Minhyuk Sung

Adobe Systems (United States), Kookmin University

Papers

Total Citations

H-Index

About

Minhyuk Sung is a leading researcher in computer vision and robotics, with a primary focus on 3D understanding, shape analysis, and autonomous assembly. His most impactful work, "Learning 3D Part Assembly from a Single Image" (2020, 56 citations), introduces a groundbreaking approach that enables robots to infer the assembly of 3D objects from just one 2D image. This contribution directly addresses the critical challenge of task specification in autonomous assembly, bridging the gap between visual perception and robotic manipulation. By learning to decompose and reconstruct objects part-by-part, Sung’s method empowers robots to handle complex, unstructured assembly tasks without explicit programming, marking a significant leap toward general-purpose robotic intelligence. His earlier work on "Plane-dominant object reconstruction for robotic spatial augmented reality" (2011) laid foundational techniques for reconstructing planar environments, essential for projector-based augmented reality systems. Sung’s research has garnered substantial attention, with his assembly paper alone accumulating over 56 citations, reflecting its influence on both the computer vision and robotics communities. Through his innovative integration of learning-based 3D reasoning and robotic action, Sung is shaping the future of autonomous systems that can perceive, understand, and physically interact with the world.