Seeing Through the Clouds: Cloud Gap Imputation with Prithvi Foundation Model
This addresses the need for accurate data analysis in remote sensing applications, but it is incremental as it compares existing methods on a specific domain task.
The study tackled the problem of filling cloudy pixels in multispectral satellite imagery by comparing a fine-tuned Vision Transformer model with a baseline Conditional Generative Adversarial Network, finding that the ViT model achieved higher imputation accuracy as measured by metrics like structural similarity index and mean absolute error.
Filling cloudy pixels in multispectral satellite imagery is essential for accurate data analysis and downstream applications, especially for tasks which require time series data. To address this issue, we compare the performance of a foundational Vision Transformer (ViT) model with a baseline Conditional Generative Adversarial Network (CGAN) model for missing value imputation in time series of multispectral satellite imagery. We randomly mask time series of satellite images using real-world cloud masks and train each model to reconstruct the missing pixels. The ViT model is fine-tuned from a pretrained model, while the CGAN is trained from scratch. Using quantitative evaluation metrics such as structural similarity index and mean absolute error as well as qualitative visual analysis, we assess imputation accuracy and contextual preservation.