Mask Conditional Synthetic Satellite Imagery
This work addresses data scarcity in satellite imagery analysis for researchers and practitioners, offering an incremental improvement through data augmentation with synthetic imagery.
The paper tackles the problem of generating synthetic satellite imagery datasets by proposing a mask-conditional model that uses real images and land cover masks to train a generator, which then produces synthetic data for downstream tasks. The result shows that models trained on a mix of real and synthetic imagery outperform those using only real data, achieving a mean Intersection over Union (mIoU) of 0.5834 compared to 0.5235.
In this paper we propose a mask-conditional synthetic image generation model for creating synthetic satellite imagery datasets. Given a dataset of real high-resolution images and accompanying land cover masks, we show that it is possible to train an upstream conditional synthetic imagery generator, use that generator to create synthetic imagery with the land cover masks, then train a downstream model on the synthetic imagery and land cover masks that achieves similar test performance to a model that was trained with the real imagery. Further, we find that incorporating a mixture of real and synthetic imagery acts as a data augmentation method, producing better models than using only real imagery (0.5834 vs. 0.5235 mIoU). Finally, we find that encouraging diversity of outputs in the upstream model is a necessary component for improved downstream task performance. We have released code for reproducing our work on GitHub, see https://github.com/ms-synthetic-satellite-image/synthetic-satellite-imagery .