Suche

Activity detection from Smartphones and Wearables


Description Sports and activity tracking are a key  application of wearable devices, however noisy and unreliable data streams make this a challenging machine learning task
Task The interested student(s) will explore deep learning solutions on to increase the reliability and accuracy of activity tracking systems
Utilises Python and related deep learning toolkit, transfer learning, data augmentation
Requirements Preliminary knowledge in Machine Learning, Good programming skills (e.g. Python, C++)
Languages English
Supervisor Dr. Nicholas Cummins (nicholas.cummins@informatik.uni-augsburg.de)