OptiSAR-Net++: A Large-Scale Benchmark and Transformer-Free Framework for Cross-Domain Remote Sensing Visual Grounding

TL;DR

This paper introduces OptiSAR-Net++, a novel framework for cross-domain remote sensing visual grounding that leverages a new large-scale benchmark dataset, employing efficient feature decoupling and cross-modal matching techniques to improve accuracy and computational efficiency.

Contribution

The paper presents OptiSAR-RSVG, the first large-scale cross-domain remote sensing visual grounding dataset, and proposes OptiSAR-Net++, a transformer-free framework with innovative modules for enhanced semantic and spatial modeling.

Findings

Achieves state-of-the-art performance on OptSAR-RSVG and DIOR-RSVG benchmarks.

Demonstrates improved localization accuracy and efficiency over existing methods.

Effectively handles cross-domain feature modeling and fine-grained semantic discrimination.

Abstract

Remote sensing visual grounding (RSVG) aims to localize specific targets in remote sensing images using natural language expressions. However, existing methods are restricted to single-sensor domains, i.e., either optical or synthetic aperture radar (SAR), limiting their real-world applicability. In this paper, we introduce the Cross-Domain RSVG (CD-RSVG) task and construct OptSAR-RSVG, the first large-scale benchmark dataset for this setting. To tackle the challenges of cross-domain feature modeling, computational inefficiency, and fine-grained semantic discrimination, we propose OptiSAR-Net++. Our framework features a patch-level Low-Rank Adaptation Mixture of Experts (PL-MoE) for efficient cross-domain feature decoupling. To mitigate the substantial computational overhead of Transformer decoding frameworks, we adopt a CLIP-based contrastive paradigm and further incorporate dynamic adversarial negative sampling, thereby transforming generative regression into an efficient cross-modal matching process. Additionally, a text-guided dual-gate fusion module (TGDF-SSA) and a region-aware auxiliary head are introduced to enhance semantic-visual alignment and spatial modeling. Extensive experiments demonstrate that OptiSAR-Net++ achieves SOTA performance on both OptSAR-RSVG and DIOR-RSVG benchmarks, offering significant advantages in localization accuracy and efficiency. Our code and dataset will be made publicly available.