A novel image annotation based on one-class SVM

In this paper, a novel AIA scheme based on one-class SVM is proposed. Compared with the previous SVM-based annotation method, one-class SVM is triggered out for remarkably boosting the reliability of SVM with less users' labeling effort involved. This guarantees that the proposed one-class SVM based annotation scheme integrates the discriminative classification with the generative model to mutually complete their advantages. In addition, not only the relevance model between the visual content of images and the textual keywords but also the property of keyword correlation is exploited in the proposed AIA scheme. Particularly, to establish an enhanced correlation network among keywords, both co-occurrence based and WordNet based correlation techniques are well fused and are able to be helpful for benefiting from each other. The experimental results reveal that the better annotation performance can be achieved at less labeled training images.