Jemo Maeng

Gwangju Institute of Science and Technology (GIST) Advisor: Prof. Kyoobin Lee

Ph.D. Course in Department of AI

Gwangju Institute of Science and Technology (GIST) Advisor: Prof. Kyoobin Lee

M.S. in Department of AI

Harbin Institute of Technology (HIT), China

B.S. in Computer Science and Technology

Development of Reconnaissance Robots Having Amphibious Mobility

Institute of Civil-Military Technology Cooperation (ICMTC), South Korea  ·  Grant No. 22-CM-GU-08

TL;DRLeading the perception stack for a robot that crosses between water and land, where the sensor domain shifts abruptly at the shoreline. Targets: 5 object classes at 65% detection and 90% recognition rate.

Acquisition and Labelling of Robot Manipulation Task Data in Laboratory Environments

Korea Institute of Robotics & Technology Convergence (KIRO), South Korea

TL;DRLeading the design and collection of a labelled manipulation dataset — trajectories, contacts and object states — for learning-based robot manipulation.

Development of Tactile Intelligence in Robotic Hands Based on Tactile Data Learning for Manipulating Irregular Multiple Types of Objects

Ministry of Trade, Industry and Energy (MOTIE), South Korea  ·  Technology Innovation Program No. 00442029

TL;DRLearning a tactile representation that lets a multi-fingered hand grasp and reorient unseen, irregular objects without a prior CAD model. Targets: 85% peg-in-hole success over 5 object types, 90% object recognition.

PoC of 3D Model-based Object Recognition and AR Visualisation with an RGB-D Camera

Hyundai, South Korea

TL;DRReal-time registration of observed RGB-D parts against reference CAD models, driving an AR overlay for on-line industrial part identification and pose estimation.

Development of Micro Intelligent Drone for Autonomous Mission in Communication/GPS-denied Environments

Institute of Civil-Military Technology Cooperation (ICMTC), South Korea  ·  Grant No. 23-CM-AI-08

TL;DRRGB-IR-LiDAR sensor-fusion object detection from drone views, plus construction of the multi-modal database behind it. Targets: ≥0.85 mAP@0.5 on targets as small as 30×30 px, holding ≥0.65 mAP under image degradation.

Hair and Dust Recognition for Robot Vacuum Cleaners

LG Electronics, South Korea

TL;DROn-device segmentation of hair and dust for a domestic robot vacuum, constrained to ≤5M parameters — ≥85% pixel accuracy at ≤10% false-positive rate, trained on a purpose-built 10K-image dataset.

Test-time Domain Adaptation for Mobile Robot Operation under Adverse Conditions such as Varying Weather

LG Electronics, South Korea

TL;DRLed a two-step continual test-time adaptation method that keeps a mobile robot's perception working as the input distribution drifts with the weather — 14% mean corruption error on CIFAR-10-C.

Code

Development of Swarm USV Autonomous Navigation Algorithm

Agency for Defense Development (ADD), South Korea

TL;DRMaritime object detection and situational awareness feeding the navigation stack of a swarm of unmanned surface vehicles.

Room Layout Detection for Estimating the Home Environment

LG Electronics, South Korea

TL;DRA ceiling segmentation model and dataset for robust floor-plan generation under heavy occlusion, at 85% layout pixel accuracy — the work behind SMRNet.

Deep Reinforcement Learning-based Human-Robot Collaboration for Unstructured Tasks such as Assembly

Ministry of Trade, Industry and Energy (MOTIE), South Korea

TL;DRClosing the simulation-to-reality gap so that policies trained in simulation transfer to a real collaborative assembly cell.

Electrical Facility Safety Management through Life-cycle History Management

Korea Institute of Energy Technology Evaluation and Planning (KETEP), South Korea

TL;DRDeep-learning detection and type classification of corona discharge for automated cable safety inspection — later published at IAS 2023 and awarded the MOTIE Minister's Award.

Applying Artificial Intelligence to Underground Facility Exploration

KEPCO KDN, South Korea

TL;DRObject detection inside low-light underground power tunnels, made robust at the feature level via patch-level cycle consistency rather than by enhancing the image — light enough to run on-board a robot with no comms. Built the cart-mounted multi-camera capture rig and authored the contract report.

Multimodal Semantic Segmentation Challenge, 4th Maritime Computer Vision Workshop

The 43rd IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2026)

🥇 First Prize

The 3rd Electrical Safety Excellent Paper Competition MOTIE

Minister's Award, Ministry of Trade, Industry and Energy (MOTIE)

Teaching Assistant GIST

Deep Learning (RT5101, Spring)

Research Intern GIST

Artificial Intelligence Laboratory (AILAB)