Education

– KAIST (Ph.D in Mechanical Engineering, 2020.8~)

– KAIST (M.S. in Mechanical Engineering, 2018.8~2020.8)

– UIUC (B.S. in Mechanical Engineering, Minor in Computer Science, 2018.5)

Research

– Omni-directional Image Object Detection/Recognition

-Semantic Segmentation

Recent Selected Publications 

• Towards Real-world Event-guided Low-light Video Enhancement and Deblurring
• Phase Concentration and Shortcut Suppression for Weakly Supervised Semantic Segmentation
• Diffusion-Guided Weakly Supervised Semantic Segmentation
• T4P: Test-Time Training of Trajectory Prediction via Masked Autoencoder and Actor-specific Token Memory
• High Accuracy Real-time Multi-Gas Identification by Batch-Uniform Gas Sensor Array and Deep Learning Algorithm
• Artificial Olfactory Neuron for an In-sensor Neuromorphic Nose 
• Ultra-Low-Power E-Noise System Based on Multi-Micro-LED-Integrated, Nanostructured Gas Sensors and Deep Learning
• Deep Learning-based Gas Identification by Time-variant Illumination of a Single Micro LED-embedded Gas Sensor
• 딥러닝 네트워크 기반 철판 연마율 인식
• Exploring Pixel-level Self-supervision for Weakly Supervised Semantic Segmentation
• EOMVS : Event-based Omnidirectional Multi-View Stereo
• 권혁준, 정재석, 윤국진, “중첩 영역을 이용한 대응점 기반 다시점 영상 색상 보정 방법의 강건화”, 제33회 영상처리 및 이해에 관한 워크샵, 2021
• SpherePHD: Applying CNNs on 360° Images with Non-Euclidean Spherical PolyHeDron Representation
• Yeon-Kun Lee, Jae-Seok Jeong, Won-June Cho, Jong-Seob Yun, Kuk-Jin Yoon, “SpherePHD: Applying CNNs on a Spherical PolyHeDron Representation of 360◦ Images”, Computer Vision and Pattern Recognition(CVPR), 2019
• 이연건, 정재석, 윤종섭, 조원준, 윤국진, “정 20면체 기반 360도 이미지 표현 및 CNN 적용 방법”, 제31회 영상처리 및 이해에 관한 워크샵, 2019

멀티모달인식 기반으로 일상생활환경의 다양한 물체를 파지, 조작하고 도구 활용 작업이 가능한 로봇 작업 제어 기술 개발