Filters








2,186 Hits in 3.2 sec

Multi-concept learning with large-scale multimedia lexicons

Lexing Xie, Rong Yan, Jun Yang
2008 2008 15th IEEE International Conference on Image Processing  
This paper aims to answer two questions related to multi-concept learning: does a large-scale lexicon help concept detection? how many concepts are enough?  ...  Multi-concept learning is an important problem in multimedia content analysis and retrieval.  ...  In this paper, we investigate multi-concept learning, i.e., how to leverage large-scale multimedia lexicons for semantic concept detection.  ... 
doi:10.1109/icip.2008.4712213 dblp:conf/icip/XieYY08 fatcat:7h2pim3wcjfsppg37yptw6wqxi

Extracting Semantics from Multimedia Content: Challenges and Solutions [chapter]

Lexing Xie, Rong Yan
2008 Signals and Communication Technology  
We then present challenges for each of the five components along with their existing solutions: designing multimedia lexicons and using them for concept detection, handling multiple media sources and resolving  ...  correspondence across modalities, learning structured (generative) models to account for natural data dependency or model hidden topics, handling rare classes, leveraging unlabeled data, scaling to large  ...  This effort has lead to a Large-Scale Concept Ontology for Multimedia (LSCOM) [61] , and an interim result of this has resulted in 39 high-level features (concepts) definitions and annotations dubbed  ... 
doi:10.1007/978-0-387-76569-3_2 fatcat:jul6fw7esfaurct6erjnvpcq6q

Semantic Image and Video Indexing in Broad Domains

Marcel Worring, Guus Schreiber
2007 IEEE transactions on multimedia  
Fan [1] defines a hierarchical learning method which explicitly takes the structure into account. Learning a large lexicon of concepts is a computationintensive process.  ...  New machine learning techniques, such as one-class classifiers, have been developed which deal with the special characteristics of multimedia data with sparse and heterogeneous examples.  ... 
doi:10.1109/tmm.2007.898913 fatcat:xagm5plk2nbsvh3nm4thgeqp5i

Multimedia content analysis and search

HongJiang Zhang
2009 Proceedings of the seventeen ACM international conference on Multimedia - MM '09  
• Many models developed -Machine learning is the core -Success in relative small-scale image databases • Given a feature space, can we identify high-level concepts with small semantic gaps?  ...  -Images with small semantic gaps are selected and clustered by a confidence map and content-context similarity matrix -Mine a concept lexicon with small semantic gaps and high cooccurrences from the surrounding  ...  25% of search results pointing to totally irrelevant websites; 30% of searches were given up at the end, due to non satisfying results; 35% of the users are not happy with search results; 40% of the users  ... 
doi:10.1145/1631272.1631274 dblp:conf/mm/Zhang09 fatcat:c6bd77qd7rgxbf4f7ynwyvsg7a

Content-Based Video Retrieval in Historical Collections of the German Broadcasting Archive [chapter]

Markus Mühling, Manja Meister, Nikolaus Korfhage, Jörg Wehling, Angelika Hörth, Ralph Ewerth, Bernd Freisleben
2016 Lecture Notes in Computer Science  
The uniqueness and importance of the video material stimulates a large scientific interest in the video content.  ...  It consists of video analysis algorithms for shot boundary detection, concept classification, person recognition, text recognition and similarity search.  ...  While query-by-content based on low-level features turned out to be insuf- ficient to search successfully in large-scale multimedia databases, image repre- sentations learned by deep neural networks greatly  ... 
doi:10.1007/978-3-319-43997-6_6 fatcat:3v5tz7pju5bedpnsp23ar4rr5q

Semantic concept-based query expansion and re-ranking for multimedia retrieval

Apostol (Paul) Natsev, Alexander Haubold, Jelena Tešić, Lexing Xie, Rong Yan
2007 Proceedings of the 15th international conference on Multimedia - MULTIMEDIA '07  
In particular, we explore the utility of a fixed lexicon of visual semantic concepts for automatic multimedia retrieval and re-ranking purposes.  ...  A multi-modal fusion step leverages the identified relevant concepts to expand and re-rank the results. Concept-based expanded and re-ranked results Multimedia Repository  ...  Most large-scale multimedia search systems typically rely on text-based search over media metadata such as surrounding html text, anchor text, titles and abstracts.  ... 
doi:10.1145/1291233.1291448 dblp:conf/mm/NatsevHTXY07 fatcat:lqjfylum3bgwtkepponnjx6f3i

Searching Videos in Visual Semantic Spaces [chapter]

Eric Zavesky, Zhu Liu, Dave Gibbon, Behzad Shahraray
2010 Semantic Computing  
from the LSCOM lexicon.  ...  In this chapter, we first introduce state of the art of visual semantics extraction technologies and then present our work on fast exploration of a large set of video data using 374 semantic concepts derived  ...  Encouraging ideas for new video processing methods, the Disruptive Technology Office (DTO) sponsored the Large-Scale Concept Ontology for Multimedia (LSCOM) workshop to develop an expanded multimedia  ... 
doi:10.1002/9780470588222.ch16 fatcat:mpwszar34vfcfeunfrqptexkr4

Exploiting Concept Association to Boost Multimedia Semantic Concept Detection

Sheng Gao, Xinglei Zhu, Qibin Sun
2007 2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07  
We evaluate its efficiency on the task of semantic concept detection on the large-scale news video dataset from TRECVID 2005 development set.  ...  In the paper we study the efficiency of semantic concept association in multimedia semantic concept detection.  ...  Large Scale Concept Ontology for Multimedia), we have large-scale annotated video corpus for semantic concept lexicon of moderate size.  ... 
doi:10.1109/icassp.2007.366074 dblp:conf/icassp/GaoZS07 fatcat:6tsqvroki5ggjf3s4wuegofxga

Pattern Mining in Visual Concept Streams

Lexing Xie, Shih-fu Chang
2006 2006 IEEE International Conference on Multimedia and Expo  
Results also show that the majority of the target concepts are better predicted with temporal or combination hypotheses, and there are novel concepts found that are not part of the original lexicon.  ...  This paper presents the first effort on temporal pattern mining in the large concept space.  ...  These data come from the Large Scale Concept Ontology for Multimedia (LSCOM) challenge workshop, they provide a valuable basis for assessing the mining strategies in video.  ... 
doi:10.1109/icme.2006.262457 dblp:conf/icmcs/XieC06 fatcat:huogrqs6yzcvzc52sekwnc4kmm

Interactive multi-user video retrieval systems

Marco Bertini, Alberto Del Bimbo, Andrea Ferracani, Lea Landucci, Daniele Pezzatini
2011 Multimedia tools and applications  
In this paper we present two interactive multi-user systems for video search and browsing.  ...  The second system implements a multiuser collaborative application within a single location, exploiting multi-touch devices.  ...  Automatic video annotation systems are based on large sets of concept classifiers [50] , typically based on supervised machine learning techniques such as SVMs.  ... 
doi:10.1007/s11042-011-0888-9 fatcat:jjujazyudbechkzojnkfp3i34e

Visual Event Detection using Multi-Dimensional Concept Dynamics

Shahram Ebadollahi, Lexing Xie, Shih-fu Chang, John Smith
2006 2006 IEEE International Conference on Multimedia and Expo  
Visual events are viewed as stochastic temporal processes in the semantic concept space.  ...  In this concept-centered approach to visual event modeling, the dynamic pattern of an event is modeled through the collective evolution patterns of the individual semantic concepts in the course of the  ...  We employ the 39 semantic concepts of LSCOM-Lite [8] , which are the interim results of the effort in developing a Large-Scale Concept Ontology for Multimedia (LSCOM) [10] .  ... 
doi:10.1109/icme.2006.262691 dblp:conf/icmcs/EbadollahiXCS06 fatcat:zit6pffvfrh6dol2ek36exfzhy

The MediaMill TRECVID 2005 Semantic Video Search Engine (Draft Version)

Cees G. M. Snoek, Jan C. van Gemert, Jan-Mark Geusebroek, Bouke Huurnink, Dennis C. Koelma, Giang P. Nguyen, Ork de Rooij, Frank J. Seinstra, Arnold W. M. Smeulders, Cor J. Veenman, Marcel Worring
2005 TREC Video Retrieval Evaluation  
The results show that an optimal strategy for generic multimedia analysis is one that learns from the training set on a per-concept basis which tactic to follow.  ...  We performed a large set of experiments (runid: B vA).  ...  We advocate that the ideal multimedia retrieval system should first learn a large lexicon of concepts, based on multimedia analysis, to be used for the initial search.  ... 
dblp:conf/trecvid/SnoekGGHKNRSSVW05 fatcat:wymhnhhlsvh6fnpnqgsnecmgie

Zero-Shot Event Detection Using Multi-modal Fusion of Weakly Supervised Concepts

Shuang Wu, Sravanthi Bondugula, Florian Luisier, Xiaodan Zhuang, Pradeep Natarajan
2014 2014 IEEE Conference on Computer Vision and Pattern Recognition  
We leverage video and image collections with free-form text descriptions from widely available web sources to learn a large bank of concepts, in addition to using several off-the-shelf concept detectors  ...  In this paper, we present a general framework for the zeroshot learning problem of performing high-level event detection with no training exemplars, using only textual descriptions.  ...  concept space using statistics learned on a large text corpus.  ... 
doi:10.1109/cvpr.2014.341 dblp:conf/cvpr/WuBLZN14 fatcat:ep7juxwv2jgubhlx3ezgyjomrq

Recent Advances and Challenges of Semantic Image/Video Search

Shih-Fu Chang, Wei-Ying Ma, Arnold Smeulders
2007 2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07  
We present an overview of recent advances and major challenges in image and video search, with a specific focus on large-scale semantic concept detection and indexing.  ...  Such semantic indexing paradigm has been driven by the increasing availability of the large resources of corpora, novel labeling approaches, innovative image features, and machine learning techniques for  ...  To define a suitable set of semantic concepts, a recent effort has also been completed to define a Large-Scale Concept Ontology for Multimedia (LSCOM) [2] .  ... 
doi:10.1109/icassp.2007.367292 dblp:conf/icassp/ChangMS07 fatcat:wgnw233c4ndbtc5zrvrb4g5wqe

Toward an Enhancement of Textual Database Retrieval Using NLP Techniques [chapter]

Asanee Kawtrakul, Frederic Andres, Kinji Ono, Chaiwat Ketsuwan, Nattakan Pengphon
2001 Lecture Notes in Computer Science  
The quality of service over a large-scale network is provided by using AHYDS-Active HYpermedia Delivery System-framework.  ...  This paper presents the project named VLSHDS-Very Large Scale Hypermedia Delivery System. The quality of textual information search is enhanced by using NLP techniques.  ...  An Overview of the Very Large Scale Hypermedia Delivery Systems The key architectural components in the VLSHDS platform used as textual database platform is shown in Figure 1 .  ... 
doi:10.1007/3-540-45399-7_15 fatcat:sstaj5qr7navfinnfhc2gkr33q
« Previous Showing results 1 — 15 out of 2,186 results