This ReadMe should provide a basic Overview of the shown work and explain the context. It is a collective Portfolio of different Projects I worked on in the past.
This program is part of a project I finished at the nuclear physics department of Uppsala University. With an alternative approach to conventional Neural Networks, this algorithm offers a more probabilistic option to perform a discrimination of Signal and Background-events for a given dataset. In my case, this was a dataset collected at the BESIII-facility containing hadron-decay events. But the program is designed to be easily adjustable for any other dataset.
The Single-File Script is the final version of the training of the neural net and at the same time the final deliverable.
As a personal project this program learns specific faces based on given images to recognize them later. It was build to run on a Raspberry Pi 3 B+ with Camera Kit. Unkown faces are marked as "unkown"
The project is supposed to be part of a larger security system that I worked on, but was replaced by an alternative solution integrated in the Home Assistant Software.
For an internship I build this GUI to speed up the data analysis of IMU data. The IMU in my use case were strapped on a Turtle Bot to detect anomalies in the driving when driving over different surfaces.
The GUI was build with Qt and Python. And also included a tab to generate synthetic data as a base line for the LSTM training.
As part of a course in Computer Graphics I chose to build a Renderer to investigate different aspects of PBR textured objects. The program was written in C++ with OpenGL. A small menu on the top right allows to turn different textures on and off or set one specific value for roughness and metallicness texture.