The course aims to provide practical knowledge for deep learning-based processing and analysis of remote sensing images. The course will start by introducing the individual engineering steps related to setting up a dataset and pre-processing it efficiently. Then, a training pipeline will be implemented, and a variety of deep learning experiments will be conducted. Relevant data and storage formats will be introduced, alongside modern deep learning methods and architectures, giving particular attention to efficiency, flexibility and seamless conduction of experiments in Python to enable efficient large-scale remote sensing image analytics with deep learning. Practical applications will be provided throughout the course.
- Trainer/in: Mathis Jürgen Adler
- Trainer/in: Tom Oswald Burgert
- Trainer/in: Kai Norman Clasen
- Trainer/in: Begüm Demir
- Trainer/in: Leonard Wayne Hackel
- Trainer/in: Jonas Klotz