Christian Potthast

Embedded Systems (United States), University of Southern California

Papers

Total Citations

199

H-Index

About

Christian Potthast is a leading researcher in active perception and robotic vision, focusing on how autonomous systems can intelligently gather visual data to understand their environments. His primary contributions lie in developing probabilistic frameworks for next-best-view (NBV) estimation, enabling robots to strategically position cameras to maximize information gain in cluttered, unstructured settings. His most influential work, "A probabilistic framework for next best view estimation in a cluttered environment" (2013, 153 citations), provides a foundational method for active object recognition, allowing robots to iteratively select optimal viewpoints to reduce uncertainty. Potthast further advanced this field by unifying online feature selection with view planning in "Active multi-view object recognition: A unifying view on online feature selection and view planning" (2016, 27 citations), demonstrating how robots can simultaneously decide which features to observe and where to look. His earlier research on eye-in-hand camera systems (2011, 10 citations) laid critical groundwork for high-dimensional robotic arm control in automatic data acquisition. Potthast’s work has significantly impacted autonomous exploration, 3D reconstruction, and robotic manipulation, establishing him as a key figure in active vision and sensor planning for intelligent systems.