Information-Theoretic Active Learning for Content-Based Image Retrieval

TL;DR

This paper introduces ITAL, an active learning method that maximizes mutual information to improve content-based image retrieval, effectively handling user feedback and model updates for better retrieval accuracy.

Contribution

It presents a novel mutual information-based batch active learning approach for image retrieval, incorporating user behavior modeling and data structure considerations.

Findings

Achieves state-of-the-art performance on MIRFLICKR and ImageNet datasets.

Effectively models user feedback, including incorrect labels and unnameable instances.

Demonstrates flexibility and robustness across various datasets.

Abstract

We propose Information-Theoretic Active Learning (ITAL), a novel batch-mode active learning method for binary classification, and apply it for acquiring meaningful user feedback in the context of content-based image retrieval. Instead of combining different heuristics such as uncertainty, diversity, or density, our method is based on maximizing the mutual information between the predicted relevance of the images and the expected user feedback regarding the selected batch. We propose suitable approximations to this computationally demanding problem and also integrate an explicit model of user behavior that accounts for possible incorrect labels and unnameable instances. Furthermore, our approach does not only take the structure of the data but also the expected model output change caused by the user feedback into account. In contrast to other methods, ITAL turns out to be highly flexible and provides state-of-the-art performance across various datasets, such as MIRFLICKR and ImageNet.