| CARVIEW |
Spatial-Temporal Super-Resolution of Satellite Imagery via Conditional Pixel Synthesis
Abstract
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.
Model
An illustration of our proposed framework (discriminator omitted). Details can be found in the paper.
Results
Samples from all models on the Texas housing dataset. Our models show advantages in both sample quality and structural detail consistency with the ground truth, especially in areas with house or pool construction (zoomed in with colored boxes).
Our method is able to generate images given a low resolution image timestamped either before or after the high resolution image.
Samples from all models on the Functional Map of the World crop field dataset.
BibTeX
@inproceedings{he2021spatial,
title={Spatial-Temporal Super-Resolution of Satellite Imagery via Conditional Pixel Synthesis},
author={He, Yutong and Wang, Dingjie and Lai, Nicholas and Zhang, William and Meng, Chenlin and Burke, Marshall and Lobell, David B. and Ermon, Stefano},
year={2021},
month={December},
abbr={NeurIPS 2021},
booktitle={Neural Information Processing Systems},
}