A kernel-based active learning strategy for content-based image retrieval

Active learning methods have attracted many researchers in the content-based image retrieval (CBIR) community. In this paper, we propose an efficient kernel-based active learning strategy to improve the retrieval performance of CBIR systems using class probability distributions. The proposed method learns for each class a nonlinear kernel which transforms the original feature space into a more effective one. The distances between user's request and database images are then learned and computed in the kernel space. Experimental results show that the proposed kernel-based active learning approach not only improves the retrieval performances of kernel distance without learning, but also outperforms other kernel metric learning methods.