ProVG: Progressive Visual Grounding via Language Decoupling for Remote Sensing Imagery

TL;DR

ProVG introduces a progressive, decoupled language grounding framework for remote sensing imagery that enhances localization accuracy by leveraging fine-grained linguistic cues through a coarse-to-fine alignment process.

Contribution

It proposes a novel decoupling of linguistic cues and a progressive cross-modal modulation scheme tailored for remote sensing visual grounding tasks.

Findings

ProVG achieves state-of-the-art results on RRSIS-D and RISBench benchmarks.

The framework effectively utilizes spatial relations and object attributes for improved localization.

Extensive experiments validate the superiority of the proposed method over existing approaches.

Abstract

Remote sensing visual grounding (RSVG) aims to localize objects in remote sensing imagery according to natural language expressions. Previous methods typically rely on sentence-level vision-language alignment, which struggles to exploit fine-grained linguistic cues, such as \textit{spatial relations} and \textit{object attributes}, that are crucial for distinguishing objects with similar characteristics. Importantly, these cues play distinct roles across different grounding stages and should be leveraged accordingly to provide more explicit guidance. In this work, we propose \textbf{ProVG}, a novel RSVG framework that improves localization accuracy by decoupling language expressions into global context, spatial relations, and object attributes. To integrate these linguistic cues, ProVG employs a simple yet effective progressive cross-modal modulator, which dynamically modulates visual attention through a \textit{survey-locate-verify} scheme, enabling coarse-to-fine vision-language alignment. In addition, ProVG incorporates a cross-scale fusion module to mitigate the large-scale variations in remote sensing imagery, along with a language-guided calibration decoder to refine cross-modal alignment during prediction. A unified multi-task head further enables ProVG to support both referring expression comprehension and segmentation tasks. Extensive experiments on two benchmarks, \textit{i.e.}, RRSIS-D and RISBench, demonstrate that ProVG consistently outperforms existing methods, achieving new state-of-the-art performance.