I’m Seongwon Lee, a assistant professor at the School of Electrical Engineering at Kookmin University. I am currently the principal investigator of the Computational Intelligence Lab (CILAB). Previously, I was a researcher in the Visual Intelligence Laboratory at ETRI. Before that, I received my Ph.D. and B.S. degrees in Electrical and Electronic Engineering from Yonsei University in 2023 and 2016, respectively, advised by Professor Euntai Kim.

I am interested in building algorithms and machines that can understand the world and communicate with humans. My research focuses on learning deep networks that can solve a wide range of problems using a few general interfaces.

🎉 News

(Mar 2026) 💼 I have started a project on vision-based cargo and cargo label recognition on robots in logistics warehouse environments in collaboration with BStar Robotics.
(Mar 2026) 💼 I has been selected for the Outstanding Young Scientist Grant from the National Research Foundation of Korea (NRF) for the project, Vision-Language Navigation Agents via Cyclic Expansion of Web-Scale Video, Augmented Simulation, and the Real World.
(Feb 2026) 📝 Two papers have been accepted to CVPR 2026.
(Jan 2026) 📝 One paper has been accepted to ICRA 2026.
(Sep 2025) 🎤 Giving an invited talk about Neural Rendering: Creating Reality with AI @ Yonsei University.
(Sep 2025) 📝 One paper has been published in IEEE Transactions on Intelligent Vehicles (TIV) (IF: 14.0, JCR Top 2%).
(Jul 2025) 🎤 Giving an invited talk about Neural Rendering Using Vision and LiDAR and Its Application in Map Representation @ SOSLAB.
(Jul 2025) 🎤 Giving an invited talk about Recent Advances in Spatial AI: The Present and Future of Spatial Intelligence @ Gachon University.
(Jun 2025) 💼 I have started a project related to 3D Reconstruction in collaboration with KETI.
(Apr 2025) 💼 I have started a project related to Image Generation in collaboration with ETRI.
(Apr 2025) 🎤 Giving an invited talk about Neural Rendering: NeRF and 3D Gaussian Splatting @ KETI.
(Mar 2025) 🎤 Giving an invited talk about Match and Extend: Visual-Based Matching and 3D Reconstruction Technology @ ETRI.
(Feb 2025) 📝 One paper has been accepted to ICRA 2025.
(Nov 2024) 📝 One paper has been published in IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI) (IF: 20.6, JCR Top 1%).
(Oct 2024) 📝 One paper has been accepted to IEEE Access (IF: 3.4, JCR Top 40%).
(Aug 2024) 👨‍🏫 I am starting my assistant professorship in Kookmin University.
(Jul 2024) 📝 One paper has been accepted to ECCVW 2024.
(Jan 2024) 🎤 Giving an invited talk about recent computer vision research trends @ Yonsei University. [slide]
(Oct 2023) 🎤 Giving an invited talk about image retrieval @ Jeonbuk National University. [slide]
(Aug 2023) 📝 One paper has been accepted to BMVC 2023.
(Feb 2023) 📝 One paper has been accepted to CVPR 2023.
(Nov 2022) 🏆 I won the Qualcomm Innovation Fellowship 2022.
(Oct 2022) 🏆 I won the Excellence Vision Research Award at the 1st Yonsei AI Workshop.
(Mar 2022) 🏆 A paper titled “Correlation Verification for Image Retrieval” was selected for Oral presentation in CVPR 2022.
(Mar 2022) 📝 Two papers have been accepted to CVPR 2022.
(Jul 2021) 📝 One paper has been accepted to ICCV 2021.

📚 Selected Publications

Google Scholar full list, * denotes corresponding author

Rethinking Pose Refinement in 3D Gaussian Splatting under Pose Prior and Geometric Uncertainty

Mangyu Kong, Seongwon Lee*, Euntai Kim*

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Jun 2026

[arXiv]

HypeVPR: Exploring Hyperbolic Space for Perspective to Equirectangular Visual Place Recognition

Suhan Woo, Seongwon Lee, Jinwoo Jang, Euntai Kim*

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Jun 2026

[arXiv]

VG3T: Visual Geometry Grounded Gaussian Transformer

Junho Kim, Seongwon Lee*

IEEE International Conference on Robotics and Automation (ICRA), Jun 2026

[arXiv]

Location-Aware Transformer Network for Bird's Eye View Semantic Segmentation

Suhan Woo, Minseong Park, Youngjo Lee, Seongwon Lee, Euntai Kim*

Transactions on Intelligent Vehicles (TIV), Sep 2025

[paper]

Fast Global Localization on Neural Radiance Field

Mangyu Kong, Jaewon Lee, Seongwon Lee*, Euntai Kim*

IEEE International Conference on Robotics and Automation (ICRA), May 2025

[arXiv]

Correlation Verification for Image Retrieval and Its Memory Footprint Optimization

Seongwon Lee, Hongje Seong, Suhyeon Lee, Euntai Kim*

Transactions on Pattern Analysis and Machine Intelligence (TPAMI), Mar 2025

[paper]

Language-only Efficient Prompt Learning for Zero-shot Composed Image Retrieval

Seongwon Lee, Yong-Ju Lee*

European Conference on Computer Vision Workshop (ECCVW (Green-FOMO Workshop)), Oct 2024

RoomNeRF: Representing Empty Room as Neural Radiance Fields for View Synthesis

Mangyu Kong, Seongwon Lee, Euntai Kim*

The British Machine Vision Conference (BMVC), Nov 2023

[page] [paper] [supp] [poster] [video]

Revisiting Self-Similarity: Structural Embedding for Image Retrieval

Seongwon Lee, Suhyeon Lee, Hongje Seong, Euntai Kim*

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Jun 2023

[page] [paper] [supp] [poster] [video]

Correlation Verification for Image Retrieval

Seongwon Lee, Hongje Seong, Suhyeon Lee, Euntai Kim*

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Jun 2022

Selected as an Oral Presentation (Oral Presentation Acceptance Rate: 342/8161, 4.2%)

Winner at Qualcomm Innovation Fellowship 2022

Excellence Vision Research Award at 1st Yonsei AI Workshop

Encouragement Award at 2023-1 Graduate School of Yonsei University Thesis Award

[page] [paper] [supp] [poster] [video] [slide] [arXiv]

WildNet: Learning Domain Generalized Semantic Segmentation from the Wild

Suhyeon Lee, Hongje Seong, Seongwon Lee, Euntai Kim*

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Jun 2022

Finalist at Qualcomm Innovation Fellowship 2022

[page] [paper] [supp] [arXiv]

Hierarchical Memory Matching Network for Video Object Segmentation

Hongje Seong, Seoung Wug Oh, Joon-Young Lee, Seongwon Lee, Suhyeon Lee, Euntai Kim*

IEEE/CVF International Conference on Computer Vision (ICCV), Oct 2021

[page] [paper] [supp] [arXiv]

🏆 Honors and Awards

(Jul 2023) 2023-1 Graduate School of Yonsei University Thesis Award (2023)
- Encouragement Award - Correlation Verification for Image Retrieval (CVPR 2022)
(Nov 2022) Qualcomm Innovation Fellowship 2022
- Winner ($4,000) - Correlation Verification for Image Retrieval (CVPR 2022)
(Oct 2022) 1st Yonsei AI Workshop (2022)
- Excellence Vision Research Award - Correlation Verification for Image Retrieval (CVPR 2022)

💼 Projects

(Mar 2026 - Present) Vision-based Cargo and Cargo Label Recognition on Robots in Logistics Warehouse Environments
- Funding from BStar Robotics
(Mar 2026 - Present) Vision-Language Navigation Agents via Cyclic Expansion of Web-Scale Video, Augmented Simulation, and the Real World
- Funding from National Research Foundation of Korea (NRF)
(Apr 2025 - Nov 2025) Development of High-Fidelity Vehicle Representation via 3D Gaussian Splatting
- Funding from Korea Electronics Technology Institute (KETI)
(Jun 2025 - Nov 2025) Development of Brain Signal Based Image Generation
- Funding from Electronics and Telecommunications Research Institute (ETRI)
(Aug 2023 - Aug 2024) Development of AI Autonomy and Knowledge Enhancement for AI Agent Collaboration
- Funding from Information & Communications Technology Promotion (IITP)
(Apr 2022 - Dec 2022) Development of Deep Learning-Based Fine-Matching Technology using ToF Depth Sensor
- Funding from LG Electronics
(May 2021 - Dec 2021) Development of Deep Learning-Based Image Retrieval Technology using ToF Depth Sensor
- Funding from LG Electronics
(Apr 2020 - Feb 2021) Development of Front-Down SLAM Technology Based on ToF Depth Sensor
- Funding from LG Electronics
(Apr 2019 - Dec 2019) Development of Depth Map Generation Technology using Continuous Stereo Images
- Funding from Hyundai AutoEver
(Oct 2017 - Feb 2019) Development of Integration Module Technology for Localization and Autonomous Driving Control Based on Multi-modal Sensor for Wheel Drive Service Robot
- Funding from Ministry of Trade, Industry and Energy (MOTIE)
(Oct 2016 - Oct 2017) Development of Robot Autonomous Driving Technology using Laser Scanner
- Funding from Hitachi-LG Data Storage
(May 2016 - Apr 2020) Development of Robot Intelligence Technology for Mobility with Learning Capability toward Robust and Seamless Indoor and Outdoor Autonomous Navigation
- Funding from Ministry of Trade, Industry and Energy (MOTIE)

📝 Academic Services

Invited Talks
- (Sep 2025) Neural Rendering: Creating Reality with AI
  - School of Electrical and Electronic Engineering, Yonsei University
- (Jul 2025) Neural Rendering Using Vision and LiDAR and Its Application in Map Representation
  - SOSLAB
- (Jul 2025) Recent Advances in Computer Vision: Foundation Models, Vision-Language Models and 3D Neural Representations
  - School of Computing, Gachon University
- (Apr 2025) Neural Rendering: NeRF and 3D Gaussian Splatting
  - Korea Electronics Technology Institute (KETI)
- (Mar 2025) Match and Extend: Visual-Based Matching and 3D Reconstruction Technology
  - Electronics and Telecommunications Research Institute (ETRI)
- (Jan 2024) Recent Advances in Computer Vision: Foundation Models, Vision-Language Models and 3D Neural Representations [slide]
  - School of Electrical and Electronic Engineering, Yonsei University
- (Oct 2023) Introduction to Image Retrieval: Local Features & beyond [slide]
  - Department of Electronic Engineering, Jeonbuk National University
Reviewer
- Conference
  - IEEE/CVF Computer Vision and Pattern Recognition Conference (CVPR)
  - European Conference on Computer Vision (ECCV)
  - IEEE/CVF International Conference on Computer Vision (ICCV)
  - IEEE International Conference on Robotics and Automation (ICRA)
  - IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
  - Asian Conference on Computer Vision (ACCV)
- Journal
  - IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
  - IEEE Transactions on Automation Science and Engineering (T-ASE)
  - International Journal of Control, Automation and Systems (IJCAS)