Can Location Embeddings Enhance Super-Resolution of Satellite Imagery?
This addresses the need for higher-resolution satellite imagery for applications like urban planning and disaster response, though it appears incremental as it builds on existing GAN and diffusion model techniques.
The paper tackles the problem of poor generalization in satellite imagery super-resolution across diverse geographic regions by proposing a framework that incorporates geographic context through location embeddings. The method shows significant improvements over state-of-the-art methods on building segmentation tasks.
Publicly available satellite imagery, such as Sentinel- 2, often lacks the spatial resolution required for accurate analysis of remote sensing tasks including urban planning and disaster response. Current super-resolution techniques are typically trained on limited datasets, leading to poor generalization across diverse geographic regions. In this work, we propose a novel super-resolution framework that enhances generalization by incorporating geographic context through location embeddings. Our framework employs Generative Adversarial Networks (GANs) and incorporates techniques from diffusion models to enhance image quality. Furthermore, we address tiling artifacts by integrating information from neighboring images, enabling the generation of seamless, high-resolution outputs. We demonstrate the effectiveness of our method on the building segmentation task, showing significant improvements over state-of-the-art methods and highlighting its potential for real-world applications.