Abstract:Content-based image retrieval (CBIR) has attracted much attention during the past decades for its potential practical applications to image database management. A variety of relevance feedback (RF) schemes have been designed to bridge the gap between low-level visual features and high-level semantic concepts for an image retrieval task. In the process of RF, it would be impractical or too expensive to provide explicit class label information for each image. Instead, similar or dissimilar pairwise constraints between two images can be acquired more easily. However, most of the conventional RF approaches can only deal with training images with explicit class label information. In this paper, we propose a novel discriminative semantic subspace analysis (DSSA) method, which can directly learn a semantic subspace from similar and dissimilar pairwise constraints without using any explicit class label information. In particular, DSSA can effectively integrate the local geometry of labeled similar images, the discriminative information between labeled similar and dissimilar images, and the local geometry of labeled and unlabeled images together to learn a reliable subspace. Compared with the popular distance metric analysis approaches, our method can also learn a distance metric but perform more effectively when dealing with high-dimensional images. Extensive experiments on both synthetic datasets and a real-world image database demonstrate the effectiveness of the proposed scheme in improving the performance of CBIR.

Incremental semi-supervised subspace learning for image retrieval.

Image retrieval based on incremental subspace learning

Learning A Semantic Space from User'S Relevance Feedback for Image Retrieval

Learning an Image Manifold for Retrieval

Active subspace learning

A semantic subspace learning method to exploit relevance feedback log data for image retrieval

Image Retrieval Based on Semi-Supervised Orthogonal Discriminant Embedding

Spectral regression: a unified subspace learning framework for content-based image retrieval.

Learning a Maximum Margin Subspace for Image Retrieval

Iterative Locality Preserving Projection for Image Retrieval

Semisupervised SVM Batch Mode Active Learning with Applications to Image Retrieval

Image Retrieval Algorithm Based on Enhanced Relational Graph.

Semi-supervised Learning for Image Retrieval Using Support Vector Machines

Asymmetric Bagging and Random Subspace for Support Vector Machines-Based Relevance Feedback in Image Retrieval

Semi-supervision and Active Relevance Feedback Algorithm for Content-Based Image Retrieval

A Linear Subspace Learning Algorithm For Incremental Data

Similarity Measure Learning for Image Retrieval Using Feature Subspace Analysis

Subspace Clustering and Label Propagation for Active Feedback in Image Retrieval

Semantic Hashing with Image Subspace Learning

Discriminative Semantic Subspace Analysis for Relevance Feedback.

Semi-supervised Dimensionality Reduction for Image Retrieval