My bachelor Thesis: Scale estimation for online Visual Object Tracking based on convolutional features.
For the below projects, I provide a brief write-up of what the purpose of the project is, what I learned while working on the project and what the key outcome is. Contrary to the blog-posts, the goal behind these write-ups is not to fully explain the implementation or underlying principle, but rather to illustrate parts of my skill-set. I don't explicitly list all my projects here, as some are featured either elsewhere on this site or somewhere on the internet already.
Animated visualization of neural activity in different neural networks, made with Python.
Using an Oculus Rift alongside a Leap Motion, we investigated interaction techniques for virtual hands.
Study visit at the Intelligent Robotics group in Umeå, Sweden: A comprehensive interface for Softbanks Pepper robot.
Write-up coming (maybe) soon...:
Encoder/Decoder for saliency prediction
Prediction of salient image regions with an Encoder/Decoder CNN architecture.
Malicious ad classification
Classification of advertisement creatives into good or malicious based on metadata.
Unity Strategy Game
Coursework for the Interactive Game Development master level course at UHH.