Weighted Contrastive Hashing

TL;DR

This paper introduces Weighted Contrastive Hashing (WCH), a novel unsupervised hashing method that improves image retrieval by leveraging fine-grained semantic relations and a mutual attention module to address limitations of previous contrastive learning approaches.

Contribution

The paper proposes a new WCH method with a mutual attention module and patch-based similarity aggregation to enhance unsupervised hashing performance.

Findings

WCH outperforms existing methods on benchmark datasets.

The mutual attention module improves feature information utilization.

Patch-based similarity enhances semantic relation modeling.

Abstract

The development of unsupervised hashing is advanced by the recent popular contrastive learning paradigm. However, previous contrastive learning-based works have been hampered by (1) insufficient data similarity mining based on global-only image representations, and (2) the hash code semantic loss caused by the data augmentation. In this paper, we propose a novel method, namely Weighted Contrative Hashing (WCH), to take a step towards solving these two problems. We introduce a novel mutual attention module to alleviate the problem of information asymmetry in network features caused by the missing image structure during contrative augmentation. Furthermore, we explore the fine-grained semantic relations between images, i.e., we divide the images into multiple patches and calculate similarities between patches. The aggregated weighted similarities, which reflect the deep image relations, are distilled to facilitate the hash codes learning with a distillation loss, so as to obtain better retrieval performance. Extensive experiments show that the proposed WCH significantly outperforms existing unsupervised hashing methods on three benchmark datasets.