GeoTorchAI: A Framework for Training and Using Spatiotemporal Deep Learning Models at Scale

A framework for quality assessment of exercises in physical rehabilitation based on skeletal joint displacements collected with a motion capture system.

[PAKDD 2021] SST-GNN: Simplified Spatio-temporal Traffic forecasting model using Graph Neural Network

[Information Fusion 2026] A collection of deep learning models for remote sensing spatiotemporal fusion.

PurpleAirSF: datasets for realistic air quality forecasting

Neural Networks and Gaussian Processes: an integrated approach to spatio-temporal modelling for air quality

Geostatistical modeling of urban air pollutants (NO₂ and PM₁₀) across Los Angeles County using the Hidden Dynamic Geostatistical Model (HDGM) framework implemented in MATLAB with the D-STEM package. Includes full data processing, model fitting, cross-validation, and spatial prediction workflow.