Spatial-Temporal Super-Resolution of Satellite Imagery via Conditional Pixel Synthesis
This work addresses the challenge of scaling satellite-based tasks like population measurement and biodiversity monitoring by providing a cost-effective solution for generating high-resolution imagery, though it is incremental as it builds on existing conditional synthesis methods.
The paper tackles the problem of generating high-resolution satellite imagery from low-resolution inputs to address the scarcity and cost of high-resolution data, achieving photo-realistic quality and outperforming baselines on object counting, especially in rapidly changing geographic areas.
High-resolution satellite imagery has proven useful for a broad range of tasks, including measurement of global human population, local economic livelihoods, and biodiversity, among many others. Unfortunately, high-resolution imagery is both infrequently collected and expensive to purchase, making it hard to efficiently and effectively scale these downstream tasks over both time and space. We propose a new conditional pixel synthesis model that uses abundant, low-cost, low-resolution imagery to generate accurate high-resolution imagery at locations and times in which it is unavailable. We show that our model attains photo-realistic sample quality and outperforms competing baselines on a key downstream task -- object counting -- particularly in geographic locations where conditions on the ground are changing rapidly.