A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Filters
K-Space at TRECvid 2006
2006
TREC Video Retrieval Evaluation
Preserved Fulltext
In this paper we describe the K-Space participation in TRECVid 2006. K-Space participated in two tasks, highlevel feature extraction and search. ...
Acknowledgments The research leading to this paper was supported by the European Commission under contract FP6-027026 (K-Space). ...
Conclusion We have presented the K-Space participation in TRECVid 2006. ...
dblp:conf/trecvid/WilkinsAFHJKMMO06
fatcat:b4kgh4uhqbhpbk3yoxvfo5y6xe
Content-based video retrieval: Three example systems from TRECVid
2008
International journal of imaging systems and technology (Print)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
which demonstrate the state of the art in interactive, content-based retrieval of video shots, and these three are just three of the more than 20 systems developed for the 2007 iteration of the annual TRECVid ...
Figure 1 . 1 User interface for Dublin City University K-Space Search System. [Color figure can be viewed in the online issue, which is available at www.interscience. wiley.com.] ...
Figure 2 . 2 User interface for Dublin City University/K-Space Search System. [Color figure can be viewed in the online issue, which is available at www. interscience.wiley.com.] ...
doi:10.1002/ima.20150
fatcat:zq4xgfb5lfgfpaohrnpir4wnw4
Merging storyboard strategies and automatic retrieval for improving interactive video search
2007
Proceedings of the 6th ACM international conference on Image and video retrieval - CIVR '07
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
For TRECVID 2006 the output of automatic search was included for the first time with storyboards, both as an option for an interactive user and in a different run as the sole means of access. ...
Further, the full Informedia interface with automatic search results as an option along with other query mechanisms scored significantly better than all other TRECVID 2006 interactive search systems. ...
This collection was automatically divided into shots at the Fraunhofer (Heinrich Hertz) Institute in Berlin, producing a common shot reference of 79,484 TRECVID shots used in scoring across TRECVID 2006 ...
doi:10.1145/1282280.1282351
dblp:conf/civr/ChristelY07
fatcat:morsepvdybgjnefd2pdlc7auzy
Multi-Lingual Broadcast News Retrieval
2006
TREC Video Retrieval Evaluation
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
In this notebook paper we describe the technical details of the submissions to TRECVID 2006 from CMU Informedia team. ...
High-level feature extraction We submitted 6 runs for TRECVID 2006 high level feature evaluation, as shown in Table 1 . There were 61901 labeled shots for the 39 concepts of the LSCOM-Lite set. ...
In TRECVID 2006, we predict fusion set and test data (the new trecvid 2006 data) by the models we built from our training set. ...
dblp:conf/trecvid/HauptmannCCDLY006
fatcat:wjl7ftqu4fem5cmfaet7vmq2sm
Towards optimal bag-of-features for object categorization and semantic video retrieval
2007
Proceedings of the 6th ACM international conference on Image and video retrieval - CIVR '07
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2008; you can also visit the original URL.
The file type is application/pdf.
By incorporating the BoF with color and texture features, an improvement of 50% is reported on TRECVID-2006 dataset. ...
On both PASCAL-2005 and TRECVID-2006 datasets, our BoF setting generates competitive performance compared to the state-of-the-art techniques. ...
K linear (x, y) = x T y. (3) K d−RBF (x, y) = e −ρd(x,y) , (4) where d(x, y) can be chosen to be any distance in the feature space. ...
doi:10.1145/1282280.1282352
dblp:conf/civr/JiangNY07
fatcat:ww26ir6h5ng7nfxcfbnvqilhbu
IBM Research TRECVID-2006 Video Retrieval System
2006
TREC Video Retrieval Evaluation
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
In this paper, we describe the IBM Research system for indexing, analysis, and retrieval of video as applied to the TREC-2006 video retrieval benchmark. ...
We used different partitioning for TRECVID 2005 and TRECVID 2006 training, and we leveraged both to build final models in 2006. ...
benchmark: up to 53.43 % for 2005 and 21.54 % 2006 TRECVID topics. ...
dblp:conf/trecvid/CampbellHEJNNSS06
fatcat:t2wpc5feqnedljhgcxy7lzx4hq
UEC at TRECVID 2007 High Level Feature Task
2007
TREC Video Retrieval Evaluation
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
In this paper, we describe our approach and results for high-level feature extraction task at TRECVID 2007. This year, we adopted late fusion of several types of features. ...
data for Run 2 and Run 6, only TRECVID 2006 training data for Run 4, and only TRECVID 2007 training data for the others. ...
INTRODUCTION In TRECVID 2006, we extracted single type of feature vectors from each test images, and classified test keyframes by the machine learning method such as support vector machines (SVM). ...
dblp:conf/trecvid/LiuTY07
fatcat:xxauisfe2rddjn3apn36gkjjqi
TRECVid 2007 Experiments at Dublin City University
2007
TREC Video Retrieval Evaluation
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2016; you can also visit the original URL.
The file type is application/pdf.
In this paper we describe our retrieval system and experiments performed for the automatic search task in TRECVid 2007. ...
Made use of query-time generation of retrieval expert coefficients for fusion. • F A 2 DCU-ImgOnlyEnt5 : Automatic generation of retrieval expert coefficients for fusion at index time. • F A 2 DCU-imgOnlyEntHigh3 ...
Acknowledgements This work was partially supported by Science Foundation Ireland under grant 03/IN.3/I361 and by the European Commission under contract FP6-027026 (K-Space). ...
dblp:conf/trecvid/WilkinsAJOS07
fatcat:kgcibmabzrd7xf6znqyzfsbo5u
Improving Automatic Video Retrieval with Semantic Concept Detection
[chapter]
2009
Lecture Notes in Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
The experiments use the common video data and sets of queries from three successive TRECVID evaluations. ...
Due to the limited space, the search topics are not listed here, but are available in the TRECVID guidelines documents at http://www-nlpir.nist. ...
TRECVID The video material and the search topics used in these experiments are from the TRECVID evaluations [2] in 2006-2008. ...
doi:10.1007/978-3-642-02230-2_49
fatcat:xrfn3gpljffxnm7mhyx4iqnyge
Glasgow University at TRECVID 2007
2007
TREC Video Retrieval Evaluation
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
In this paper we describe our experiments in the automatic search task of TRECVid 2007. For this we have implemented a new video search technique based on SIFT features and manual annotation. ...
Acknowledgements The research leading to this paper was supported by European Commission under contracts FP6-027026(K-Space) and FP6-027122(Salero). ...
The first author acknowledges the kind help rendered by Frank Hopfgartner at various stages of submission. ...
dblp:conf/trecvid/PunithaJ07
fatcat:rpyp6wtlvjbsrkgb35ll2roami
University of Central Florida at TRECVID 2006 High-Level Feature Extraction and Video Search
2006
TREC Video Retrieval Evaluation
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
In this paper, we describe our experiments in high-level features extraction and interactive topic search tasks of TRECVID 2006. ...
plots the performance comparison of our runs to all the TRECVID 2006 runs in term of average precision. ...
In each run, the temporal "K-nearest neighbor" method is used as the final step in the relevant feedback. ...
dblp:conf/trecvid/LiuZBOKNBS06
fatcat:7itptghdzbbynmadnuip2m3iea
K-Space at TRECVID 2007
2007
TREC Video Retrieval Evaluation
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2016; you can also visit the original URL.
The file type is application/pdf.
Acknowledgments The research leading to this paper was supported by the European Commission under contract FP6-027026 (K-Space). ...
In contrast to TRECVid 2006 the face detection was applied only both to the keyframes and the K-frames within a shot. ...
This paper describes the K-Space participation in TRECVid 2007. ...
dblp:conf/trecvid/WilkinsABJLKMOSAOBGHELVPVMKSAMSBPCIHGSCSP07
fatcat:xvegdrcqmjfp3avqitwnq3rhgq
[Invited Paper] Semantic Indexing for Large-Scale Video Retrieval
2016
ITE Transactions on Media Technology and Applications
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
We also show the TokyoTech system using Gaussian-mixture-model (GMM) supervectors and deep convolutional neural networks (CNNs) with its experimental evaluation at TRECVID 2014. ...
This paper reviews TRECVID activities with these techniques for semantic indexing. ...
For example, the best system at the TRECVID 2014 is a hybrid system 70) of bag-of-words and deep learning. ...
doi:10.3169/mta.4.209
fatcat:pewjuvxonfgxnm5onv5e5xyvlu
Inexpensive fusion methods for enhancing feature detection
2007
Signal processing. Image communication
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
At multiple stages throughout these frameworks, various pieces of evidence can be fused together in order to boost performance. ...
Our approaches, validated experimentally on TRECVid data, are designed to be complementary to existing frameworks and can be regarded as possible replacements for the more computationally expensive combination ...
We would like to extend our thanks to our TRECVid K-Space partners who provided the specific feature detectors used throughout this series of experiments. ...
doi:10.1016/j.image.2007.05.012
fatcat:mm3q3o6nd5f55nezucj2fxiblu
Inexpensive Fusion Methods for Enhancing Feature Detection
2007
2007 International Workshop on Content-Based Multimedia Indexing
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
At multiple stages throughout these frameworks, various pieces of evidence can be fused together in order to boost performance. ...
Our approaches, validated experimentally on TRECVid data, are designed to be complementary to existing frameworks and can be regarded as possible replacements for the more computationally expensive combination ...
We would like to extend our thanks to our TRECVid K-Space partners who provided the specific feature detectors used throughout this series of experiments. ...
doi:10.1109/cbmi.2007.385400
dblp:conf/cbmi/WilkinsAOS07
fatcat:j7pcn7crzrgktfhm2rsnc2td2m
« Previous
Showing results 1 — 15 out of 533 results