My research lies in the intersection of AI, Computer Vision, and Human-Centred Computing. In particular, I am interested in how we can develop general methods and tools that enable autonomous systems to abstract, understand, reason about, and learn from multimodal (human) Interactions, with the aim to assist humans in their everyday personal and professional tasks. From a technical point of view this includes for instance: -- Commonsense Abstractions of Space and Motion -- Declarative Reasoning and Learning about Visuospatial Dynamics -- Nurosymbolic Integration of AI and Vision -- Visual Perception (e.g., involving eye-tracking)