Jongkeum Choi
Papers
1
Total Citations
8
H-Index
1
About
Jongkeum Choi is a rising researcher in computer vision and robotics, with a primary focus on 3D scene understanding and depth perception for challenging object categories. Their most notable contribution is the development of SAID-NeRF (Segmentation-AIDed NeRF), a pioneering framework that addresses the long-standing problem of acquiring accurate depth information for transparent objects using standard RGB-D cameras. This work, published in 2024 and already garnering 8 citations, introduces a novel approach that leverages Neural Radiance Fields (NeRF) guided by semantic segmentation to complete depth maps where traditional sensors fail. By tackling the inherent ambiguity of transparent surfaces—which typically cause depth sensors to produce erroneous or missing data—Choi's research bridges a critical gap between perception and manipulation in robotics. Their methodology demonstrates how combining implicit neural representations with explicit segmentation cues can yield robust depth estimates without relying on synthetic training data. This achievement holds significant promise for applications in autonomous manipulation, augmented reality, and industrial inspection, where reliable depth sensing of glass, plastic, and other transparent materials is essential. Choi's work represents an important step toward more capable and perceptive robotic systems operating in real-world environments.
Research Focus
Key Achievements
Top Papers
- 1