Probabilistic Morphable Models
-

Open online course


1. March - 24. May 2021

Course overview

Probabilistic Morphable Models

Probabilistic Morphable Models are a fully probabilistic approach to model-based image analysis, based on:

  • the theory of Gaussian processes for modelling shapes
  • Markov chain Monte Carlo (MCMC) methods for model fitting
  • a software framework for efficient development of image analysis solutions.

This course teaches the theoretical basis of shape modelling and model fitting and provides an introduction on how to use the software framework Scalismo. The participants will apply the learned concepts and methods in two course projects:

  • Project 1: 3D Reconstruction of the full femur From partical surfaces.
  • Project 2: (Interactive-) Model-based segmentation of 3D CT images.

Course organization

The course consists of two parts:

The first part, will focus on the basic theory of shape modelling using Gaussian processes. The participants are also given a practical introduction to shape modelling in Scalismo.

The second part will focus on Bayesian inference and model fitting using Markov Chain Monte Carlo methods and how efficient samplers for shape and image analysis can be formulated and implemented in Scalismo.

  • All the course material (videos, article, software tutorials) will be offered directly on this website.
For both parts of the course, the instructors will answer questions regarding the course content of the week in an online forum.

Target audience

The intended audience for the course is beginning graduate students and researchers from academia and industry interested in shape and image analysis. While the examples are predominately medical, the taught principles and techniques are directly applicable to applications in computer vision.

Prerequisite is knowledge of basic probability theory and statistics as well as linear algebra. Participants should have good programming skills in at least one programming language (ideally Java, C++ or Python).

Programme

Online course of FutureLearn
1. March 2021 - 26. April 2021

From To Subject
Mon 1. March     Sun 7. March     Basic concepts of shape modelling and the scalismo software.
Mon 8. March Sun 14. March Building shape models: fundamental concepts in theory and practice.
Mon 15. March Sun 2. March Probabilistic aspects of shape models.
Mon 22. March Sun 28. March Modelling with Gaussian Processes
Mon 29. March Sun 5. April Reconstructing missing parts
Mon 6. April Sun 12. April Fitting models to data
Mon 13. April Sun 19. April Model-based image analysis
Mon 20. April Sun 26. April Course Project

Online Course on Probabilistic fitting
27 April 2021 - 24 Mai 2021

From To Subject
Mon 27. April     Sun 3. Mai     Analysis by synthesis - Bayesian inference
Mon 4. Mai Sun 10. Mai Markov Chain Monte Carlo Methods - Basic ideas and methods
Mon 11. Mai Sun 17. Mai Markov Chain Monte Carlo Methods for shape model fitting
Mon 18. Mai Sun 24. Mai Designing effective proposals for shape model fitting

Registration

To registration for the online course on FutureLearn, please go directly to FutureLearn course page. There is no need to register for the second part of the course.

For questions regarding the course, please send an email to the address below.

marcel.luethi@unibas.ch