Obin Kwon

square_profile.png

I'm a postdoc in KIMLAB (Kinetic Intelligent Machine LAB), University of Illinois Urbana-Champaign. I recently got Ph.D. in ECE, Seoul National University, Korea. I was in Robot Learning Labatory (RLLAB) and was supervised by Songhwai Oh . Before the PhD course, I received a B.S. degree in Electrical and Computer Engineering from SNU in 2018.
Previously I was in internship at Naver LABS in Jan.-June 2023. I also worked as a robotics researcher in Sequor Robotics, Inc. from Oct 2023 to Jun 2024.

Research Keywords : Embodied AI, Robot Learning 🤖 with Computer Vision 👀, Machine Learning🧠




My research interests are lie in translating sensory information into effective robot action policies . Previously, I focused on developing novel scene representations from vision signals and creating robot policies that effectively use these representations to perform high-level tasks. Looking forward, I am enthusiastic about exploring a wide spectrum of new frontiers in robotics, encompassing multi-modal perception, human-robot interaction, and advanced decision-making processes.

News

Aug 01, 2024 🎓 I received the Distinguished Ph.D. Dissertation Award from ECE, Seoul National University! 🎓
Jul 08, 2024 I started my role as a postdoctoral research associate in KIMLAB@UIUC, working with Joohyung Kim.
Jun 30, 2024 Our paper “RNR-Nav: A Real-World Visual Navigation System Using Renderable Neural Radiance Maps” is accepted to IROS 2024, as oral presentation.
Jun 26, 2024 I successfully defended my Ph.D Thesis! :sparkles:
May 13, 2024 I attended ICRA 2024, presenting my paper “WayIL:Image-based Indoor Localization with Wayfinding Maps

Publications

  1. coming_soon.gif
    Spatially-Conditional 3D Furniture Generation Model for Indoor Scene Generation
    Jeongho Park, Obin Kwon, and Songhwai Oh
    In 2024 24th International Conference on Control, Automation and Systems (ICCAS), Oct 2024
    Scene Representation
    Scene Generation
  2. RNR-Nav: A Real-World Visual Navigation System Using Renderable Neural Radiance Maps
    Minsoo Kim, Obin Kwon, Howoong Jun, and Songhwai Oh
    In IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Oct 2024 (Oral Presentation)
    Scene Representation
    Visual Navigation
    Visual Localization
    NeRF
  3. WayIL:Image-based Indoor Localization with Wayfinding Maps
    Obin Kwon, Dongki Jung, Youngji Kim, Soohyun Ryu, Suyong Yeon, Songhwai Oh, and Donghwan Lee
    In 2024 International Conference on Robotics and Automation (ICRA), May 2024
    Scene Representation
    Visual Localization
  4. Renderable Neural Radiance Map for Visual Navigation
    Obin Kwon, Jeongho Park, and Songhwai Oh
    In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Jun 2023 (Highlight)
    Scene Representation
    Visual Navigation
    NeRF
  5. areq_overview.png
    Attention-Based Randomized Ensemble Multi-Agent Q-Learning
    Jeongho Park, Obin Kwon, and Songhwai Oh
    In 2023 23rd International Conference on Control, Automation and Systems (ICCAS), Jun 2023
    Multi-Agent
    Reinforcement Learning
  6. Topological Semantic Graph Memory for Image-Goal Navigation
    Nuri Kim, Obin Kwon, Hwiyeon Yoo, Yunho Choi, Jeongho Park, and Songhwai Oh
    In Proceedings of The 6th Conference on Robot Learning, 14–18 dec 2022 (Oral Presentation)
    Scene Representation
    Visual Navigation
    Memory Building
    Imitation Learning
    Reinforcement Learning
  7. Visually Grounding Language Instruction for History-Dependent Manipulation
    Hyemin Ahn, Obin Kwon, Kyungdo Kim, Jaeyeon Jeong, Howoong Jun, Hongjung Lee, Dongheui Lee, and Songhwai Oh
    In 2022 International Conference on Robotics and Automation (ICRA), 14–18 dec 2022
    Manipulation
    Memory Building
    Language Instruction
  8. Image-goal navigation algorithm using viewpoint estimation
    Obin Kwon, and Songhwai Oh
    In International Conference on Control, Automation and Systems (ICCAS), 14–18 dec 2021
    Visual Navigation
    Reinforcement Learning
  9. geometric_overview.png
    Geometric Understanding of Reward Function in Multi-Agent Visual Exploration
    Minyoung Hwang, Obin Kwon, and Songhwai Oh
    Oct 2021
    Multi-Agent
    Reinforcement Learning
    Visual Navigation
  10. Visual Graph Memory With Unsupervised Representation for Visual Navigation
    Obin Kwon, Nuri Kim, Yunho Choi, Hwiyeon Yoo, Jeongho Park, and Songhwai Oh
    In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Oct 2021
    Visual Navigation
    Scene Representation
    Memory Building
    Imitation Learning
    Reinforcement Learning
  11. ltm_overview.png
    Image-Goal Navigation Algorithm using Viewpoint Estimation
    Obin Kwon, and Songhwai Oh
    In 2021 21st International Conference on Control, Automation and Systems (ICCAS), Oct 2021
    Visual Navigation
    Memory Building
    Reinforcement Learning