Filters








14 Hits in 1.7 sec

FXPAL at TRECVID 2005

Matthew Cooper, John Adcock, Robert Chen, Hanning Zhou
2005 TREC Video Retrieval Evaluation  
In 2005 FXPAL submitted results for 3 tasks: shot boundary detection (section 1, page 1), high-level feature extraction (section 2, page 5), and interactive search (section 3, page 7).  ...  The bottom columns show average TRECVID and average FXPAL results. Figure 3 : 3 Figure 3: MAP performance of all interactive search submissions with FXPAL submissions shown in red.  ...  System MAP score Notes FXPAL 1 LCN 0.348 FXPAL 0 LN 0.345 Post-processing identical to 2004 text-only system LSA1 FXPAL 4 LC 0.339 FXPAL 2 RAN 0.329 FXPAL 3 RPN 0.323 FXPAL 5 RA 0.311 User ID-ed shots  ... 
dblp:conf/trecvid/0002ACZ05 fatcat:5oes2joegja73jhxdvcauitumq

Video shot boundary detection: Seven years of TRECVid activity

Alan F. Smeaton, Paul Over, Aiden R. Doherty
2010 Computer Vision and Image Understanding  
year (2005) as an example.  ...  Automatic SBD was one of the tracks of activity within the annual TRECVid benchmarking exercise, each year from 2001 to 2007 inclusive.  ...  Precision and recall were the primary measures used in the presentation of results at TRECVid.  ... 
doi:10.1016/j.cviu.2009.03.011 fatcat:w5kluejxi5bfrfrwq7jku7jgda

FXPAL at TRECVID 2006

Matthew Cooper, John Adcock, Francine Chen
2006 TREC Video Retrieval Evaluation  
Figure 2 : 2 Figure 2: Visualizing the performance of FXPAL shot boundary systems relative to other TRECVID participants: mean performance for SB05 (left) and SB06 (right). 1 . 1 FXPAL1LN: LSA text query  ...  This is similar to our relative performance in 2004 and 2005. Figure 3 shows the MAP performance of the FXPAL runs against the entire set of interactive search submissions.  ... 
dblp:conf/trecvid/0002AC06 fatcat:6nyxbq3zuvgqxg5irl6lnsd4zi

FXPAL Interactive Search Experiments for TRECVID 2007

John Adcock, Jeremy Pickens, Matthew Cooper, Lisa Anthony, Francine Chen, Pernilla Qvarfordt
2007 TREC Video Retrieval Evaluation  
In 2007 FXPAL submitted results for two tasks: rushes summarization and interactive search. The rushes summarization task has been described at the ACM Multimedia workshop [1] .  ...  The other submitted runs (FXPAL CO*) employed a novel real-time, multi-user, collaborative search system.  ...  We have yet to perform statistical significance analysis on these results but similarly small differences in 2005 proved to have a very high level of statistical significance.  ... 
dblp:conf/trecvid/AdcockPCACQ07 fatcat:pujhbmwt4vah7bq3hmajblrlim

Recent Advances and Open Issues of Digital Image/Video Search

Shih-Fu Chang
2007 Eighth International Workshop on Image Analysis for Multimedia Interactive Services (WIAMIS '07)  
Research Intel AT&T FXPAL University of Amsterdam Oxford University Nanyang Technological University, Singapore National Taiwan University Tsinghua University KDDI, Japan Dublin City University  ...  E C --error rate of independent detector for i C Insufficient Precision of Social Tags834 concepts defined, 449 concepts annotated Labeled over 61,000 shots of TRECVID 2005 data set 33 Million judgments  ... 
doi:10.1109/wiamis.2007.70 dblp:conf/wiamis/Chang07 fatcat:poprgo5kwbbtddugdbpf3zjosq

TRECVid 2005 Experiments at Dublin City University

Colum Foley, Cathal Gurrin, Gareth J. F. Jones, Hyowon Lee, Sinéad McGivney, Noel E. O'Connor, Sorin Sav, Alan F. Smeaton, Peter Wilkins
2005 TREC Video Retrieval Evaluation  
In this paper we describe our experiments in the automatic and interactive search tasks and the BBC rushes pilot task of TRECVid 2005.  ...  Interactive Search System This year marks a departure from the standard DCU TRECVid search systems where the user interface was rendered by a web browser.  ...  Preliminary results from this and some analysis will be presented at the TRECVid workshop.  ... 
dblp:conf/trecvid/FoleyGJ0MOSSW05 fatcat:6qitz4g3mvc3tmzeljuwrgtpze

Shot Boundary Detection and Low-Level Feature Extraction Experiments for TRECVID 2005

Kazunori Matsumoto, Masaru Sugano, Masaki Naito, Keiichiro Hoashi, Haruhisa Kato, Masami Shshibori, Kenji Kita, Fumiaki Sugaya, Yasuyuki Nakajima
2005 TREC Video Retrieval Evaluation  
Short gradual cuts are trained from the subset of 2005 develop. bs-9: Result of bs-1's grad and that of bs-7's cut are merged. bs-10: Result of bs-1's grad and that of bs-8's cut are merged.  ...  bs-1: Compressed domain approach, which corresponds to the best options in TRECVID 2004 with newly introduced luminance adaptive threshold. bs-2: Compressed domain approach with newly introduced luminance  ...  2005 (RunID=1) and TRECVID 2004 (RunID=5).  ... 
dblp:conf/trecvid/MatsumotoSNHKSK05 fatcat:6ultu46wt5cnfjn2vsmzaixjo4

The trecvid 2007 BBC rushes summarization evaluation pilot

Paul Over, Alan F. Smeaton, Philip Kelly
2007 Proceedings of the international workshop on TRECVID video summarization - TVS '07  
It was carried out under the auspices of TRECVID.  ...  Procedures for developing ground truth lists of important segments from each video were developed at Dublin City University and applied to the BBC video.  ...  Ireland under grant number 03/IN.3/I361 and the team at Dublin City University for creating the ground truth, to several sites for mirroring the video data to allow distribution to participating groups  ... 
doi:10.1145/1290031.1290032 dblp:conf/mm/OverSK07 fatcat:yscn3od5rzhrjl3horbrzxpctq

Efficient targeted search using a focus and context video browser

Ork De Rooij, Marcel Worring
2012 ACM Transactions on Multimedia Computing, Communications, and Applications (TOMCCAP)  
Examples from recent TRECVID benchmarks include the Informedia Storyboard interface [Christel and Yan 2007] and the FXPal MediaMagic [Adcock et al. 2007 ], which use a recognizable grid layout to browse  ...  In literature, and at benchmarks like TRECVID, we see a variety of systems, each with unique approaches to interactive search.  ... 
doi:10.1145/2379790.2379793 fatcat:rbeuvtv3d5ftlhsw3slbdr7js4

Scenario based dynamic video abstractions using graph matching

JeongKyu Lee, JungHwan Oh, Sae Hwang
2005 Proceedings of the 13th annual ACM international conference on Multimedia - MULTIMEDIA '05  
In order to evaluate the effectiveness of the proposed SBD using graph matching (GM-SBD), we compare it with the SBD proposed by FXPAL Lab (FX-PAL) [1] , which has one of the best methods in TRECVID 2004  ...  In case of gradual changes, GM-SBD is more accurate than FXPAL.  ... 
doi:10.1145/1101149.1101328 dblp:conf/mm/LeeOH05 fatcat:lo2l75bwkngm3aekmoqn3bd7r4

TRECVID 2016: Evaluating Video Search, Video Event Detection, Localization, and Hyperlinking

George Awad, Jonathan G. Fiscus, David Joy, Martial Michel, Alan F. Smeaton, Wessel Kraaij, Georges Quénot, Maria Eskevich, Robin Aly, Roeland Ordelman, Marc Ritter, Gareth J. F. Jones (+2 others)
2016 TREC Video Retrieval Evaluation  
The Video-to-Text work has been partially supported by Science Foundation Ireland (SFI) as a part of the Insight Centre at DCU (12/RC/2289).  ...  F1504054U; Science Foundation Ireland (SFI) as a part of the ADAPT Centre at DCU (13/RC/2106), the Dutch National Research Programme COMMIT/ and the CLARIAH (www.clariah.nl) project.  ...  For participants in this Grand Challenge task, the majority of participants used CNNs and RNNs just as in VTT at TRECVid 2016, but unlike TRECVid, the ACM MM Grand Challenge setup does not give an opportunity  ... 
dblp:conf/trecvid/AwadFJMSKQEAORJ16 fatcat:ly333heczrgrbfubo4wfr7qrgy

Beyond actions: Exploring the discovery of tactics from user logs

Jiyin He, Pernilla Qvarfordt, Martin Halvey, Gene Golovchinsky
2016 Information Processing & Management  
In addition, as the resulting analysis is at a tactical level we reduce potential issues surrounding the need for interpretation of low-level user actions for log analysis.  ...  Alternatively, in lab studies user activities can be analysed at a higher level, such as search tactics and strategies, abstracted away from detailed interface implementation.  ...  He was instrumental for this work by bringing all of us together during a few months at FXPAL, where the foundation of this research project was laid.  ... 
doi:10.1016/j.ipm.2016.05.007 fatcat:c4q3divevzdz7h3akx2oq4p6wi

Combining Relevance Information in a Synchronous Collaborative Information Retrieval Environment [chapter]

Colum Foley, Alan F. Smeaton, Gareth J.F. Jones
Collaborative and Social Information Retrieval and Access  
Two versions of Físchlár-DiamondTouch were evaluated in TRECVid 2005 (Foley et al., 2005) , one which provided for increased awareness amongst users and one which was designed for improved group efficiency  ...  In an effort to improve collaborative search effectiveness through "algorithmically-mediated collaboration", the "Cerchiamo" system was developed by the FXPAL TRECVid team (Adcock et al., 2007) .  ... 
doi:10.4018/978-1-60566-306-7.ch007 fatcat:wc7bxilmsrep7gp6n5bfhkcfk4

Multimodal Indexing of Presentation Videos

Michele Merler
2017
This work aims at recognizing visual content "in the wild", where the system cannot rely on any additional information besides the video itself.  ...  Recently researchers at FXPAL have introduced Talkminer [Adcock et al., 2010] , which is the first large scale fully automatic presentation video indexing system.  ...  I want to thank Yu-Gang Jiang and Yadong Mu for the great help and collaboration on the Trecvid MED tasks.  ... 
doi:10.7916/d8rv0vxx fatcat:biib47liqvgu3bqesf54avdjhu