The performance of image retrieval with SVM active learning is known to be poor when started with few labeled images only. In this paper, the problem is solved by incorporating the unlabelled images into the bootstrapping of the learning process. In this work, the initial SVM classifier is trained with the few labeled images and the unlabelled images randomly selected from the image database. Both theoretical analysis and experimental results show that by incorporating unlabelled images in the bootstrapping, the efficiency of SVM active learning can be improved, and thus improves the overall retrieval performance.
Bootstrapping SVM active learning by incorporating unlabelled images for image retrieval
01.01.2003
2797509 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Bootstrapping SVM Active Learning by Incorporating Unlabelled Images for Image Retrieval
British Library Conference Proceedings | 2003
|Learning with constrained and unlabelled data
IEEE | 2005
|New Unlabelled Process Oils for the Tyre and Rubber Industry
British Library Conference Proceedings | 1999
|Stochastic exploration and active learning for image retrieval
British Library Online Contents | 2007
|