Jakob Mittelberger

Papers

Total Citations

H-Index

About

Jakob Mittelberger is a researcher at the forefront of robotic perception and manipulation, with a primary focus on affordance detection—the ability of machines to understand how objects can be used or interacted with from visual data. His most notable contribution is the development of **Dynamic-Tree Capsule Networks**, a novel architecture that addresses a critical limitation of conventional convolutional neural networks: their inability to encode spatial hierarchies and parts-to-whole relationships. By introducing this approach, Mittelberger has advanced the field of autonomous robotics, enabling more nuanced and context-aware visual understanding. His 2022 paper on this topic has garnered early citations, reflecting its growing influence among researchers seeking to bridge the gap between perception and action. This work stands out for its theoretical elegance and practical promise, offering a pathway toward more robust robotic systems that can reason about object affordances in dynamic environments. Mittelberger’s research is particularly relevant for students and engineers working on embodied AI, computer vision, and intelligent robotics, as it tackles a foundational challenge in making machines interact with the world more like humans do.