proband for the given dataset and have been computed on a per-utterance basis rather than based on samples to give equal weight to each misclassified utterance regardless  ... 
doi:10.17192/z2010.0465 fatcat:qik4of45xbbrjm3zgwxo7dhrcu