I’m a Master student at Karlsruhe Institute of Technology with a background in Electrical Engineering.
Currently, I am working on unsupervised discovery of parts and segmentation with Prof. Björn Ommer in Heidelberg (Master Thesis).
In my spare time I organize events through Hack & Söhne and take photos.
Follow me on github.
Publications and Projects
Poster Presentation at ISMRM 2018
S. Braun, X. Chen, B. Odry, B. Mailhe, M. Nadar, “Motion Detection and Quality Assessment of MR images with Deep Convolutional DenseNets”, Proc. Intl. Soc. Mag. Reson. Med. 26:2715 (2018), Paris
Poster Presentation at ISMRM 2018
S. Braun, P. Ceccaldi, X. Chen, B. Odry, B. Mailhe, M. Nadar, “Wasserstein GAN for Motion Artifact Reduction of MR images”, Proc. Intl. Soc. Mag. Reson. Med. 26:4093 (2018), Paris
Bachelor Thesis
German title: Simulation einer Lichtfeld Kamera mit Blendenmodulation. English translation: Simulation of lightfield cameras using aperture modulation.
FPGA Programming Project (VHDL)
FPGA Programming Project (VHDL) as part of my semester project.
Semester project: 50W Boost Converter circuit design
Build a 50W DC-DC Boost Converter from scratch.
Micro-controller programming : controller for 200W DC-DC buck converter
Program a micro-controller to control a 200W DC-DC buck converter.
Awards
Prize for Best Visualization
We won the prize for best visualization at the Munich RE Datathon 2018. The challenge was to estimate children’s height from point clouds. This would revolutionize the child growth monitoring in poor countries and would be a huge step towards eliminating global hunger. Our approach combined 2D and 3D data, using state of the art Computer Vision and point cloud processing.
Winner of EIT Digital Deephack Hackathon Hamburg
We won the EIT Digital Deephack Hackathon Hamburg 2019 with our project ScrubberWatch. The idea behind ScrubberWatch is to detect the level of emissions of incoming ships in the Hamburg Harbour using Computer Vision.
Teaching and Workshops
Workshop at GPN 19
I gave a workshop at Gulasch Programmiernacht 2019 on Machine Learning Workflow Tools together with Leander Kurscheidt.
Deep Learning Workshop
A workshop I did through Hack & Söhne together with Maximilian Franz and Leander Kurscheidt. We wanted to make very low level concepts clear and not just do tensorflow.train().
TiTrAa Workshop
After my bachelor thesis I was frustrated that the writing workflow is so hard to learn. That’s why I made a workshop about it. Why is it called TiTrAa? TiTrAa stands for ‘Tipps und Tricks zu Abschlussarbeit’ and is German for ‘thesis workflow hacks’.
Mathematics Tutorials
I gave math tutorials from 2014 to 2016. It is important to get feedback about your teaching methods which is why I wrote an app to monitor the quality of my tutorials.
Some other things I do and did
Open Codes Hackathon
I was part of the organizer team and a speaker for the Open Codes Hackathon. The Open Codes Hackathon was the largest student-run hackathon in Germany with international 200 participants.
Machine Learning Karlsruhe
ML-KA hosts weekly paper discussions group meetings (PDG) about Machine Learning. The classic pdg is Wednesdays. We recently started a theory-PDG branch that focusses on theoretical aspects of Machine Learning.