Filters








17,410 Hits in 3.8 sec

Mobile multi-view object image search

Fatih Çalışır, Muhammet Baştan, Özgür Ulusoy, Uğur Güdükbay
2016 Multimedia tools and applications  
We performed a comprehensive analysis of early and late fusion approaches using various similarity functions, on an existing single view and a new multiview object image database.  ...  Multi-view images of objects acquired by the mobile clients are processed and local features are sent to a server, which combines the query image representations with early/late fusion methods and returns  ...  All these works use multiple query images on single view databases for performance improvement; they do not utilize multi-view queries on multi-view databases.  ... 
doi:10.1007/s11042-016-3659-9 fatcat:gd6jasw2rfaujpsm7hk5nhqg6u

Mobile Multi-View Object Image Search [article]

Fatih Calisir, Muhammet Bastan, Ozgur Ulusoy, Ugur Gudukbay
2018 arXiv   pre-print
We performed a comprehensive analysis of early and late fusion approaches using various similarity functions, on an existing single view and a new multi-view object image database.  ...  Multi-view images of objects acquired by the mobile clients are processed and local features are sent to the server, which combines the query image representations with early/late fusion methods based  ...  All these works use multiple query images on single view databases for performance improvement; they do not utilize multi-view queries on multi-view databases.  ... 
arXiv:1507.08861v2 fatcat:cqnzukmstnbf3heah2jaii75bi

Reifying and optimizing collection queries for modularity

Paolo G. Giarrusso
2012 Proceedings of the 3rd annual conference on Systems, programming, and applications: software for humanity - SPLASH '12  
We propose SQUOPT, the Scala Query Optimizer, a deep embedding of the Scala collection library performing collection-specific optimizations automatically without external tools or compiler extensions.  ...  Conventional collection libraries do not perform automatic collection-specific optimizations.  ...  In particular, our optimizer automatically performs various collection-specific algebraic optimizations, such as map fusion, selection pushdown, automatic indexing and query unnesting.  ... 
doi:10.1145/2384716.2384735 dblp:conf/oopsla/Giarrusso12 fatcat:vtm74dlhqnebppntjcf65ln3qm

Reifying and optimizing collection queries for modularity

Paolo G. Giarrusso, Klaus Ostermann, Michael Eichberg, Tillmann Rendel, Christian Kästner
2012 Proceedings of the 3rd annual conference on Systems, programming, and applications: software for humanity - SPLASH '12  
We propose SQUOPT, the Scala Query Optimizer, a deep embedding of the Scala collection library performing collection-specific optimizations automatically without external tools or compiler extensions.  ...  Conventional collection libraries do not perform automatic collection-specific optimizations.  ...  In particular, our optimizer automatically performs various collection-specific algebraic optimizations, such as map fusion, selection pushdown, automatic indexing and query unnesting.  ... 
doi:10.1145/2384716.2384747 dblp:conf/oopsla/GiarrussoOERK12 fatcat:tybqrtontzfbhd66tvol4uve4i

Automatic video annotation through search and mining

Emily Moxley, Tao Mei, Xian-Sheng Hua, Wei-Ying Ma, B.S. Manjunath
2008 2008 IEEE International Conference on Multimedia and Expo  
Then, the transcripts associated with these similar videos are mined to extract keywords for the query.  ...  This work aims to exploit the overlap in content of news video to automatically annotate by mining similar videos that reinforce, filter, and improve the original annotations.  ...  MAX fusion selects the best performing individual metric for a shot, and outperforms the automatic fusion modalities.  ... 
doi:10.1109/icme.2008.4607527 dblp:conf/icmcs/MoxleyMHMM08 fatcat:uaycmhpjlnac5lmwhwarhyc5pm

Analysing the performance of visual, concept and text features in content-based video retrieval

Mika Rautiainen, Timo Ojala, Tapio Sepp�nen
2004 Proceedings of the 6th ACM SIGMM international workshop on Multimedia information retrieval - MIR '04  
Weighted fusion of text, concept and visual features improved the performance over text search baseline.  ...  System performance was evaluated with 25 pre-defined search topics using average precision. In visual search, multiple examples improved the results over single example search.  ...  (left) Interactive interface for content-based browsing of video shots (right) Query formulation view that was used to formulate manual topic queries  ... 
doi:10.1145/1026711.1026744 dblp:conf/mir/RautiainenOS04 fatcat:bqae3m3ujbb4xjmjp2zdqk63oq

Waseda_Meisei at TRECVID 2017: Ad-hoc Video Search

Kazuya Ueki, Koji Hirakawa, Kotaro Kikuchi, Tetsuji Ogawa, Tetsunori Kobayashi
2017 TREC Video Retrieval Evaluation  
Our best fully automatic run achieved a mAP of 15.9%, which ranked second among all participants. 1 System Description Concept bank To provide good coverage for the given query phrases, we built a large  ...  For this year's AVS task, we submitted both manually assisted and fully automatic runs.  ...  Our future work will be focused on improving the performance of the fully automatic search system by effectively selecting visually valuable concepts.  ... 
dblp:conf/trecvid/UekiHKOK17 fatcat:5bzye36bvzeevgrhueap2hsa4q

Image retrieval based on colour and improved NMI texture features

Anyu Du, Liejun Wang, Jiwei Qin
2019 Automatika  
Therefore, the proposed algorithm has better retrieval performance and robustness compared with the existing image retrieval algorithms based on multi-feature fusion. ARTICLE HISTORY  ...  improved by 13.4% compared with the similar algorithm; on the FD-XJ dataset, the retrieval accuracy is improved by 17.7% compared with the similar algorithm.  ...  At this stage, the query image matches the image in the database; when the matching process is completed, an image with similar matching values is selected in the database and the image query.  ... 
doi:10.1080/00051144.2019.1645977 fatcat:a7hkfzi3yzhmlofrv4uupwoyee

Enhanced automatic 2D–3D conversion using retinex in machine learning framework

Jose L. Herrera, Carlos R. del-Blanco, Narciso Garcia
2015 2015 International Symposium on Consumer Electronics (ISCE)  
Our experimental results indicate that the inclusion of a retinex based stage for the query image and the dataset images improves the performance of the system on commonlyused databases and for different  ...  In this paper, we present an approach for automatically convert images from 2D to 3D.  ...  previous approach to efficiently use large databases through a hierarchical search, and also improved the performance of the algorithm by using an adaptive number of images in the fusion process of the  ... 
doi:10.1109/isce.2015.7177825 dblp:conf/isce/HerreradG15 fatcat:lhwy4hsrpjc3xhakaevuy3jkri

Late Semantic Fusion Approaches for Multimedia Information Retrieval with Automatic Tag Generation

Pooja Shinde, J.V. Shinde
2015 International Journal of Computer Applications  
Five late semantic fusion approaches are used for text based and content based image retrieval of any dataset.  ...  The logistic regression relevance feedback algorithm is used to determine the similarity between the images from the dataset to the query.  ...  They proposed semantic combination models for textual and visual information fusion can significantly improve multimedia retrieval performances while being more robust with respect to the choice of the  ... 
doi:10.5120/ijca2015906966 fatcat:qh64szjxd5bkheeiynex2mq3by

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 evaluate the system against the TRECVID 2004 automatic video search task and find that the automatically discovered query classes give an improvement of 18% in MAP over hand-defined query classes used  ...  We further combine these performance features with the semantic features of the queries during clustering in order to make discovered classes meaningful.  ...  It works particularly well when the images being sought in the database have consistent visual features, compared with each other and the query image, which are distinct from other images in the database  ... 
doi:10.1145/1101149.1101339 dblp:conf/mm/KennedyNC05 fatcat:g6lspob5vfemfl7zcewuywqsvi

Learning the semantics of multimedia queries and concepts from a small number of examples

Apostol (Paul) Natsev, Milind R. Naphade, Jelena TešiĆ
2005 Proceedings of the 13th annual ACM international conference on Multimedia - MULTIMEDIA '05  
One task involves answering queries with a few examples. The other involves learning models for semantic concepts, also with a few examples.  ...  In both tasks we find that applying the combination hypothesis across both modeling techniques and a variety of features results in enhanced performance over any of the baseline models, as well as in improved  ...  The performance improvement in the case of late model fusion is 7% and the improvement in the case of late model fusion is 17%.  ... 
doi:10.1145/1101149.1101288 dblp:conf/mm/NatsevNT05 fatcat:2fkovjvzd5azjh764mv4fzlfpe

Facial marks: Soft biometric for face recognition

Anil K. Jain, Unsang Park
2009 2009 16th IEEE International Conference on Image Processing (ICIP)  
., freckles, moles, and scars) to improve face recognition and retrieval performance.  ...  Facial marks can be used in three ways: i) to supplement the features in an existing face matcher, ii) to enable fast retrieval from a large database using facial mark based queries, and iii) to enable  ...  The fusion of facial marks with a state-of-the-art face matcher (FaceVACS) improves the rank-1 face recognition performance on a public domain as well as an operational database.  ... 
doi:10.1109/icip.2009.5413921 dblp:conf/icip/JainP09 fatcat:f6nkb63zgnap3ncqhqkctc2xru

Automatic ranking of information retrieval systems using data fusion

Rabia Nuray, Fazli Can
2006 Information Processing & Management  
This could improve the effectiveness of automatic ranking. Based on this intuition, we introduce a new method for the selection of systems to be used for data fusion.  ...  This approach provides even higher correlations with the humanbased results. We demonstrate that our approach outperforms the previously proposed automatic ranking methods.  ...  We thank two anonymous referees for their remarks which significantly improved the paper and NIST (http://www.nist.gov/) for making the TREC data available.  ... 
doi:10.1016/j.ipm.2005.03.023 fatcat:ggvpdzgegzb75etkq5j6fv6ruq

A Learning-Based Similarity Fusion and Filtering Approach for Biomedical Image Retrieval Using SVM Classification and Relevance Feedback

M. Rahman, S. Antani, G. Thoma
2011 IEEE Transactions on Information Technology in Biomedicine  
In this framework, the probabilistic outputs of a multiclass support vector machine (SVM) classifier as category prediction of query and database images are exploited at first to filter out irrelevant  ...  Experiment is performed in a diverse collection of 5 000 biomedical images of different modalities, body parts, and orientations.  ...  ACKNOWLEDGMENT The authors would like to thank the CLEF [28] , [29] organizers and the Radiological Society of North America (RSNA), for making the database available.  ... 
doi:10.1109/titb.2011.2151258 pmid:21690019 fatcat:j76tzyp62vgatkt6kkga6dqfxy
« Previous Showing results 1 — 15 out of 17,410 results