DEMO: A Statistical Perspective for Efficient Image-Text Matching

TL;DR

This paper introduces DEMO, a novel unsupervised hashing method for image-text matching that uses distribution divergence and consistency learning to improve semantic structure accuracy and retrieval performance.

Contribution

DEMO employs a statistical approach with distribution divergence and collaborative consistency learning to enhance image-text matching accuracy.

Findings

Outperforms state-of-the-art methods on benchmark datasets

Robust semantic similarity structure construction

Effective self-supervised consistency enforcement

Abstract

Image-text matching has been a long-standing problem, which seeks to connect vision and language through semantic understanding. Due to the capability to manage large-scale raw data, unsupervised hashing-based approaches have gained prominence recently. They typically construct a semantic similarity structure using the natural distance, which subsequently provides guidance to the model optimization process. However, the similarity structure could be biased at the boundaries of semantic distributions, causing error accumulation during sequential optimization. To tackle this, we introduce a novel hashing approach termed Distribution-based Structure Mining with Consistency Learning (DEMO) for efficient image-text matching. From a statistical view, DEMO characterizes each image using multiple augmented views, which are considered as samples drawn from its intrinsic semantic distribution. Then, we employ a non-parametric distribution divergence to ensure a robust and precise similarity structure. In addition, we introduce collaborative consistency learning which not only preserves the similarity structure in the Hamming space but also encourages consistency between retrieval distribution from different directions in a self-supervised manner. Through extensive experiments on three benchmark image-text matching datasets, we demonstrate that DEMO achieves superior performance compared with many state-of-the-art methods.