Asymmetric Semi-Supervised Boosting Scheme for Interactive Image Retrieval

Abstract

Support vector machine (SVM) active learning plays a key role in interactive content-based image retrieval (CBIR) community. However, regular SVM active learning is challenged by what we call the small example problem and the asymmetric distribution problem. This paper attempts to integrate merits of semi-supervised learning, ensemble learning, and active learning into interactive CBIR. Concretely, unlabeled images are exploited to facilitate boosting by helping augment diversity among base SVM classifiers, and then learned ensemble model is used to identify most informative images for active learning. In particular, a bias-weighting mechanism is developed to guide ensemble model to pay more attention on positive images than negative images. Experiments on 5000 Corel images show that proposed method yields better retrieval performance by an amount of 0.16 in mean average precision compared to regular SVM active learning, which is more effective than some existing improved variants of SVM active learning.