# Seeing Through the Clouds: Cloud Gap Imputation with Prithvi Foundation Model

arXiv: https://arxiv.org/abs/2404.19609 · HTML: https://mlsanity.com/paper/2404.19609/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 14 references that are in the map.
- First posted on arXiv 30 Apr 2024.

## Builds on (most cited first, 11 of 11)

- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Image-to-Image Translation with Conditional Adversarial Networks](https://mlsanity.com/paper/1611.07004/) (2016)
- [SatMAE: Pre-training Transformers for Temporal and Multi-Spectral Satellite Imagery](https://mlsanity.com/paper/2207.08051/) (2022)
- [SatCLIP: Global, General-Purpose Location Embeddings with Satellite Imagery](https://mlsanity.com/paper/2311.17179/) (2023)
- [Foundation Models for Generalist Geospatial Artificial Intelligence](https://mlsanity.com/paper/2310.18660/) (2023)
- [GEO-Bench: Toward Foundation Models for Earth Monitoring](https://mlsanity.com/paper/2306.03831/) (2023)
- [Mission Critical -- Satellite Data is a Distinct Modality in Machine Learning](https://mlsanity.com/paper/2402.01444/) (2024)
- [Patch-Based Image Inpainting with Generative Adversarial Networks](https://mlsanity.com/paper/1803.07422/) (2018)
- [A Billion-scale Foundation Model for Remote Sensing Images](https://mlsanity.com/paper/2304.05215/) (2023)
- [Deep Adversarial Training for Multi-Organ Nuclei Segmentation in Histopathology Images](https://mlsanity.com/paper/1810.00236/) (2018)
- [Synthesizing Optical and SAR Imagery From Land Cover Maps and Auxiliary Raster Data](https://mlsanity.com/paper/2011.11314/) (2020)

## Built on by (most cited first, 1 of 1)

- [From Surface Forecasting to Observability Forecasting: A Latent World Model for Cloud-Aware EO Monitoring](https://mlsanity.com/paper/2607.13651/) (2026)

## Abstract (arXiv)

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.