RSGPT: A remote sensing vision language model and benchmark

Abstract

The emergence of large-scale Large Language Models (LLMs), with GPT-4 as a prominent example, has significantly propelled the rapid advancement of Artificial General Intelligence (AGI) and sparked the revolution of Artificial Intelligence 2.0. In the realm of remote sensing, there is a growing interest in developing large vision language models (VLMs) specifically tailored for data analysis in this domain. However, current research predominantly revolves around visual recognition tasks, lacking comprehensive, high-quality image–text datasets that are aligned and suitable for training large VLMs, which poses significant challenges to effectively training such models for remote sensing applications. In computer vision, recent research has demonstrated that fine-tuning large vision language models on small-scale, high-quality datasets can yield impressive performance in visual and language understanding. These results are comparable to state-of-the-art VLMs trained from scratch on massive amounts of data, such as GPT-4. Inspired by this captivating idea, in this work, we build a high-quality Remote Sensing Image Captioning dataset (RSICap) that facilitates the development of large VLMs in the remote sensing field. Unlike previous remote sensing datasets that either employ model-generated captions or short descriptions, RSICap comprises 2,585 human-annotated captions with rich and high-quality information. This dataset offers detailed descriptions for each image, encompassing scene descriptions (e.g., residential area, airport, or farmland) as well as object information (e.g., color, shape, quantity, absolute position, etc.). To facilitate the evaluation of VLMs in the field of remote sensing, we also provide a benchmark evaluation dataset called RSIEval. This dataset consists of human-annotated captions and visual question–answer pairs, allowing for a comprehensive assessment of VLMs in the context of remote sensing. We are actively engaged in expanding the scale of these two datasets to cover a broader spectrum of remote sensing image understanding tasks, further enhancing their utility and applicability. Our dataset and codes will be released at https://github.com/Lavender105/RSGPT.