A Novel Semi-Supervised Learning for Collaborative Image Retrieval

Content-based image retrieval (CBIR) solutions with regular Euclidean metric usually cannot achieve satisfactory performance due to the semantic gap. Hence, relevance feedback has been adopted as a promising approach to improve the search performance. In this paper, we propose a novel idea of learning with historical relevance feedback log-data, and adopt a new methodology called "Collaborative Image Retrieval" (CIR). To effectively search the log data,we propose a novel semi-supervised distance metric learning technique, called "Laplacian Regularized Metric Learning" (LRML), for learning robust distance metrics for CIR. Different from previous methods, the proposed LRML method integrates both log data and unlabeled data information through an effective graph regularization framework. We show that reliable metrics can be learned from real log data even they may be noisy and limited at the beginning stage of a CIR system.