Finding More Relevance: Propagating Similarity on Markov Random Field for Image Retrieval

TL;DR

This paper introduces a Markov random field-based method that propagates visual similarities to improve image retrieval accuracy by capturing both low-level features and relational context among images.

Contribution

It presents a novel similarity propagation technique on MRF that enhances image correspondence and can be integrated into existing BoW systems for better retrieval performance.

Findings

Significantly improves retrieval accuracy on Oxford-5K, Oxford-105K, and Paris-6K datasets.

Outperforms current state-of-the-art methods in image retrieval.

Effectively incorporates relational information among images to refine similarity measures.

Abstract

To effectively retrieve objects from large corpus with high accuracy is a challenge task. In this paper, we propose a method that propagates visual feature level similarities on a Markov random field (MRF) to obtain a high level correspondence in image space for image pairs. The proposed correspondence between image pair reflects not only the similarity of low-level visual features but also the relations built through other images in the database and it can be easily integrated into the existing bag-of-visual-words(BoW) based systems to reduce the missing rate. We evaluate our method on the standard Oxford-5K, Oxford-105K and Paris-6K dataset. The experiment results show that the proposed method significantly improves the retrieval accuracy on three datasets and exceeds the current state-of-the-art retrieval performance.