Summerschool on Probabilistic Morphable Models

A follow-up on the online course on statistical shape modelling

June 6th - June 10th, 2017
Basel, Switzerland

Endorsed by

Probabilistic Morphable Models

Overview

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 for practial applications in medical image analysis and 2D face image analysis.

Course organization

The course consists of two parts: The basic theory on shape modelling and practical introduction to Scalismo is given as a open online course and taken individually by the participants. Participants of this course, who submit a solution to the course project, are invited to attend the summer school, which is hosted at the University of Basel. In the summer school, participants will learn about probabilistic methods for model fitting, and can deepen their knowledge on shape modelling in practical projects.

Target audience

The intended audience for the course are beginning graduate students and researchers in computer vision and medical image analysis, who plan to apply statistical shape models as part of their work.

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

Information

Department of Mathematics and Computer Science Location of the Apaliving Basel Budget Hotel

Venue

Classes will take place at the Deparment of Mathematics and Computer Science of the University of Basel (Lecture hall 00.003, Spiegelgasse 1.)

Basel

Basel is the second largest town in Switzerland. It can be reached by plane, by either flying directly to Basel Airport, or to Zurich (1 hour train ride to Basel).

Registration fee

The summerschool is free of charge, but all participants need to organize their stay (travel, accomodation, food) by themselves.

Accomodation

Our main recommondation is the Apaliving Basel Budget Hotel. The price per night for a single room approximately Fr. 90. In the same neighborhood you can also find the following, slighly cheaper options:

All these hotels are within walking distance of each other and located a 15 minutes bus ride from the Department of Mathematics and Computer Science.

Instructors

Lecturers

Sandro Schönborn
Sandro Schönborn is a senior reseracher at the University of Basel. His research focuses on probabilistic methods for fitting statistical shape models, in particular sampling methods.
Sandro Schönborn
Thomas Vetter
Thomas Vetter is the founder and head of the Graphics and Vision Research Group of the University of Basel. Thomas Vetter’s research at the University of Basel is directed at automated synthesis and analysis of images.
Thomas Vetter
Marcel Lüthi
Marcel Lüthi is a senior-researcher at the University of Basel. His research focuses on the use of Gaussian Processes for shape modelling for application in medical image analysis.
Marcel Lüthi

Tutors

Ghazi Bouabene
Ghazi Bouabene is a senior resarcher at the Graphics and Vision Resarch Group. He is one of the main developers of Scalismo, and has been working in several projects that aim at bringing shape modelling into industrial applications.
Ghazi Bouabene
Andreas Schneider
Andreas Schneider is a PhD Student at the Graphics and Vision research group. He works on face-image analysis, in particular the synthesis of realistic skin texture.
Andreas Schneider
Thomas Gerig
Thomas Gerig is a PhD Student at the Graphics and Vision research group. His research focuses on shape modelling using Gaussian processes, in particular using multi-scale methods for modelling shape deformations.
Thomas Gerig
Andreas Morél-Forster
Andreas Morél-Forster is a Postdoctoral Researcher at the Graphics and Vision research group. His research focuses on shape model fitting, using MCMC methods, with applications in both face image analysis and medical image analysis.
Andreas Morél-Forster
fe

Programme

Online course (Feb 2017 - May 2017)

From To Subject
Mon 2/20     Sun 2/26    Basic concepts of shape modelling and the scalismo software.
Mon 2/27 Sun 3/5 Building shape models: fundamental concepts in theory and practice.
Mon 3/6 Sun 3/12Probabilistic aspects of shape models.
Mon 3/13 Sun 3/19Modelling with Gaussian Processes
Mon 3/20 Sun 3/26 Reconstructing missing parts
Mon 3/27 Sun 4/2Fitting models to data
Mon 4/3 Sun 4/9 Model-based image analysis
Mon 4/9 Sun 4/30 Course Project

Summerschool (June 6 - 10, 2017)

Day Time Subject
Tue 6/6MorningLecture: Analysis by Synthesis. Basics of computer graphics.
Tue 6/6AfternoonExercise: Working with Scalismo-faces. Morphable Models and rendering
Lecture: Bayesian modelling
Wed 7/6MorningLecture: Probabilistic Model Fitting using MCMC
Wed 7/6AfternoonExercise: Introduction to course project, Landmark fitting using MCMC
Thu 8/6MorningLecture: 2D Face image analysis
Thu 8/6AfternoonLecture: Debugging MCMC
Exercise: Course project
Fri 9/6MorningLecture: Gaussian processes
Lecture: Non-rigid registration
Fri 9/6AfternoonExercise: Course Project
Sat 10/6MorningProject presentations
Sat 10/6AfternoonSocial event

In the course project, the participants will implement their own pipeline for recovering the 3D shape of a face from a 2D picture.

Classes start at 9.00 and take place in the lecture hall 00.003 (Department of Mathematics and Computer Science, Spiegelgasse 1.)

Registration

To register for the summer school, fill in the registration from below. Deadline for the registration is March, 20. 2017. For additional questions please send us an email to the address given below.

Registration form

marcel.luethi@unibas.ch

Notifications of acceptance will be provided after submission of the course project, which is part of the online course. The latest possible submission of the course project is April, 30. 2017.