We are very excited and proud to announce that today our student Amber Mulder won the Geomatics best thesis award 2020 at this year’s Geomatics day! Her thesis titled ‘Semantic segmentation of RGB-Z aerial imagery using convolutional neural networks’ was supervised by Balázs Dukai and Ravi Peters and done in cooperation with the company Readar. The jury praised the work as being of ‘exceptional quality’ and commented it was quite rare to beat state of the art results in a MSc thesis. And despite the fact that the code was not open sourced due to the collaboration with a commercial company, the jury said that it would still be reproducable due to the clarity and completeness of the thesis.
Congratulations Amber, you did an excellent job! 🥳
For those interested, you can find a link to her thesis below.