Image Retrieval Over Networks: Active Learning Using Ant Algorithm
IEEE Transactions on MultimediaPublished 1 November 2008Open access
David Picard, Matthieu Cord, Arnaud Revel
Citations24
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
A framework for distributed content based image retrieval with online learning based on ant-like mobile agents, showing that learning both the similarity function and the localization of the relevant images leads to a significant improvement.
Abstract
International audience
Keywords
Computer Science
International Journal of Computer VisionDistinctive Image Features from Scale-Invariant Keypoints
55,266 Citations2004David Lowe
This paper presents a method for extracting distinctive invariant features from images that can be used to perform reliable matching between different views of an object or scene and can robustly identify objects among clutter and occlusion while achieving near real-time performance.
Machine LearningSupport-vector networks
40,463 Citations1995Corinna Cortes, Vladimir Vapnik
High generalization ability of support-vector networks utilizing polynomial input transformations is demonstrated and the performance of the support- vector network is compared to various classical learning algorithms that all took part in a benchmark study of Optical Character Recognition.
Machine LearningSupport-Vector Networks
33,035 Citations1995Corinna Cortes, Vladimir Vapnik
High generalization ability of support-vector networks utilizing polynomial input transformations is demonstrated and the performance of the support- vector network is compared to various classical learning algorithms that all took part in a benchmark study of Optical Character Recognition.
IEEE Transactions on Systems Man and Cybernetics Part B (Cybernetics)Ant system: optimization by a colony of cooperating agents
11,908 Citations1996Marco Dorigo, Vittorio Maniezzo +1 more
ACM Computing SurveysMachine learning in automated text categorization
7,899 Citations2002Fabrizio Sebastiani
This survey discusses the main approaches to text categorization that fall within the machine learning paradigm and discusses in detail issues pertaining to three different problems, namely, document representation, classifier construction, and classifier evaluation.
IEEE Transactions on Pattern Analysis and Machine IntelligenceContent-based image retrieval at the end of the early years
6,023 Citations2000A.W.M. Smeulders, Marcel Worring +3 more
The working conditions of content-based retrieval: patterns of use, types of pictures, the role of semantics, and the sensory gap are discussed, as well as aspects of system engineering: databases, system architecture, and evaluation.
Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE<title>QBIC project: querying images by content, using color, texture, and shape</title>
1,760 Citations1993Carlton W. Niblack, Ron Barber +7 more
The main algorithms for color texture, shape and sketch query that are presented, show example query results, and discuss future directions are presented.
Support vector machine active learning for image retrieval
1,333 Citations2001Simon Tong, Edward Yi Chang
This work proposes the use of a support vector machine active learning algorithm for conducting effective relevance feedback for image retrieval and achieves significantly higher search accuracy than traditional query refinement schemes after just three to four rounds of relevance feedback.
Elsevier eBooksHeterogeneous Uncertainty Sampling for Supervised Learning
1,142 Citations1994David Lewis, Jason Catlett
This work test the use of one classifier (a highly efficient probabilistic one) to select examples for training another (the C4.5 rule induction program) and finds that the uncertainty samples yielded classifiers with lower error rates than random samples ten times larger.
NatureInspiration for optimization from social insect behaviour
906 Citations2000Eric Bonabeau, Marco Dorigo +1 more
Research in social insect behaviour has provided computer scientists with powerful methods for designing distributed control and optimization algorithms that tend to exhibit a high degree of flexibility and robustness in a dynamic environment.
Applied Physics Letters10.1162/153244302760185243
898 Citations2000
Experimental results showing that employing the active learning method can significantly reduce the need for labeled training instances in both the standard inductive and transductive settings are presented.
The MIT Press eBooksThe Dynamics of Collective Sorting: Robot-Like Ants and Ant-Like Robots
883 Citations1991J. L. Deneubourg, Simon Goss +4 more
A distributed sorting algorithm, inspired by how ant colonies sort their brood, is presented for use by robot teams, offering the advantages of simplicity, flexibility and robustness.
ComputerContent based image retrieval systems
875 Citations1995Venkat N. Gudivada, Vijay V. Raghavan
Application areas in which CBIR is a principal activity are numerous and diverse and with the recent interest in multimedia systems, CBIR has attracted the attention of researchers across several disciplines.
Communications of the ACMSeven good reasons for mobile agents
872 Citations1999Danny B. Lange, Mitsuru Oshima
A mobile agent has the unique ability to transport itself from one system in a network to another in the same network, which allows it to move to a system containing an object with which it wants to interact and then to take advantage of being in the the same host or network as the object.
Lecture notes in computer scienceBlobworld: A System for Region-Based Image Indexing and Retrieval
761 Citations1999Chad Carson, Megan Thomas +3 more
This work indexes the blob descriptions using a lower-rank approximation to the high-dimensional distance to make large-scale retrieval feasible, and shows encouraging results for both querying and indexing.
Proceedings of the ninth ACM international conference on Multimedia - MULTIMEDIA '01Support vector machine active learning for image retrieval
486 Citations2001Simon Tong, Edward Yi Chang
Kluwer Academic Publishers eBooksDistributed Information Retrieval
387 Citations2005Jamie Callan
A broad and diverse group of experimental results is presented to demonstrate that the algorithms are effective, efficient, robust, and scalable.
Impact of search engines on page popularity
285 Citations2004Junghoo Cho, Sourashis Roy
This paper analytically estimates how much longer it takes for a new page to attract a large number of Web users when search engines return only popular pages at the top of search results and shows that search engines can have an immensely worrisome impact on the discovery of new Web pages.
SVMs for Histogram Based Image Classification
221 Citations1999Olivier Chapelle, Patrick Haffner +1 more
This paper shows that Support Vector Machines (SVM) can generalize well on difficult image classification problems where the only features are high dimensional histograms and observes that a simple remapping of the input xi → x a i improves the performance of linear SVMs to such an extend that it makes them a valid alternative to RBF kernels.
IEEE Transactions on Knowledge and Data EngineeringEmergent semantics through interaction in image databases
192 Citations2001Simone Santini, Amarnath Gupta +1 more
It is argued that images don't have an intrinsic meaning, but that they are endowed with a meaning by placing them in the context of other images and by the user interaction.
A Survey of Content-Based Image Retrieval Systems
169 Citations2002Remco C. Veltkamp, Mirela Tănase
In this chapter, some technical aspects of current content-based image retrieval systems are surveyed.
DSpace@MIT (Massachusetts Institute of Technology)Bayesian models for visual information retrieval
77 Citations2000Nuno Vasconcelos, Andrew Lippman
A Bayesian architecture is shown to generalize a significant number of previous recognition approaches, solving some of the most challenging problems faced by these: joint modeling of color and texture, objective guidelines for controlling the trade-off between feature transformation and feature representation, and unified support for local and global queries without requiring image segmentation.
Image retrieval with relevance feedback: from heuristic weight adjustment to optimal learning methods
71 Citations2002Tao Huang, Xiang Sean Zhou
The kernel-based biased discriminant analysis (KBDA) is proposed to fit the unique nature of relevance feedback as a biased classification problem and provides a trade-off between discriminant transform and regression.
ACM Transactions on Information SystemsDistributed content-based visual information retrieval system on peer-to-peer networks
68 Citations2004Irwin King, Cheuk Hang Ng +1 more
The design and implementation of DISCOVIR: DIStributed COntent-based Visual Information Retrieval system using the Peer-to-Peer (P2P) Network is presented and a Firework Query Model for distributed information retrieval is proposed.
Image and Vision ComputingStochastic exploration and active learning for image retrieval
47 Citations2006Matthieu Cord, Philippe H. Gosselin +1 more
A two-step—exploration, classification—interactive strategy designed for category retrieval using stochastic image selection process instead of the usual strategy based on a similarity score ranking.
CBIR in Distributed Databases using a Multi-Agent System
15 Citations2006David Picard, Matthieu Cord +1 more
A new architecture for image retrieval in distributed image databases, based on multi-agent, systems, inspired by "ant-agents", is introduced, which uses labels provided by the user for learning both the searched category of images and die path to the most relevant databases.
On-line learning by active sampling using orthogonal decision support vectors
13 Citations2002Jong-Min Park
The result of the active sampling near the boundary using orthogonal decision support vectors is shown in comparison with active learning using random selection in the multi-dimensional decision hyperplane.
A distributed content-based search engine based on mobile code
12 Citations2005Volker Röth, Ulrich Pinsdorf +1 more
This work presents the prototype of the secure and distributed search engine, which dynamically pushes content based feature extraction to image providers, and gives a description of the search engine's architecture and implementation, quantitative evaluation results, and a discussion of related security mechanism for content protection and server security.
Information RetrievalDistributed Information Retrieval
12 Citations2004David A. Grossman, Ophir Frieder
The use of a single machine with multiple processors to improve performance is discussed, and possible constraints prohibiting the centralization of the data include data security, their sheer volume prohibiting their physical transfer, their rate of change, political and legal constraints, as well as other proprietary motivations.
Web-agents inspired by ethology: a population of "ant"-like agents to help finding user-oriented information
12 Citations2004A. Revel
A Web-search ant-agent inspired by ethology and robotics is presented and its implementation on a set of FIPAOS platforms is detailed and useful results in route finding and rerouting are shown.
From Robots to Web-Agents: Building Cognitive Software Agents for Web-Information Retrieval by Taking Inspiration from Experience in Robotics
11 Citations2005Arnaud Revel
Web-search mobile agents inspired by ethology and robotics and the implementation on a set of FI-PAOS platforms is given and unintuitive results concerning agent mobility are given.
Journal of Interconnection NetworksPerformance Analysis of Mobile Agents in Mobile Distributed Information Retrieval System – A Quantitative Case Study
5 Citations2004Yu Jiao, Ali R. Hurson
A quantitative performance evaluation of systems implemented using the client/server and agent-based models in terms of response time, energy consumption, and communication cost is presented.