Dr. Dominic Spata

I am Dominic Spata, a computer science and machine learning expert living in Witten, Germany. Originally, I graduated from the Ruhr University Bochum with a master's degree in computer science. Even then my studies focused on machine learning, especially as applied to aspects of driving and traffic. I've since completed a doctoral degree researching the use of neural networks for automotive radar perception at the Bergische Universität Wuppertal. I am currently continuing that work as a full-time machine learning engineer at automotive supplier Aptiv at their site in Wuppertal.

Privately, I'm a fan of games of all forms (be they sports, board games, computer games, or virtual reality), like to read and write fantasy, dabble in web and graphics design¹, and have an interest in pretty much anything and everything related to programming. Another big part of my life is exercise. I like to keep active with a variety of activities like yoga, traditional weight training, swimming, and more.

¹This website was created by myself from scratch in pure HTML and CSS.

Title: Action Classification Using a Combination of Hough Voting and Random Forest

Advisors: PD Dr. Rolf Würtz, Dipl.-Inform. André Ibisch

This thesis deals with the classification of human behaviour using a recent method. Given a video of a single human performing some distinct action as well as annotations indicating that human's position, it extracts 3D feature patches based on the dense visual flow of the image material. A random forest regressor is used to estimate the probability of a certain action being localised at a certain position and time in the video. These probabilities are consolidated using a Hough voting scheme to yield the final classification result.

Title: Generation of Natural Traffic Sign Images Using Domain Translation with Cycle-Consistent Generative Adversarial Networks

Advisors: Jun.-Prof. Sebastian Houben, Daniela Horn M.A. M.Sc.

This thesis deals with the generation of novel traffic sign images with the goal of supplementing the German Traffic Sign Benchmark dataset. A custom two-step method for image generation is explored, which first derives prototype images from standardised traffic sign diagrams and then enhances theses prototypes into natural images using an unpaired image-to-image translation system. The translation architecture of choice is a recent invention by the name of cycle-consistent generative adversarial networks (CycleGANs), which combine the powerful generative adversarial framework with CNN-based image-to-image mappings.

Title: Toward Automotive Safety and Autonomy with Machine-Learning-Assisted Radar Perception

Advisors: Prof. Dr.-Ing. Anton Kummert, Prof. Dr.-Ing. Tobias Meisen