Deep Learning & Computer Vision Researcher
Semantic Segmentation • Robust Perception • Knowledge Distillation • Edge Deployment

I am a deep learning and computer vision researcher focused on making autonomous driving perception reliable under adverse conditions such as weather and illumination changes.
My work centers on semantic segmentation, spanning RGB-D representation learning, knowledge distillation, and deployment-oriented model design.

I go beyond offline benchmarks by integrating on-device inference into ROS2 and Nav2 pipelines and validating models in real system contexts.
I make design decisions based on measured edge-device constraints such as latency, FPS, and model size, and iterate through evidence-driven experiments.
In team projects, I have built end-to-end pipelines from data preparation and modeling to evaluation and real-device deployment, delivering service-ready prototypes.

Deep Learning / CV: Python, PyTorch, OpenCV
On-device Deployment: ONNX, TensorRT, OpenVINO, Android Studio, Kotlin
Robotics Integration: ROS2, Nav2, SLAM
Dev / MLOps: Linux, Git, Docker
Embedded Platforms: Jetson TX2, Jetson Orin Nano, Raspberry Pi 5, Android
Vision Datasets: CamVid, KITTI, IDD-Lite, WM-811K
Other Data / Tasks: ETD (Trajectory)

CACKD	VD-MKDF

ROS2 & VLM Autonomous Robot with VD-MKDF	Smart Appliance Multimodal AI	Wafer Defect Classification	CPS Factory Automation Digital Twin

News Big Data Hackathon	LG Aimers Season 8	ETRI Vehicle Trajectory Prediction Challenge	Hecto Vehicle Classification Competition

Image Processing	Deep Learning	Embedded Systems

Robotics	Digital Twin	Paper Reviews