Ashwin T
ABSTRACT
Inherent subjectivity in user's perception of an image has motivated the use of relevance feedback (RF) in the image desigined output's retrieval process. RF techniques interactively determine the user's query concept, given the user's relevance judgments on a set of images. In this paper we propose a robust technique that utilizes non-relevant images to efficiently discover the relevant search region. A similarity metric, estimated using the relevant images is then used to rank and retrieve database images in the relevant region. The partitioning of the feature space is achieved by using a piecewise linear decision surface that separates the relevant and non-relevant images. Each of the hyperplanes constituting the decision surface is normal to the minimum distance vector from a non-relevant point to the convex hull of relevant points. Experimental results demonstrate significant improvement in retrieval performance for the small feedback size scenario over two well established RF algorithms.
- Y. Rui, T. S. Huang, and S. Mehrotra. Content-based image retrieval with relevance feedback in MARS. Proc. Int'l Conf. on Image Processing, 1997.Google Scholar
Cross Ref
- X. S. Zhou and T. S. Huang. Exploring the nature and variants of relevance feedback. Proc. IEEE Workshop on Content-Based Access of Image and Video Libraries, Hawaii, 2001. Google ScholarDigital Library
- Y. Ishikawa, R. Subramanya, and C. Faloustos. MindReader: Query databases through multiple examples. Proc. VLDB Conf., New York, 1998. Google ScholarDigital Library
- X. S. Zhou and T. S. Huang. Small sample learning during multimedia retrieval using BiasMap. Proc. IEEE Conf. Computer Vision and Pattern Recognition, Hawaii, 2001. Google ScholarDigital Library
- S. Tong and E. Chang. Support vector machine active learning for image retrieval. Proc. ACM Multimedia, Ottawa, 2001. Google ScholarDigital Library
- Ashwin T. V., Rahul Gupta, and Sugata Ghosal. Adaptable similarity search using non-relevant information. Proc. VLDB Conf., Hong Kong, 2002. Google ScholarDigital Library
- David G. Luenberger. Introduction to linear and non-linear programming, pp. 262--270. Addison-Wesley Publishing Company, 1973.Google Scholar
- Ashwin T. V., Rahul Gupta, and Sugata Ghosal. Leveraging non-relevant images to enhance image retrieval performance. IBM Research Report, RI02022, http://www.research.ibm.com/resources/paper_search.shtml, 2002.Google Scholar
- T. Joachims. SV Mlight, http://svmlight.joachims.org/.Google Scholar
Recommendations
Adaptable similarity search using non-relevant information
VLDB '02: Proceedings of the 28th international conference on Very Large Data BasesMany modern database applications require content-based similarity search capability in numeric attribute space. Further, users' notion of similarity varies between search sessions. Therefore online techniques for adaptively refining the similarity ...
A statistical correlation model for image retrieval
MULTIMEDIA '01: Proceedings of the 2001 ACM workshops on Multimedia: multimedia information retrievalA bigram correlation model for image retrieval is proposed, which captures the semantic relationship among images in a database from simple statistics of users' relevance feedback information. It is used in the post-processing of image retrieval results ...
Relevance feedback methods for logo and trademark image retrieval on the web
SAC '06: Proceedings of the 2006 ACM symposium on Applied computingRelevance feedback is the state-of-the-art approach for adjusting query results to the needs of the users. This work extends the existing framework of image retrieval with relevance feedback on the Web by incorporating text and image content into the ...
Comments