Practical Internship / Thesis: Spatial Reasoning for Partially Observed Industrial Scenes

This role involves researching methods for spatial reasoning in industrial image processing, focusing on scenarios where objects are obscured or sensors have limited coverage. The position requires evaluating and prototyping approaches using generative models, vision-language models, or probabilistic methods to draw conclusions from incomplete visual data. The goal is to assess the suitability of selected approaches for real-world industrial applications through a proof-of-concept prototype.