Jungtaek Kim

Pohang University of Science and Technology

Papers

Total Citations

H-Index

About

Jungtaek Kim is a researcher at the forefront of 3D vision and geometric deep learning, with a focus on combinatorial shape generation and robotic assembly. His most-cited work, "Combinatorial 3D Shape Generation via Sequential Assembly" (2020, 7 citations), tackles a fundamental challenge in robotics and computer vision: generating practical blueprints for constructing 3D shapes from volumetric primitives. By addressing the combinatorial explosion inherent in sequential assembly, Kim's research moves beyond greedy methods to enable more efficient and scalable shape generation—a critical step for applications in automated manufacturing and embodied AI. His contributions bridge the gap between geometric reasoning and practical robotics, offering insights that inspire further work in structured 3D representation learning. With a growing citation impact, Kim's work is recognized for its innovative approach to combining discrete optimization with continuous shape modeling. As a researcher advancing the intersection of geometry, learning, and robotics, he is shaping the future of how machines understand and build the physical world.