Query by dialog: an interactive approach to pictorial querying

We describe a prototype system, SQD (sketch-based query by dialog), designed for retrieving images by content according to shape similarity. At present, only shapes corresponding to closed contours have been considered. We represent them using normalized polar coordinates contour functions, from which image key features are extracted. The query system interface is based on the integration of the query-by-visual-example and the progressive-query approaches. A sketch-based query allows users to submit a drawn example of the target object. Whenever the system response does not match the searched object, users can continue the query process refining their request. The query specification and refinement cycle is iterated until the target object or a set of objects are obtained. The interface keeps track of the user's requests and of the system answers, so that it is always possible to fully control the query process and to analyze the system similarity evaluation.

[1]  Shi-Kuo Chang Pictorial Information Systems , 1981, Computer.

[2]  Patrick M. Kelly,et al.  CANDID: comparison algorithm for navigating digital image databases , 1994, Seventh International Working Conference on Scientific and Statistical Database Management.

[3]  Shi-Kuo Chang,et al.  Visual Languages , 1986, Management and Information Systems.

[4]  Antonin Guttman,et al.  R-trees: a dynamic index structure for spatial searching , 1984, SIGMOD '84.

[5]  Maria Francesca Costabile,et al.  Reality bites-progressive querying and result visualization in logical and VR spaces , 1994, Proceedings of 1994 IEEE Symposium on Visual Languages.

[6]  Alberto Del Bimbo,et al.  Visual image retrieval by elastic deformation of object sketches , 1994, Proceedings of 1994 IEEE Symposium on Visual Languages.

[7]  Robert R. Korfhage,et al.  Visualization of a Document Collection: The VIBE System , 1993, Inf. Process. Manag..

[8]  Alex Pentland,et al.  Photobook: tools for content-based manipulation of image databases , 1994, Electronic Imaging.

[9]  Jon Louis Bentley,et al.  Data Structures for Range Searching , 1979, CSUR.

[10]  Bruce Tognazzini,et al.  Tog on Interface , 1992 .

[11]  Hideyuki Tamura,et al.  Image database systems: A survey , 1984, Pattern Recognit..

[12]  Freddy Fierens,et al.  Interactive outlining: an improved approach using active contours , 1993, Electronic Imaging.

[13]  William I. Grosky,et al.  Index-based object recognition in pictorial data management , 1990, Comput. Vis. Graph. Image Process..

[14]  Rama Chellappa,et al.  Stochastic models for closed boundary analysis: Representation and reconstruction , 1981, IEEE Trans. Inf. Theory.

[15]  Filson H. Glanz,et al.  An Autoregressive Model Approach to Two-Dimensional Shape Classification , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  Paul Suetens,et al.  A software environment for image database research , 1992, J. Vis. Lang. Comput..

[17]  Toshikazu Kato,et al.  Query by Visual Example - Content based Image Retrieval , 1992, EDBT.

[18]  Shi-Kuo Chang,et al.  Image Information Systems: Where Do We Go From Here? , 1992, IEEE Trans. Knowl. Data Eng..

[19]  Ch Chen,et al.  Pattern recognition and artificial intelligence , 1976 .

[20]  S. Joy Mountford,et al.  The Art of Human-Computer Interface Design , 1990 .

[21]  Alan J. Dix Que Sera Sera - The Problem of the Future Perfect in Open and Cooperative Systems , 1994, BCS HCI.