Our Projects
[NRF] Intelligent Scene Coordinate Regression Model for Robust LiDAR Localization
Sep 2025 - 2026
This project develops an Intelligent Scene Coordinate Regression (SCR) framework for LiDAR localization that overcomes current issues of storage, privacy, and scalability by introducing dynamic object filtering.
[CMES] Robust Non-Rigid 3D Alignment for Thermally and Physically Deformed Objects
Jul 2025 - 2025
This project focuses on performing non-rigid registration for 3D objects deformed by heat or physical force by establishing correspondence matches between shapes and then applying a non-rigid ICP algorithm for final alignment.
[MOTIE] Development of Multimodal Image-Language Model and On-device Model Compression Technology for Industrial Application
Apr 2025 - Dec 2025
This project aims to develop Hyper-CLIP, a multimodal embedding framework that effectively aligns visual and textual semantics on hyperspectral imagery. Building upon this foundation, we further introduce Hyper-LLaVA, optimized for efficient deployment in real-world industrial environments.
[MOTIE] Interactive humanoid platform with transcendental sensory augmentation
Mar 2025 - Dec 2025
The goal of this study is to develop a next-generation humanoid platform capable of bidirectional communication using extrasensory information to autonomously execute human-taught commands and enhance human cognitive and sensory functions.
[NRF] Research Center for Precision Medicine Platform Based on Smart Hemo-Dynamic Index
Jun 2022 - Feb 2029
The aim of the project is to create a mixed-reality (MR) platform for calculating smart hemodynamics index in real-time to support patients with cardiovascular diseases.
[KEIT] Development of Industrial Intelligent Technology for Smart Factory
Jan 2023 - Dec 2026
The project is aimed towards developing an accurate algorithm for image-based anomaly detection algorithm that can be applied across a range of different products.
[KRIT] Space Signal Intelligence Payload Data Analysis Technology
Nov 2022 - Feb 2028
This project comprises developing an algorithm for type classification of data sent from geostationary satellites. It involves byte-data augmentation using transformers.
[Hyundai Motor] Development of a Few Shot-based AI Algorithm for Augmentation and Generation of Defect Data
Jan 2025 - Jun 2025
The project aims to develop an AI algorithm that can generate defect data for industrial applications using few-shot learning.
[NRF] Privacy-Preserving 3D Reconstruction and Camera Localization
Mar 2022 - Feb 2025
We are developing core algorithms for 3D reconstruction and camera localization that can address user's privacy concerns.
[NRF] Development of AI Technologies for Indoor and Outdoor Long-Range Voice/Audio/Video Detection
Jan 2023 - Feb 2025
The project is aimed towards developing an accurate algorithm for image-based anomaly detection algorithm that can be applied across a range of different products.
[Samsung Electronics] Continual Learning and Data Augmentation for Anomaly Detection
Mar 2022 - Feb 2025
The goal is to adopt a continual learning approach for data augmentation of defect patterns in semiconductor wafers.
[Hyundai NGV] AI for Design of Vehicles
Apr 2022 - Apr 2023
The goal is to design a network that can detect changes in vehicle design just by looking at the cross-section of rear air flow.
[Blue Asia] Violence detection
Jan 2022 - Feb 2022
Developed an algorithm for improving accuracy of violence detection in public transportation systems
[KIST] Sketch-based person recognition
Sep 2021 - Dec 2021
Developed a method for accurately recognizing people from their sketches, thereby reducing burdens for police officers.
[Bagel Labs] 3D human pose estimation
Sep 2021 - Dec 2021
Investigated in improving accuracy of real-time 3D human pose estimation with use of a commercial depth camera.
Restoring potteries from fragments
Jul 2018 - Dec 2020
Developed an efficient pipeline for reassembling potteries from their constituent fragments.
Advanced Integrated Intelligence for Identification (AIID)
Jul 2018 - Jul 2021
Developed face frontalization and face-masking pipelines for robust face recognition