PMPGuard: Catching Pseudo-Matched Pairs in Remote Sensing Image-Text Retrieval

TL;DR

This paper introduces PMPGuard, a novel framework for remote sensing image-text retrieval that effectively handles pseudo-matched pairs by using cross-modal gated attention and positive-negative awareness mechanisms, improving robustness and accuracy.

Contribution

The paper proposes PMPGuard, a new retrieval framework with gating and awareness mechanisms to mitigate pseudo-matched pairs in remote sensing image-text retrieval.

Findings

Achieves state-of-the-art results on RSICD, RSITMD, and RS5M datasets.

Demonstrates robustness in handling noisy and mismatched data.

Outperforms existing methods in real-world scenarios.

Abstract

Remote sensing (RS) image-text retrieval faces significant challenges in real-world datasets due to the presence of Pseudo-Matched Pairs (PMPs), semantically mismatched or weakly aligned image-text pairs, which hinder the learning of reliable cross-modal alignments. To address this issue, we propose a novel retrieval framework that leverages Cross-Modal Gated Attention and a Positive-Negative Awareness Attention mechanism to mitigate the impact of such noisy associations. The gated module dynamically regulates cross-modal information flow, while the awareness mechanism explicitly distinguishes informative (positive) cues from misleading (negative) ones during alignment learning. Extensive experiments on three benchmark RS datasets, i.e., RSICD, RSITMD, and RS5M, demonstrate that our method consistently achieves state-of-the-art performance, highlighting its robustness and effectiveness in handling real-world mismatches and PMPs in RS image-text retrieval tasks.