Lode Vuegen

​Lode obtained his bachelor and master degree in Electronics at the KH Kempen in Belgium in 2012.  His thesis research involved a comparative study on sound classification algorithms. Supported by a SB-IWT grant he currently is researching novel acoustical classification and descriptive models for human monitoring. At first a benchmark study was performed using traditional classfier models that were learned using a supervised strategy with state-of-the art sound features on public available data sets and in-house recorded multi-channel acoustical data. He currently is working on a novel semi-supervised methodology based on Non-Negative Matrix Deconvolution that allows to build classifier models with state-of-the-art classification performance but with less requirements regarding annotation.  While developing novel algorithms he takes into account that the algorithms should be able to run on an embedded system with limited resources. 

Supervisors: dr. Peter Karsmakers, prof. dr.Bart Vanrumste (ESAT-STADIUS) & prof. dr. Hugo van Hamme (ESAT-PSI)