Quentin Paletta

I am a research fellow at the Climate Office of the European Space Agency (ESA) in Harwell in collaboration with the Philab of the European Space Research Institute in Frascati, where I work on computer vision and machine learning for solar energy forecasting. 🌤️

Previously at the University of Cambridge, I have developped deep learning algorithms for advancing solar nowcasting based on cloud cover observations from sky cameras and weather satellites.

Also working on solar energy meteorology? Don't hesitate to contact me to join our community on Slack! (80+ researchers and engineers from around the world)

Google Scholar  /  Research Gate  /  Twitter  /  Github

Email  /  CV  /  LinkedIn

profile photo


I'm interested in computer vision, machine learning, earth observations and solar energy. Much of my research is about improving solar forecasting tools from cloud cover observations. Representative papers are highlighted.

Multilocation sky image dataset for solar forecasting and atmospheric science.
Yuhao Nie, Quentin Paletta, Sherrie Wang,
Tackling Climate Change with Machine Learning workshop (ICLR), 2024
paper project

A multilocation sky image dataset for solar forecasting and atmospheric sciences.

blind-date Sky image-based solar forecasting using deep learning with heterogeneous multi-location data: Dataset fusion versus transfer learning
Yuhao Nie, Quentin Paletta, Andea scott, Luis Martin Pomares, Guillaume Arbod, Sgouris Sgouridis, Joan Lasenby, Adam Brandt,
Applied Energy, 2024

Comparative study between local training, global training, and transfer learning for solar nowcasting based on sky images.

blind-date Improving cross-site generalisability of vision-based solar forecasting models with physics-informed transfer learning
Quentin Paletta, Yuhao Nie, Yves-Marie Saint-Drenan, Bertrand Le Saux,
Energy Conversion and Management, 2024

Forecasting solar energy accurately is good, generalising beyond the location where the model was trained is even better! Here, we explore diverse zero-shot and few-shot learning tasks based on a model pre-trained at another location.

blind-date SkyGPT: Probabilistic Ultra-short-term Solar Forecasting Using Synthetic Sky Images from Physics-constrained VideoGPT
Yuhao Nie, Eric Zelikman, Andea scott, Quentin Paletta, Adam Brandt,
Advances in Applied Energy, 2024

We introduce SkyGPT, a physics-informed stochastic video prediction model that is able to generate multiple possible future images of the sky with diverse cloud motion patterns, by using past sky image sequences as input.

blind-date Open-source ground-based sky image datasets for very short-term solar forecasting, cloud analysis and modeling: A comprehensive survey
Yuhao Nie, Xiatong Li, Quentin Paletta, Max Aragon, Andea scott, Adam Brandt,
Renewable and Sustainable Energy Reviews, 2024

Comprehensive survey of 72 open-source ground-based sky image datasets for very short-term solar forecasting applications.

blind-date Advances in Solar Forecasting: Computer Vision with Deep Learning
Quentin Paletta, Guillermo Terrén-Serrano, Yuhao Nie, Binghui Li, Jacob Bieker, Wenqi Zhang, Laurent Dubus, Soumyabrata Dev, Cong Feng,
Advances in Applied Energy, 2023

Literature review on the field of computer vision-based solar forecasting with a particular focus at the application of deep learning to process cloud cover observations collected by sky cameras or meteorological satellites.

Omnivision forecasting: Combining satellite and sky images for improved deterministic and probabilistic intra-hour solar energy predictions
Quentin Paletta, Guillaume Arbod, Joan Lasenby,
Applied Energy, 2023

Combining sky images and satellite observations into a single machine learning framework for intra-hour solar forecasting.

Cloud Flow Centring in Sky and Satellite Images for Deep Solar Forecasting
Quentin Paletta, Guillaume Arbod, Joan Lasenby,
8th World Conference on Photovoltaic Energy Conversion, 2022

Processing method for cloud cover videos to consistently centre the polar representation on the incoming flow of clouds using an optical flow algorithm.

SPIN: Simplifying Polar Invariance for Neural networks Application to vision-based irradiance forecasting
Quentin Paletta, Anthony Hu, Guillaume Arbod, Philippe Blanc, Joan Lasenby,
Conference on Computer Vision and Pattern Recognition (CVPR) Workshop, 2022

Comparative study between various data augmentation and scene presentation methods including the use of polar coordinates. Application to solar forecasting from sky images and satellite observations.

ECLIPSE: Envisioning CLoud Induced Perturbations in Solar Energy
Quentin Paletta, Anthony Hu, Guillaume Arbod, Joan Lasenby,
Applied Energy, 2022

Novel deep learning architecture for solar irradiance and sky image prediction.

Benchmarking of Deep Learning Irradiance Forecasting Models from Sky Images - An in-depth Analysis
Quentin Paletta, Guillaume Arbod, Joan Lasenby,
Solar Energy, 2021

Benchmarking of standard deep learning models for solar power nowcasting using sky images.

A Temporally Consistent Image-based Sun Tracking Algorithm for Solar Energy Forecasting Applications
Quentin Paletta, Joan Lasenby,
NeurIPS 2020 - Tackling Climate Change with Machine Learning workshop, 2020
project page / paper / slides

Sun tracking algorithm based on sky images only.

Design and source code from Jon Barron's website.