FUSE-RSVLM: Feature Fusion Vision-Language Model for Remote Sensing

TL;DR

This paper introduces MF-RSVLM, a novel multi-feature fusion vision-language model tailored for remote sensing, which enhances visual feature extraction and reduces forgetting, achieving state-of-the-art results across multiple RS tasks.

Contribution

The paper presents a new multi-feature fusion approach and recurrent visual feature injection scheme specifically designed for remote sensing vision-language tasks.

Findings

Achieves state-of-the-art performance on RS classification, captioning, and VQA.

Effectively captures small and complex structures in RS images.

Reduces visual forgetting during language generation.

Abstract

Large vision-language models (VLMs) exhibit strong performance across various tasks. However, these VLMs encounter significant challenges when applied to the remote sensing domain due to the inherent differences between remote sensing images and natural images. Existing remote sensing VLMs often fail to extract fine-grained visual features and suffer from visual forgetting during deep language processing. To address this, we introduce MF-RSVLM, a Multi-Feature Fusion Remote Sensing Vision--Language Model that effectively extracts and fuses visual features for RS understanding. MF-RSVLM learns multi-scale visual representations and combines global context with local details, improving the capture of small and complex structures in RS scenes. A recurrent visual feature injection scheme ensures the language model remains grounded in visual evidence and reduces visual forgetting during generation. Extensive experiments on diverse RS benchmarks show that MF-RSVLM achieves state-of-the-art or highly competitive performance across remote sensing classification, image captioning, and VQA tasks. Our code is publicly available at https://github.com/Yunkaidang/RSVLM.