Fast Interactive Image Retrieval using large-scale unlabeled data

TL;DR

This paper introduces a scalable, interactive image retrieval system that combines active learning and graph-based semi-supervised learning to efficiently identify relevant images with minimal user feedback in large datasets.

Contribution

It proposes a novel scalable GSSL approach integrated with active learning and multi-modal data to improve interactive image retrieval performance.

Findings

High F1 scores achieved with few feedback rounds

Scalable GSSL reduces complexity from O(n^3) to O(n)

Effective on large datasets like Imagenet with 1.2 million images

Abstract

An interactive image retrieval system learns which images in the database belong to a user's query concept, by analyzing the example images and feedback provided by the user. The challenge is to retrieve the relevant images with minimal user interaction. In this work, we propose to solve this problem by posing it as a binary classification task of classifying all images in the database as being relevant or irrelevant to the user's query concept. Our method combines active learning with graph-based semi-supervised learning (GSSL) to tackle this problem. Active learning reduces the number of user interactions by querying the labels of the most informative points and GSSL allows to use abundant unlabeled data along with the limited labeled data provided by the user. To efficiently find the most informative point, we use an uncertainty sampling based method that queries the label of the point nearest to the decision boundary of the classifier. We estimate this decision boundary using our heuristic of adaptive threshold. To utilize huge volumes of unlabeled data we use an efficient approximation based method that reduces the complexity of GSSL from $O(n^3)$ to $O(n)$, making GSSL scalable. We make the classifier robust to the diversity and noisy labels associated with images in large databases by incorporating information from multiple modalities such as visual information extracted from deep learning based models and semantic information extracted from the WordNet. High F1 scores within few relevance feedback rounds in our experiments with concepts defined on AnimalWithAttributes and Imagenet (1.2 million images) datasets indicate the effectiveness and scalability of our approach.