Exploring Models and Data for Remote Sensing Image Caption Generation

TL;DR

This paper introduces a new large-scale dataset and methods for generating accurate and flexible captions for remote sensing images, addressing the challenge of describing complex satellite imagery.

Contribution

It presents a novel dataset for remote sensing image captioning and explores models to improve descriptive accuracy and flexibility.

Findings

Generated captions effectively describe remote sensing images

The dataset enables comprehensive training and evaluation

Models achieve promising results in image captioning

Abstract

Inspired by recent development of artificial satellite, remote sensing images have attracted extensive attention. Recently, noticeable progress has been made in scene classification and target detection.However, it is still not clear how to describe the remote sensing image content with accurate and concise sentences. In this paper, we investigate to describe the remote sensing images with accurate and flexible sentences. First, some annotated instructions are presented to better describe the remote sensing images considering the special characteristics of remote sensing images. Second, in order to exhaustively exploit the contents of remote sensing images, a large-scale aerial image data set is constructed for remote sensing image caption. Finally, a comprehensive review is presented on the proposed data set to fully advance the task of remote sensing caption. Extensive experiments on the proposed data set demonstrate that the content of the remote sensing image can be completely described by generating language descriptions. The data set is available at https://github.com/201528014227051/RSICD_optimal