Using Simulated Data to Generate Images of Climate Change
This work addresses a data scarcity issue in generative models for a domain-specific application in climate change communication.
The paper tackles the problem of limited training data for GANs in domain adaptation by using simulated 3D environment images to enhance the MUNIT architecture, resulting in improved image generation for climate change awareness.
Generative adversarial networks (GANs) used in domain adaptation tasks have the ability to generate images that are both realistic and personalized, transforming an input image while maintaining its identifiable characteristics. However, they often require a large quantity of training data to produce high-quality images in a robust way, which limits their usability in cases when access to data is limited. In our paper, we explore the potential of using images from a simulated 3D environment to improve a domain adaptation task carried out by the MUNIT architecture, aiming to use the resulting images to raise awareness of the potential future impacts of climate change.