Procedural Synthesis of Remote Sensing Images for Robust Change Detection with Neural Networks
This addresses data scarcity for remote sensing applications, but it is incremental as it applies existing procedural generation techniques to a specific domain.
The paper tackles the problem of limited annotated data for change detection in remote sensing images by proposing a method to create realistic synthetic datasets using game development engines, resulting in improved performance and convergence of deep learning models.
Data-driven methods such as convolutional neural networks (CNNs) are known to deliver state-of-the-art performance on image recognition tasks when the training data are abundant. However, in some instances, such as change detection in remote sensing images, annotated data cannot be obtained in sufficient quantities. In this work, we propose a simple and efficient method for creating realistic targeted synthetic datasets in the remote sensing domain, leveraging the opportunities offered by game development engines. We provide a description of the pipeline for procedural geometry generation and rendering as well as an evaluation of the efficiency of produced datasets in a change detection scenario. Our evaluations demonstrate that our pipeline helps to improve the performance and convergence of deep learning models when the amount of real-world data is severely limited.