New Paper published at Sensors MDPI

Symbolic picture for the article. The link opens the image in a large view.
We are pleased to announce that our latest paper “Wearable Sensors for Activity Recognition in Ultimate Frisbee Using Convolutional Neural Networks and Transfer Learning“ is now published at Sensors MDPI.
The Paper is published open access:
We developed the first activity recognition system for Ultimate Frisbee. Therefore we trained a Convolutional Neural Network to distinguish seven different throwing techniques plus catches. Secondly, we pre-trained the same architecture on an existing volleyball dataset to investigate the potential of transfer learning for marginal sports. Thirdly, we explored the generalization capabilities of transfer learning when dealing with small-scale datasets. We show that transfer learning can be beneficial, especially when dealing with small datasets, as in marginal sports, and therefore, can improve the tracking of marginal sports.
Congratulations to all authors!
Johannes Link, Timur Perst, Maike Stöve and Bjoern Eskofier