Curriculum Vitae

Education

• M.S. in Electrical Engineering from Korea Advanced Institute of Science and Technology (KAIST) (2019)

• B.S. in Electrical Engineering from Korea University (2017)

Research Interests

• Deep Learning

• Computer Vision

Experiences

• 2019.03. - Electronics and Telecommunications Research Institute (ETRI) Researcher

• 2018.10. - 2019.01. Genietec  Technical Support

Patents

• "Method and apparatus for face recognition robust to alignment status of the face", US Patent US20220129689A1

• "Garbage dumper detection system and method using multi-characteristic information in surveillance camera environment", Korea Patent 10-2020-0041134

• "Voice Recognition Method using Multiple Channels", Korea Patent 10-2018-0144290

Publications

International

• Kimin Yun, Hyungil Kim, Kangmin Bae, and Jinyoung Moon, "Background memory-assisted zero-shot video object segmentation for unmanned aerial and ground vehicles", ETRI Journal, 2023

• Kangmin Bae, Hyung-Il Kim, Yongjin Kwon, and Jinyoung Moon, "Unsupervised Bidirectional Style Transfer Network using Local Feature Transform Module", Computer Vision and Pattern Recognition (CVPR) Workshop GCV, 2023

• Je-Seok Ham, Kangmin Bae and Jinyoung Moon, "MCIP: Multi-Stream Network for Pedestrian Crossing Intention Prediction", European Conference on Computer Vision (ECCV) Workshop AVVision, 2022

• Kangmin Bae*, Kimin Yun*, Jungchan Cho, and Yuseok Bae, "The Dataset and Baseline Models to Detect Human Postural States Robustly against Irregular Postures", International Conference on Advanced Video and Signal-based Surveillance (AVSS), 2021

• Kimin Yun, Hyungil Kim, Kangmin Bae, and Jongyoul Park, "Unsupervised Moving Object Detection through Background Models for PTZ Camera", International Conference on Pattern Recognition (ICPR), 2020

• Kangmin Bae*, Kimin Yun*, Hyungil Kim, Youngwan Lee, and Jongyoul Park, "Anti-Litter Surveillance based on Person Understanding via Multi-Task Learning", British Machine Vision Conference (BMVC), 2020

• Kang Min Bae, Minuk Ma, Hyunjun Jang, Minjeong Ju, Hyoungwoo Park, and Chang D. Yoo, "ImaGAN: Unsupervised Training of Conditional Joint CycleGAN for Transferring Style with Core Structures in Content Preserved", Asian Conference on Computer Vision (ACCV), 2018

• Jae Hyun Han*, Kang Min Bae*, Seong Kwang Hong*, Hyunsin Park, Jun-Hyuk Kwak, Hee Seung, Wang, Daniel Juhyung Joe, Jung Hwan Park, Young Hoon Jung, Shin Hur, Chang D. Yoo, and Keon Jae Lee, "Machine learning-based self-powered acoustic sensor for speaker recognition", Nano Energy, vol. 53, pp. 658-665, Nov. 2018

• Kang Min Bae, Luu Minh Tung, and Chang D. Yoo "Associative domain adaptation of depth estimation using synthetic indoor data to real indoor data", International Conference On Consumer Electronics (ICCE) Asia, 2018

Domestic

• 노승하, 배강민, 배유석, 이병대, "추적 정보를 활용한 비정상 상태 사람 인식 기법", 한국소프트웨어종합학술대회 (KSC), 2023

• 부원국, 배강민, 윤기민, 배유석, "멀티태스크 학습을 활용한 실세계 쓰러진 사람 탐지 기술 개발", 대한전자공학회 학술 대회 (하계), 2023

• 김진아, 송길재, 배강민, 윤기민, 배유석,  "학습 데이터셋의 적절한 조합과 모델 최적화로 사람 상태 이해 성능 개선 및 실세계 도메인 최적화", 한국컴퓨터종합학술대회 (KCC), 2022

• 송길재, 김진아, 배강민, 윤기민, 배유석, "효과적인 다중 카메라 배치 단위 추론과 이벤트 알람 전송을 위한 프레임워크", 한국컴퓨터종합학술대회 (KCC), 2022

• 오원석, 배강민, 배유석, "Attention 을 이용한 비식별화 이미지 dataset 복원을 위한 GAN 의 시각화 비교", 한국소프트웨어종합학술대회 (KSC), 2021

• 오원석, 배강민, 배유석, "RDID-GAN: 비식별화 이미지 데이터 복원을 통한 효과적인 학습데이터 생성", 정보과학회논문지 48 (12), pp. 1329-1334, 2021

• 조영주,  배강민,  박종열, "GAN 적대적 생성 신경망과 이미지 생성 및 변환 기술 동향", [ETRI] 전자통신동향분석, 35(4), pp. 91-102, 2020

• 배강민, 유창동, "FCN과 Spatial Attention을 이용한 Semantic Segmentation", 대한전자공학회 하계학술대회 2018

• 배강민, 유창동, "BEGAN을 이용한 LSUN과 MNIST 데이터 생성", 영상처리 및 이해에 관한 워크샵 (IPIU) 2018

• Hyo J. Han, Kangmin Bae, Chang D. Yoo, "Visual speech recognition from multi view to single view", 영상처리 및 이해에 관한 워크샵 (IPIU) 2018

• Oh Chul Kwon, Chang D. Yoo, Kangmin Bae, "Enhanced drowsy detection via simple idea", 영상처리 및 이해에 관한 워크샵 (IPIU) 2018