Abstract:Support vector machine (SVM) active learning is one popular and successful technique for relevance feedback in content-based image retrieval (CBIR). Despite the success, conventional SVM active learning has two main drawbacks. First, the performance of SVM is usually limited by the number of labeled examples. It often suffers a poor performance for the small-sized labeled examples, which is the case in relevance feedback. Second, conventional approaches do not take into account the redundancy among examples, and could select multiple examples that are similar (or even identical). In this work, we propose a novel scheme for explicitly addressing the drawbacks. It first learns a kernel function from a mixture of labeled and unlabeled data, and therefore alleviates the problem of small-sized training data. The kernel will then be used for a batch mode active learning method to identify the most informative and diverse examples via a min-max framework. Two novel algorithms are proposed to solve the related combinatorial optimization: the first approach approximates the problem into a quadratic program, and the second solves the combinatorial optimization approximately by a greedy algorithm that exploits the merits of submodular functions. Extensive experiments with image retrieval using both natural photo images and medical images show that the proposed algorithms are significantly more effective than the state-of-the-art approaches. A demo is available at http://msm.cais.ntu.edu.sg/LSCBIR/.

Active BSVM Learning for Relevance Feedback in Content-Based Sketch Retrieval

BSVM-Based Relevance Feedback for Sketch Retrieval

Sketch Retrieval and Relevance Feedback with Biased SVM Classification.

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

Semi-supervised SVM Batch Mode Active Learning for Image Retrieval

Semisupervised SVM Batch Mode Active Learning with Applications to Image Retrieval

Small Sample Incremental Biased Learning Algorithm for Sketch Retrieval

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

An improved method for support vector machine-based active feedback

Probabilistic Latent Semantic Analysis for Sketch-Based 3D Model Retrieval

Active Learning for Human Action Retrieval Using Query Pool Selection

Image Retrieval Method Based on SVMs and Active Learning

Boost Svm Active Learning For Content-Based Image Retrieval

Relevance feedback for sketch retrieval based on linear programming classification

Relevance Feedback in Content-Based Image Retrieval: Bayesian Framework, Feature Subspaces, and Progressive Learning

Support Vector Machine Learning for Image Retrieval

Applying Relevance Feedback with a Linear Programming Classifier

A Novel SVM Based Relevance Feedback Algorithm in Image Retrieval

A Boosting Method Based on SVM for Relevance Feedback in Content-Based 3D Model Retrieval

Relevance Feedback For Keyword And Visual Feature-Based Image Retrieval

Support Vector Machine Based Relevance Feedback Algorithm in Image Retrieval