Jihun Ham

Papers

Total Citations

H-Index

About

Jihun Ham is a leading researcher in robotics and machine perception, with a focus on nonlinear dimensionality reduction and multi-modal sensing for autonomous systems. His pioneering work on learning nonlinear appearance manifolds has fundamentally advanced robot localization, enabling robots to infer their pose from high-dimensional panoramic images by exploiting the underlying low-dimensional structure of visual data. This approach, detailed in his highly cited 2005 paper (27 citations), demonstrated how local geometric methods can extract meaningful spatial information from complex, high-dimensional sensory inputs, a contribution that remains influential in the fields of visual SLAM and manifold learning. Ham further expanded the boundaries of cooperative robotics through his innovative use of audible acoustic sensing for relative robot localization. His 2005 work (20 citations) introduced a method for teams of mobile robots to estimate each other's relative poses using only sound—specifically, time-of-arrival measurements from specially designed acoustic signals captured on stereo microphones. This research laid critical groundwork for low-cost, robust multi-robot coordination in environments where visual or laser-based sensing may fail. With a career marked by these foundational contributions to nonlinear manifold learning and acoustic-based perception, Jihun Ham’s work continues to inspire new generations of researchers tackling the challenges of autonomous navigation and collaborative robotics.