Scholar
Seungyeon Kim
Google Scholar ID: jNz4SZgAAAAJ
KTH Royal Institute of Technology
Citations & Impact
Citations
123
H-index
6
i10-index
4
Publications
13
Co-authors
8
list available
Contact
Publications
Resume (English only)
Academic Achievements
- Publications: - ScrewSplat: An End-to-End Method for Articulated Object Recognition (CoRL 2025) - DreamGrasp: Zero-Shot 3D Multi-Object Reconstruction from Partial-View Images for Robotic Manipulation (arXiv 2025) - Motion Manifold Flow Primitives for Task-Conditioned Trajectory Generation under Complex Task-Motion Dependencies (RA-L 2025) - Diverse Policy Learning via Random Obstacle Deployment for Zero-Shot Adaptation (RA-L 2025) - T2SQNet: A Recognition Model for Manipulating Partially Observed Transparent Tableware Objects (CoRL 2024) - Leveraging 3D Reconstruction for Mechanical Search on Cluttered Shelves (CoRL 2023) - Equivariant Motion Manifold Primitives (CoRL 2023) - SE(2)-Equivariant Pushing Dynamics Models for Tabletop Object Manipulations (CoRL 2022)
Research Experience
- Currently a postdoctoral researcher at the GeoRob Lab in EECS at KTH Royal Institute of Technology, advised by Noémie Jaquier; Previously a postdoctoral researcher at the Robotics Laboratory at Seoul National University, advised by Frank C. Park.
Education
- Ph.D.: Seoul National University, advised by Frank C. Park; M.S. and B.S.: Seoul National University, Mechanical Engineering (with a minor in Economics).
Background
- Research Interests: Developing practical solutions for intelligent robots that are adaptive and generalizable to unknown arbitrary environments. Professional Field: Mechanical Engineering (with a minor in Economics). Brief Introduction: Focused on leveraging inductive biases to enhance the performance of robots in real-world scenarios with limited data, while also adapting to various downstream manipulation tasks.
Miscellany
- Personal website provides links to email, CV, Google Scholar, Github, and YouTube.
