한밭대학교인공지능소프트웨어학과

VIR Lab
Visual Intelligence & Reliability Lab, VIR Lab

연구실 소개

VIR 연구실은 시각 지능(Visual Intelligence)과 신뢰할 수 있는 인공지능(Robust & Reliable AI)을 핵심 주제로 연구하는 연구실입니다. 컴퓨터 비전(Computer Vision)과 딥러닝(Deep Learning)을 기반으로 영상·이미지 등 시각 데이터를 이해하고 처리하는 시각 지능 기술을 연구하며, 동시에 실제 환경에서 안정적으로 동작하고 믿고 사용할 수 있는 인공지능을 만드는 것을 목표로 합니다.

지도교수 : 김영동

논문

• Seunghee Koh, Sunghyun Baek, Youngdong Kim, Junmo Kim, "Forget What Matters, Keep the Rest: Selective Unlearning of Informative Tokens", Annual Meeting of the Association for Computational Linguistics (ACL), 2026
• Jongin Lim, Youngdong Kim, Byungjai Kim, Chanho Ahn, Jinwoo Shin, Eunho Yang, Seungju Han, "BiasAdv: Bias-Adversarial Augmentation for Model Debiasing", IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023
• Youngdong Kim, Juseung Yun, Hyounguk Shon, Junmo Kim, "Joint Negative and Positive Learning for Noisy Labels", IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021
• Youngdong Kim, Junho Yim, Juseung Yun, Junmo Kim, "NLNL: Negative Learning for Noisy Labels", IEEE/CVF International Conference on Computer Vision (ICCV), 2019
• Juseung Yun, Youngdong Kim, Junho Yim, Hansang Cho, Junmo Kim, "Learning from Imbalanced and Hierarchically Dual-Labeled Data", IEEE International Conference on Consumer Electronics-Asia (ICCE-Asia), 2018
• Jaemyung Lee, Sihaeng Lee, Youngdong Kim, Janghyeon Lee, Junmo Kim, "Refine Pedestrian Detections by Referring to Features in Different Ways", IEEE Intelligent Vehicles Symposium (IV), 2017
• Donggyu Ju, Youngdong Kim, Junmo Kim, "Identity Preserving Residual-Spatial Transformer Network", International Conference on Robot Intelligence Technology and Applications (RiTA), 2017

특허

• METHOD AND DEVICE WITH IMAGE-DIFFERENCE REDUCTION PREPROCESSING (US 2025/0045884 A1), Patent Pending
• METHOD AND ELECTRONIC DEVICE WITH ADVERSARIAL DATA AUGMENTATION (US 2024/0152764 A1), Patent Pending
• METHOD AND DEVICE WITH DEFECT DETECTION (US 2024/0153070 A1), Patent Pending