Filters








2,655 Hits in 4.2 sec

Automatic discovery of query-class-dependent models for multimodal search

Lyndon S. Kennedy, Apostol (Paul) Natsev, Shih-Fu Chang
2005 Proceedings of the 13th annual ACM international conference on Multimedia - MULTIMEDIA '05  
We develop a framework for the automatic discovery of query classes for query-class-dependent search models in multimodal retrieval.  ...  We also find that some hand-defined query classes, such as "Named Person" and "Sports" do, indeed, have similarities in search method performance and are useful for query-class-dependent multimodal search  ...  QUERY CLASS DISCOVERY Having seen the limitations of query-independent fusion in multimodal search, some recent efforts in the TRECVID video search task, have implemented query-class-dependent models and  ... 
doi:10.1145/1101149.1101339 dblp:conf/mm/KennedyNC05 fatcat:g6lspob5vfemfl7zcewuywqsvi

Image Retrieval and Re-Ranking Techniques - A Survey

Mayuri D. Joshi, Revati M. Deshmukh, Kalashree N.Hemke, Ashwini Bhake, Rakhi Wajgi
2014 Signal & Image Processing An International Journal  
This re-ranking process aids the mutual transfer of information across multiple modalities to improvise search performance to facilitate authentic searching.  ...  In contrast to the existing techniques, the re-ranking method fosters the interaction between modalities to find a tranquillity which is helpful for re-ranking.  ...  Kennedy et al, [9] , proposed a query class dependent search models in multimodal retrieval for the automatic discovery of query classes.  ... 
doi:10.5121/sipij.2014.5201 fatcat:gusz6wizpbgsfgjt6asf7s44fi

A Survey on Visual Search Reranking

Thalla Shankar, Lalitha Manglaram, Murali Sadak
2014 IOSR Journal of Computer Engineering  
This paper suggests a new kind of reranking algorithm, the circular reranking, that supports the mutual exchange of information across multiple modalities for improving search performance, and follows  ...  However the problem is not trivial especially when we are considering multiple features or modalities for search in image and video retrieval.  ...  Kennedy et al, [9] , proposed a query class dependent search models in multimodal retrieval for the automatic discovery of query classes.  ... 
doi:10.9790/0661-16197881 fatcat:7txgqtlicnfcffteakolplwdji

Columbia University TRECVID-2005 Video Search and High-Level Feature Extraction

Shih-Fu Chang, Winston H. Hsu, Lyndon S. Kennedy, Lexing Xie, Akira Yanagawa, Eric Zavesky, Dong-Qing Zhang
2005 TREC Video Retrieval Evaluation  
Uses query-class-dependent weights on text retrieval, content-based image retrieval, text search against visual concepts, and Cue-X reranking M_A_1_ColumbiaM2_6: Required manual baseline run, using only  ...  The automatic run uses query-classdependent weights on text retrieval, content-base image retrieval, and text search against visual concepts .  ...  For query-class-dependent models to be effective, it is necessary to have a complementary set of tools, where particular tools (or combinations of tools) perform well for unique sets of query topics.  ... 
dblp:conf/trecvid/ChangHKXYZZ05 fatcat:k7hixb2dozeilgtobonf4okm3q

Query-Adaptive Fusion for Multimodal Search

L. Kennedy, Shih-Fu Chang, A. Natsev
2008 Proceedings of the IEEE  
We observe that the state of the art in query-adaptive retrieval frameworks for video collections is highly dependent upon the definition of classes of queries, which are groups of queries that share similar  ...  | We conduct a broad survey of query-adaptive search strategies in a variety of application domains, where the internal retrieval mechanisms used for search are adapted in response to the anticipated needs  ...  In a similar work, Yan and Hauptmann [3] also propose a framework for automatically determining class-dependent search models.  ... 
doi:10.1109/jproc.2008.916345 fatcat:izn34ieavjf3vmul55kjqq4bnu

Context-aware Service Discovery in Mobile Heterogeneous Environments

N. Blefari-Melazzi, E. Casalicchio, S. Salsano
2007 2007 16th IST Mobile and Wireless Communications Summit  
In this paper, we focus on the first requirement of SMS: how to find a mobile service matching the user needs and appropriate for the user context.  ...  The IST SMS (Simple Mobile Services) project takes up the challenge of creating innovative tools addressing the specific needs of mobile users and making it easier for individuals and small businesses  ...  The authors would like to acknowledge the contributions of their colleagues from the SMS consortium. The Community is not liable for any use that may be made of the information contained therein.  ... 
doi:10.1109/istmwc.2007.4299226 fatcat:zfmwcv2dhvejthpnldyle7lfyq

A reranking approach for context-based concept fusion in video indexing and retrieval

Lyndon S. Kennedy, Shih-Fu Chang
2007 Proceedings of the 6th ACM international conference on Image and video retrieval - CIVR '07  
We propose to incorporate hundreds of pre-trained concept detectors to provide contextual information for improving the performance of multimodal video search.  ...  We test the method on TRECVID 2005 and 2006 automatic video search tasks and find improvements in mean average precision (MAP) of 15%-30%.  ...  The views and conclusions are those of the authors, not of the US Government or its agencies.  ... 
doi:10.1145/1282280.1282331 dblp:conf/civr/KennedyC07 fatcat:7gwjpfya7reetji5ob5jfofffu

Multimodal Word Discovery and Retrieval with Phone Sequence and Image Concepts

Liming Wang, Mark A. Hasegawa-Johnson
2019 Interspeech 2019  
This paper demonstrates three different systems capable of performing the multimodal word discovery task.  ...  A multimodal word discovery system accepts, as input, a database of spoken descriptions of images (or a set of corresponding phone transcripts), and learns a lexicon which is a mapping from phone strings  ...  Acknowledgements The idea of this work originated from the 2017 JSALT workshop at Carnegie Mellon University, Pittsburgh. We would like to thank [17] for releasing the Flickr8k dataset.  ... 
doi:10.21437/interspeech.2019-1487 dblp:conf/interspeech/WangH19 fatcat:xk5x2ddt2fg7ppntj7lqj5oqie

A Survey on Multimodal Video Representation for Semantic Retrieval

J. Calic, N. Campbell, S. Dasiopoulou, Y. Kompatsiaris
2005 EUROCON 2005 - The International Conference on "Computer as a Tool"  
Furthermore, the concept of video multimodality is reevaluated and redefined in order to introduce modalities such as editing technique or affect to the audience.  ...  This paper surveys the approaches to video representation, focusing on semantic analysis for content-based indexing and retrieval.  ...  ACKNOWLEDGEMENTS The work reported in this paper has formed part of the activity of the WG3 within IST COST292 action in semantic multimodal analysis of digital media whose funding and support is gratefully  ... 
doi:10.1109/eurcon.2005.1629877 fatcat:e25kjkfqx5fplht6rrnavhwy54

IBM Research TRECVID-2006 Video Retrieval System

Murray Campbell, Alexander Haubold, Shahram Ebadollahi, Dhiraj Joshi, Milind R. Naphade, Apostol Natsev, Joachim Seidl, John R. Smith, Katya Scheinberg, Jelena Tesic, Lexing Xie
2006 TREC Video Retrieval Evaluation  
This year, focus of the system improvement was on ensemble learning and fusion for both high-level feature detection task and the search task.  ...  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 tried out three different multimodal fusion approaches-a query-independent non-weighted fusion approach, and two query-class-dependent fusion approaches using strict and fuzzy query class assignments  ... 
dblp:conf/trecvid/CampbellHEJNNSS06 fatcat:t2wpc5feqnedljhgcxy7lzx4hq

TRECVID 2006 by NUS-I2R

Tat-Seng Chua, Shi-Yong Neo, Yantao Zheng, Hai-Kiat Goh, Sheng Tang, Yang Xiao, Ming Zhao, Sheng Gao, Xinglei Zhu, Lekha Chaisorn, Qibin Sun
2006 TREC Video Retrieval Evaluation  
As for the automated search task, our emphases this year are: 1) integration of HLF: query-analysis to match query to possible related HLF and fuse results from various participating groups in the HLF  ...  For HLF task, we develop 2 methods to perform automated concept annotation: (a) fully machine learning approach using SVM, LDF and GMM; and (b) Bi-gram model for Pattern Discovery and Matching.  ...  FULLY AUTOMATIC SEARCH TASK This is our third year participating in the TRECVID automatic search task.  ... 
dblp:conf/trecvid/ChuaNZGTX0GZCS06 fatcat:45ttmjvwd5c2fjbqjiq7exeu2i

Flexible Fashion Product Retrieval Using Multimodality-Based Deep Learning

Yeonsik Jo, Jehyeon Wi, Minseok Kim, Jae Yeol Lee
2020 Applied Sciences  
This paper proposes a new fashion product search method using multimodality-based deep learning, which can support more flexible and efficient retrieval by combining faceted queries and fashion image-based  ...  Typically, fashion product searching in online shopping malls uses meta-information of the product.  ...  Conflicts of Interest: The authors declare no conflicts of interest.  ... 
doi:10.3390/app10051569 fatcat:2qeztv2rbzaljdm7yukgwgixqe

Interactive Search and Exploration in Online Discussion Forums Using Multimodal Embeddings [article]

Iva Gornishka, Stevan Rudinac, Marcel Worring
2019 arXiv   pre-print
We show these representations to be useful not only for categorizing users, but also for automatically generating user and community profiles.  ...  In this paper we present a novel interactive multimodal learning system, which facilitates search and exploration in large networks of social multimedia users.  ...  However, separately modeling the individual modalities does not only yield sparse representations in some of them, but it also fails to capture important dependencies between them.  ... 
arXiv:1905.02430v1 fatcat:4sy7tdnknjct5ksuqag5ygdoxq

Integrating, Indexing and Querying the Tangible and Intangible Cultural Heritage Available Online: The QueryLab Portal

Maria Teresa Artese, Isabella Gagliardi
2022 Information  
Moreover, with the introduction of the ICH-Light model, specifically studied for the mapping of intangible heritage, data from inventories of this domain can also be harvested, indexed and integrated into  ...  The purpose of the developed portal is to provide query tools managing the web services provided by cultural heritage databases in a transparent way, allowing the user to make a single query and obtain  ...  Multimodal Search Engine The QueryLab platform defines and implements a multimodal search engine, to increase user satisfaction in searching, browsing, viewing, and interacting with data from different  ... 
doi:10.3390/info13050260 fatcat:njgm2o3djngidbrkpvd7e3djhm

Knowledge Graphs and Knowledge Networks: The Story in Brief [article]

Amit Sheth, Swati Padhee, Amelie Gyrard
2020 arXiv   pre-print
The evolution of search engine responses to user queries in the last few years is partly because of the role of KGs such as Google KG.  ...  This article is an attempt to summarize the journey of KG for AI.  ...  The need for accurate temporal query responses by dominant search engines requires extracting, maintaining, and updating the temporal facts in KGs.  ... 
arXiv:2003.03623v1 fatcat:zle7g626mzeqrkmobgrr6xtfae
« Previous Showing results 1 — 15 out of 2,655 results