Paper Accepted at ICRA 2026, Top Robotics Conference / Student Kim Junho (Electrical Engineering, Class of 2019)
- 26.03.06 / 홍유민
Professor Lee Seong Won's research team from the Department of Electronic Engineering at Kookmin University (President Jeong Seung Ryul) has achieved the distinction of having a paper accepted for publication at the IEEE International Conference on Robotics and Automation (ICRA 2026), the world's most prestigious international academic conference in the field of robotics. This paper holds particular significance as it was conducted with undergraduate student Kim Jun-ho (Class of 2019, Department of Electronic Engineering) serving as the lead author.
ICRA is one of the most influential top-tier international conferences in robotics, autonomous driving, and AI-based intelligent systems, serving as a platform where leading universities and global research institutions present their core research achievements. Given its rigorous review standards and high competition rate, having an undergraduate student serve as the lead author on a paper accepted for publication is considered an exceptional achievement.
The accepted paper, titled “VG3T: Visual Geometry Grounded Gaussian Transformer,” proposes a novel feed-forward network architecture that directly predicts 3D semantic occupancy based on multi-view information. The research team aimed to solve the problem of precise spatial understanding in complex 3D environments by designing a framework that effectively integrates visual geometric information and semantic information using 3D Gaussian representations.
Notably, the proposed method achieved performance surpassing existing state-of-the-art techniques on the nuScenes dataset, a representative public benchmark in the autonomous driving field. Furthermore, by using 46% fewer primitives compared to existing methods, it significantly improved computational efficiency, demonstrating scalability and practicality for real-world autonomous driving and robotics systems.
This research was conducted under the guidance of Professor Lee Seong Won's research team at Kookmin University's CILAB (Computer Intelligence Lab), with undergraduate student Kim Junho playing a leading role in the entire process from research design to implementation and experimentation. This achievement serves as a case study demonstrating that world-class robotics and artificial intelligence research can be conducted even at the undergraduate level, and is evaluated as a result that externally validates the university's research competitiveness.
Meanwhile, this paper is scheduled to be officially presented at IEEE ICRA 2026, to be held in Vienna, Austria.
|
This content is translated from Korean to English using the AI translation service DeepL and may contain translation errors such as jargon/pronouns. If you find any, please send your feedback to kookminpr@kookmin.ac.kr so we can correct them.
|
|
Paper Accepted at ICRA 2026, Top Robotics Conference / Student Kim Junho (Electrical Engineering, Class of 2019) |
||||
|---|---|---|---|---|
|
2026-03-06
49
Professor Lee Seong Won's research team from the Department of Electronic Engineering at Kookmin University (President Jeong Seung Ryul) has achieved the distinction of having a paper accepted for publication at the IEEE International Conference on Robotics and Automation (ICRA 2026), the world's most prestigious international academic conference in the field of robotics. This paper holds particular significance as it was conducted with undergraduate student Kim Jun-ho (Class of 2019, Department of Electronic Engineering) serving as the lead author.
ICRA is one of the most influential top-tier international conferences in robotics, autonomous driving, and AI-based intelligent systems, serving as a platform where leading universities and global research institutions present their core research achievements. Given its rigorous review standards and high competition rate, having an undergraduate student serve as the lead author on a paper accepted for publication is considered an exceptional achievement.
The accepted paper, titled “VG3T: Visual Geometry Grounded Gaussian Transformer,” proposes a novel feed-forward network architecture that directly predicts 3D semantic occupancy based on multi-view information. The research team aimed to solve the problem of precise spatial understanding in complex 3D environments by designing a framework that effectively integrates visual geometric information and semantic information using 3D Gaussian representations.
Notably, the proposed method achieved performance surpassing existing state-of-the-art techniques on the nuScenes dataset, a representative public benchmark in the autonomous driving field. Furthermore, by using 46% fewer primitives compared to existing methods, it significantly improved computational efficiency, demonstrating scalability and practicality for real-world autonomous driving and robotics systems.
This research was conducted under the guidance of Professor Lee Seong Won's research team at Kookmin University's CILAB (Computer Intelligence Lab), with undergraduate student Kim Junho playing a leading role in the entire process from research design to implementation and experimentation. This achievement serves as a case study demonstrating that world-class robotics and artificial intelligence research can be conducted even at the undergraduate level, and is evaluated as a result that externally validates the university's research competitiveness.
Meanwhile, this paper is scheduled to be officially presented at IEEE ICRA 2026, to be held in Vienna, Austria.
|
||||
