DDFAV: Remote Sensing Large Vision Language Models Dataset and Evaluation Benchmark

TL;DR

This paper introduces DDFAV, a high-quality remote sensing dataset for large vision language models, along with an instruction set and an evaluation method to improve and assess model performance in remote sensing tasks.

Contribution

The paper presents a novel remote sensing LVLM dataset, a tailored instruction set, and an evaluation benchmark to address hallucinations and performance issues in remote sensing applications.

Findings

DDFAV dataset enhances remote sensing LVLM training.

RSPOPE evaluation method effectively measures hallucinations.

Zero-shot performance varies across different LVLMs.

Abstract

With the rapid development of large vision language models (LVLMs), these models have shown excellent results in various multimodal tasks. Since LVLMs are prone to hallucinations and there are currently few datasets and evaluation methods specifically designed for remote sensing, their performance is typically poor when applied to remote sensing tasks. To address these issues, this paper introduces a high quality remote sensing LVLMs dataset, DDFAV, created using data augmentation and data mixing strategies. Next, a training instruction set is produced based on some high-quality remote sensing images selected from the proposed dataset. Finally, we develop a remote sensing LVLMs hallucination evaluation method RSPOPE based on the proposed dataset and evaluate the zero-shot capabilities of different LVLMs. Our proposed dataset, instruction set, and evaluation method files are available at https://github.com/HaodongLi2024/rspope.