Miles Freeman

Papers

Total Citations

H-Index

About

Miles Freeman is an emerging researcher specializing in robot perception, articulated object understanding, and kinematic reasoning. His most notable work, "Learning to Infer Kinematic Hierarchies for Novel Object Instances" (2022), addresses a fundamental challenge in robotic manipulation: enabling machines to understand the structure and motion of objects they have never encountered before. This research makes a significant leap beyond prior approaches by inferring *complete* kinematic hierarchies — identifying an object's parts, their possible movements, and how those movements are coupled — without relying on predefined templates or category-specific assumptions. This capability is critical for deploying robots in real-world environments filled with diverse, unfamiliar objects. With 9 citations since its publication, the work has already begun attracting attention within the robotics and computer vision communities, a promising trajectory for a relatively recent contribution. Freeman's research sits at the intersection of structured perception and physical reasoning, tackling problems that are essential for the next generation of intelligent robotic systems capable of interacting flexibly and safely with the physical world.