Visual-Grounded Hierarchical Perception for Remote Sensing Visual Question Answering

Remote sensing visual question answering (RSVQA) leverages vision and natural language for intelligent Earth observation. While multimodal large language models (LLMs) demonstrate broad potential in remote sensing (RS) multitask frameworks, their heavy reliance on textual priors from extensive conversational pretraining and neglect of the hierarchical nature of RS scenes limit their performance. To address these limitations, this article proposes a visual-grounded hierarchical perception (VGHP) framework that activates a pretrained LLM’s open-world knowledge under the joint guidance of visual evidence and linguistic semantics for reliable answer generation. Specifically, a visual-grounded language encoder (VGLE) dynamically binds question semantics to the visual context via cross-modal attention at every encoding layer, effectively mitigating the language prior bias by enforcing vision-conditioned reasoning. Concurrently, a hierarchical vision perceiver (HVP) is designed to capture multigranular visual information, routing local details and global semantics across feature hierarchies in a vision tower, thereby addressing multiscale perception challenges in RS images. Extensive experiments on three RSVQA benchmarks showcase the superiority of the proposed VGHP, achieving state-of-the-art accuracy while maintaining computational efficiency.