EGO: a personalized multimedia management and retrieval tool. (English) Zbl 1104.68460

Summary: The problems of Content-Based Image Retrieval (CBIR) systems can be attributed to the semantic gap between the low-level data representation and the high-level concepts the user associates with images, on the one hand, and the time-varying and often vague nature of the underlying information need, on the other. These problems can be addressed by improving the interaction between the user and the system. In this article, we sketch the development of CBIR interfaces and introduce our view on how to solve some of the problems these interfaces present. To address the semantic gap and long-term multifaceted information needs, we propose a ’retrieval in context’ system, EGO. EGO is a tool for the management of image collections, supporting the user through personalization and adaptation. We will describe how it learns from the user’s personal organization, allowing it to recommend relevant images to the user. The recommendation algorithm is described, which is based on relevance feedback techniques. Additionally, we provide results of a performance analysis of the recommendation system and of a preliminary user study.


68P20 Information storage and retrieval of data
68M10 Network design and communication in computer systems
68W40 Analysis of algorithms


ImageGrouper; EGO
Full Text: DOI Link


[1] Bates, Online Review 13 pp 407– (1989)
[2] Information retrieval interaction. London: Taylor Graham; 1992.
[3] Affective computing. Cambridge, MA: MIT Press; 1997.
[4] Interface techniques for making searching for information more effective. In: CHI’2003, Best Practices and Future Visions for Search UIs: A Workshop; 2003.
[5] Smeulders, IEEE Trans Patt Anal Mach Intell 22 pp 1349– (2000)
[6] , . Automatic image annotation and retrieval using cross-media relevance models. In: Proc Ann Int ACM SIGIR Conf on Research and Development in Information Retrieval (SIGIR’03); 2003.
[7] Zhou, IEEE Multimedia 9 pp 23– (2002)
[8] Santini, IEEE Trans Knowl Data Eng 13 pp 337– (2001)
[9] ter Hofstede, Comput J 39 pp 255– (1996)
[10] Rui, IEEE Trans Circ Syst Video Technol 8 pp 644– (1998)
[11] . Optimizing learning in image retrieval. In: IEEE Proc Conf on Computer Vision and Pattern Recognition (CVPR-00), June 2000, Los Alamitos: IEEE Computer Society Press; 2000. pp 236–245.
[12] , . Query refinement for multimedia similarity retrieval in MARS. In: Proc ACM Int Conf on Multimedia, Orlando, Florida; 1999. pp 235–238.
[13] , . ImageGrouper: Search, annotate and organize images by groups. In: Proc Fifth Int Conf on Visual Information Systems (VISual 2002). Lecture Notes in Computer Science, Vol. 2314. Berlin: Springer Verlag; 2002. pp 129–142. · Zbl 1051.68754
[14] Zhou, ACM Multimedia Syst J 8 pp 536– (2003)
[15] , . An adaptive approach towards content-based image retrieval. In: Proc Third Int Workshop on Content-Based Multimedia Indexing (CBMI’03), Rennes, France, September 2003. pp 119–126.
[16] , . An approach for implicitly detecting information needs. In: Proc Int Conf on Information and Knowledge Management (CIKM 2003), November 2003. pp 504–507.
[17] . The ostensive model of developing information needs. In: Proc Int Conf on Conceptions of Library and Information Science; 1996. pp 251–268.
[18] . The art of search: A study of art directors. In: Proc ACM Int Conf on Human Factors in Computing Systems (CHI’92); 1992. pp 157–163.
[19] Markkula, Inform Retriev 1 pp 259– (2000)
[20] Santini, IEEE Trans Multimedia 7 pp 26– (2000) · Zbl 05099380
[21] Chen, IEEE Trans Image Process 9 pp 442– (2000)
[22] , , . Does organisation by similarity assist image browsing? In: Proc ACM Int Conf on Human Factors in Computing Systems; 2001. pp 190–197.
[23] Flickner, Computer 28 pp 23– (1995)
[24] . Fetch: A personalised information retrieval tool. In: Proc Conf Recherche d’Information AssistĂ©e par Ordinateur (RIAO’2004); 2004. pp 405–419.
[25] . Mediafinder: An interface for dynamic personal media management with semantic regions. In: Proc ACM Int Conf on Human Factors in Computing Systems (CHI’2003); 2003. pp 764–765.
[26] Explicit query formulation with visual keywords. In: Proc Eighth Int ACM Conf on Multimedia (Multimedia-00), October 30–November 4, 2000. New York: ACM Press; 2001. pp 407–409.
[27] Miller, Psych Rev 63 pp 81– (1956)
[28] . Similarity of color images. In: Proc SPIE: Storage and Retrieval for Image and Video Databases, February 1995. pp 381–392.
[29] , . Image processing, analysis, and machine vision, 2nd ed. Toronto: Thomson Engineering; 1998.
[30] Hu, IEEE Trans Inform Theor 8 pp 179– (1962)
[31] . Pattern recognition. San Diego, CA: Academic Press; 1999. · Zbl 0968.68540
[32] . Determining the number of clusters/segments in hierarchical clustering/segmentation algorithms. Technical Report CS-2003-18. Melbourne, FL: Florida Institute of Technology; 2003.
[33] , . Efficient similarity search and classification via rank aggregation. In: Proc ACM SIGMOD Int Conf on Management of Data. New York: ACM Press; 2003. pp 301–312.
[34] . Evidence combination for multi-point query learning in content-based image retrieval. In: Proc IEEE Sixth Int Symp on Multimedia Software Engineering (ISMSE’04), December 2004. pp 583–586.
[35] . A retrieval mechanism for semi-structured photographic collections. In: Hameurlain A, Tjoa A, editors. Proc Int Conf on Database and Expert Systems Applications, Toulouse, France, September 1997. pp 276–292.
[36] , . Spatial querying for image retrieval: A user-oriented evaluation. In: Proc Ann Int ACM SIGIR Conf on Research and Development in Information Retrieval (SIGIR’98). New York: ACM Press; 1998. pp 232–240.
[37] Paired-sample testing by ranks. In: editor. Biostatistical analysis. Englewood Cliffs, NJ: Prentice-Hall; 1984. pp 153–156.
[38] , . Towards personalized retrieval. In: Proc Second Int Workshop on Adaptive Multimedia Retrieval; 2004. pp 89–102.
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.