Incorporating prior knowledge into SVM for image retrieval
- 1 January 2004
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 2, 981-984 Vol.2
- https://doi.org/10.1109/icpr.2004.1334423
Abstract
SVM based image retrieval suffers from the scarcity of labelled samples. In this paper, this problem is solved by incorporating prior knowledge into SVM. Firstly, some prior knowledge of image retrieval is discussed and constructed. After that, the knowledge is incorporated into SVM optimization as a constraint, and a new knowledge-based target function is formulated. Based on this, a framework of image retrieval with knowledge based SVM is proposed. Experimental results demonstrate that the proposed method can effectively improve the learning and retrieval performance of SVM, especially when the number of labelled samples is small.Keywords
This publication has 5 references indexed in Scilit:
- Image retrieval with SVM active learning embedding Euclidean searchPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2004
- Update relevant image weights for content-based image retrieval using support vector machinesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Learning similarity measure for natural image retrieval with relevance feedbackIEEE Transactions on Neural Networks, 2002
- Support vector machine active learning for image retrievalPublished by Association for Computing Machinery (ACM) ,2001
- Similarity of color imagesPublished by SPIE-Intl Soc Optical Eng ,1995