Probabilistic methods for modeling and capturing human motion (2V+1U, 4.0 LP)
|SWS||Type||Course Form||CP (Effort)||Presence-Time / Self-Study|
|-||K||Lecture with exercise classes (V/U)||4.0 CP||78 h|
|1||U||Exercise class (in small groups)||14 h|
|(2V+1U)||4.0 CP||42 h||78 h|
|CP, Effort||4.0 CP = 120 h|
|Position of the semester||1 Sem. in WiSe|
|Level|| Master (General)|
|Area of study||[INF-KI] Intelligent Systems|
Possible Study achievement
- Verification of study performance: proof of successful participation in the exercise classes (ungraded)
The focus of this course is on probabilistic methods and models for IMU (inertial measurement unit) based human motion modelling and capturing, aiming at biomechanically interpretable motion estimation (e.g. 3D kinematics). After motivating this general topic via different application examples and presenting an overview of different capturing modalities, the following topics are addressed:
- Human body, kinematic models & marker-based optical motion capture.
- 3D kinematics
- Inertial measurement devices
- Probabilistic model-based sensor fusion methods
- Statistical machine learning techniques.
The techniques covered will be illustrated with practical examples from current research. The contents of the lectures are also presented in the form of laboratory exercises with implementations in Matlab and Python.
- Gustafsson, Fredrik. Statistical sensor fusion. Studentlitteratur, 2010.
- Thrun, Sebastian, Wolfram Burgard, and Dieter Fox. Probabilistic robotics. MIT press, 2005.
- Murphy, Kevin P. Machine Learning: A Probabilistic Perspective, MIT Press, 2013