392146 Machine Learning for Remote Photoplethysmography (S) (WiSe 2026/2027)

This course covers the principles of remote photoplethysmography (rPPG) and their application in video-based physiological signal extraction. In the first part of the course, students will learn the fundamentals of rPPG, including the underlying optical and physiological principles, alongside core video processing techniques and a survey of state-of-the-art methods in the field. This will also be complemented with exercises that will focus on reimplementing the papers and evaluating them on public datasets. In the second part, students will work in groups to design and implement their own rPPG method as part of a structured competition/challenge. Groups will compete against each other on a defined benchmark. Each group is expected to document their work in a written report that describes the method in a detailed and reproducible manner. Prerequisite: Foundations of Machine Learning and Deep Learning