Watch Intro Video

Welcome!

Course curriculum

  • 1

    Module 1 - Earth Observation & the Geospatial domain

    • Earth Observation introduction, principles & applications

    • Earth Observation physics & image characteristics

    • Earth Observation platforms & sensors

    • Some questions to review

  • 2

    Module 2 - Machine / Deep Learning Concepts

    • Machine Learning Concepts

    • Terminology quiz

    • Some questions to review

  • 3

    Module 3 - Diving into the Picterra platform - Imagery, and Account Basics

    • Setup Tasks

    • Diving into the Picterra platform - Imagery and Account Basics

    • One more thing!

  • 4

    Module 4 - Training your first detector

    • Training your first detector

  • 5

    Module 5 - Running your detector and managing results

    • Running your detector & managing results

    • Building Detector Exercise

    • Building Detector Solution

    • Sheep Detector Exercise

    • Sheep Detector Solution

  • 6

    Module 6 - Advanced features & Next Developments

    • Advanced Features & Next Developments

    • Road Cracks Detector Exercise

    • Road Cracks Detector Solution

    • Solar Panels Detector Exercise

    • Solar Panels Detector Solution

  • 7

    Module 7 - Picterra API & python library

    • Setup tasks

    • Picterra API & Python Library

  • 8

    Module 8 - 3rd party tools (QGIS and Tableau)

    • Setup Tasks

    • 3rd Party Tools (QGIS & Tableau)

  • 9

    Module 9 - Case studies

    • Case studies

  • 10

    Module 10 - Now it's your turn!

    • Use case brainstorming!

    • Course Survey

Instructor(s)

CTO

Frank De Morsier

Frank received the BSc degree in Electrical Engineering and the MSc degree in Information Technology with a Minor in Space Technologies from the EPFL, Switzerland, ranked as world's 12th and Europe 2nd best engineering university. He holds a PhD in machine learning applied to remote sensing imagery from the EPFL-LTS5 and worked in the private industry in the field of remote sensing image processing since 2014. He is also lecturer of the “Image Processing for Earth Observation” master course at EPFL. He has an extensive practical and theoretical knowledge of machine learning and pattern recognition for image processing. Co-founder and CTO of Picterra, his focus is on the development and applications of machine learning algorithms for multitemporal and multisensors image analysis, with a special emphasis on approaches allowing to upscale processing chains while keeping end-user interactions within the loop.

Computer Vision Scientist

Roger Fong

Roger received his B.Sc degree in Computer Science from Brown University in Providence, RI, USA in 2012. After graduating he was hired as a full-time software engineer at Apple headquarters in Cupertino, California from 2012 to 2015 where he worked on the open source WebKit framework. He finished his EPFL Master degree in Computer Science with a minor in Micro Engineering by performing a combined internship/master thesis at Picterra in association with the Computer Vision Lab from EPFL. He is now Computer Vision engineer at Picterra and focuses on the application of Machine Learning algorithms on remote sensing imagery. He has worked primarily with deep learning models and is particularly interested in image recognition and object detection applications.

Lead Software & ML Engineer

Julien Rebetez

Julien received the MSc degree in Information and Communication Technologies from the University of Applied Science Western Switzerland (HES-SO) in 2012. After an internship at Google during his studies, he worked for 4 years as a R&D engineer focused on Machine Learning at HEIG-VD, working on a number of remote sensing data analysis projects. In 2016, he joined MindMaze as a Senior Software Engineer working on a range of technologies from low-level C++ to Unity 3D games. He joined Picterra in 2018 where he works on the Picterra Platform, both on Software Architecture and on the underlying Machine Learning. He has also contributed to a number of machine learning / data science open source projects and is generally interested about the latest findings in Deep Learning research.

Welcome to the Picterra Community!

Join us on our community slack channel where you can discuss course content with your fellow students and Picterrans.