I'm a 4th year PhD student at Chalmers working on fundamental research in computer vision and autonomous systems. My interests include image and video analysis, e.g., semantic segmentation and neural network architectures dedicated to video processing, learning based sensor fusion strategies, e.g., neural network architectures designed to process the input of multiple sensors, and different strategies to train the neural networks used for these problems, e.g., unsupervised domain adaptation (UDA) and semi-supervised learning (SSL). In my first year, I studied UDA semantic segmentation, wherein a neural network makes semantic segmentation predictions for images provided by a single camera. In my second and third year, I developed training methods for multi-view object detection using a neural network capable of taking multiple images as input. The findings were central to my Licentiate thesis, defended on October 7, 2025. In my fourth year (the current project), I will investigate principled methods for making multi-view neural networks invariant to the camera poses and calibration. Such invariance would constitute an important step to widespread deployment of multi-view neural networks as it allows them to generalize to any camera setup without additional training.