Large Vision-Language Models for Remote Sensing Visual Question Answering
This work addresses the problem of interpreting complex satellite imagery for open-ended questions in remote sensing analysis, representing an incremental improvement by applying generative LVLMs to a domain-specific task.
The paper tackles the challenge of Remote Sensing Visual Question Answering (RSVQA) by proposing a generative Large Vision-Language Model (LVLM) with a two-step training strategy, achieving superior performance on the RSVQAxBEN dataset and producing more accurate, relevant, and fluent answers in human evaluations.
Remote Sensing Visual Question Answering (RSVQA) is a challenging task that involves interpreting complex satellite imagery to answer natural language questions. Traditional approaches often rely on separate visual feature extractors and language processing models, which can be computationally intensive and limited in their ability to handle open-ended questions. In this paper, we propose a novel method that leverages a generative Large Vision-Language Model (LVLM) to streamline the RSVQA process. Our approach consists of a two-step training strategy: domain-adaptive pretraining and prompt-based finetuning. This method enables the LVLM to generate natural language answers by conditioning on both visual and textual inputs, without the need for predefined answer categories. We evaluate our model on the RSVQAxBEN dataset, demonstrating superior performance compared to state-of-the-art baselines. Additionally, a human evaluation study shows that our method produces answers that are more accurate, relevant, and fluent. The results highlight the potential of generative LVLMs in advancing the field of remote sensing analysis.