Search

Physiological Feature Representation Learning


Description Physiological signals provide valuable information about the state of the body and mind. These signals have been successfully applied to emotion recognition in different scenarios, such as human-machine communication or human-human interaction.
Task The aim of this project is to generate robust and strong representations of physiological signals in order provide augmented training material for deep learning systems
Utilises Convolutional Neural Networks, auDeep, Deep Spectrum System.
Requirements Preliminary knowledge in Machine Learning, Good programming skills (e.g. Python, C++)
Languages English and German.
Supervisor Shahin Amiriparian, M. Sc. (shahin.amiriparian@informatik.uni-augsburg.de)