Heewon Kim
Assistant Professor
Global School of Media
Graduate School of Metaverse
Graduate School of Security
College of IT, Soongsil University
E-mail: hwkim (at) ssu.ac.kr
Research Interest:
Computer Vision, Deep Learning, AR/VR, Robot, Security, Media Art
Google Scholar Profile (Citations: 8000+)
Lab Homepage
Biography
Heewon Kim received the BS and PhD degrees in the Department of Electrical and Computer Engineering at Seoul National University, Korea, in 2014 and 2023. Currently, he is with the Global School of Media, College of IT, Soongsil University, Korea as an assistant professor. He was a Lieutenant of the Republic of Korea Army from 2014 to 2016, a Research Intern at Qualcomm in 2018, and a Perception Intern at NVIDIA in 2022.
His research interests are in the broad areas of Computer Vision and Machine Learning, including low-level vision, video analysis, 3D modeling and rendering, human pose and shape estimation, and robotics. He is a co-author of more than 20 journal and conference papers. He developed a very deep and high-performing network, EDSR (Enhanced Deep Super-Resolution), and won the first NTIRE2017 Single Image Super-Resolution (SR) Challenge in all categories. EDSR became a standard benchmark algorithm in SR. Recently, Prof. Kim has been working on video processing and 3D modeling/rendering problems. His works demonstrate outstanding impact on the research community with a total of 8,271 citations and an h-index of 13 on Google Scholar (as of January 18, 2024).
He received several awards, in particular, the Sang-Uk Lee prize in KCCV 2022 which is for KCCV papers from five years ago that have made a significant impact on computer vision research, Outstanding Reviewer Award in ECCV 2020, Runner-Up Award in AIM 2019 Challenge on Video Temporal Super-Resolution, and Winner Award and Best Paper Award in NTIRE 2017 Challenge Track.
He served as a reviewer more than 10 times in refereed journals and conferences including TPAMI, CVPR, ECCV, ICCV, and NeurIPS.
Education
Seoul National University
Ph.D. in Electrical and Computer Engineering, 03/2017 ~ 02/2023
Advisor: Kyoung Mu Lee and Bohyung Han
B.S. in Electrical and Computer Engineering, 03/2008 ~ 02/2014
B.S. in Management Of Technology, 03/2011 ~ 02/2014
Reserve Officers' Training Corps (ROTC), 03/2010 ~ 02/2014
Humboldt-Universität zu Berlin, 09/2012 ~ 06/2013
Visiting Student
Seoul Science High School, 03/2006 ~ 02/2008
Honors and Awards
2024 혁신한국인 & 파워코리아 대상, 교육/연구 부분, 스포츠서울
3rd place in ARNOLD Challenge at CVPR 2024 Embodied AI Workshop
Distinguished Ph.D. Dissertation Award in Seoul National University 2023
Sang-Uk Lee prize in KCCV 2022
KCCV papers from five years ago that have made a significant impact on computer vision research
Bronze prize ($5,000) in 27th Humantech paper award (2021)
Silver prize ($500) in 33rd IPIU (2021)
Outstanding reviewer in ECCV 2020
Runner-up award (2nd place) in AIM 2019 challenge on video temporal super-resolution
Excellent presentation award in VTT 2018 research highlights, AI governmental project
Best paper award in NTIRE 2017 workshop challenge track
Winner award in NTIRE 2017 workshop challenge track
Professional Experience
[2022. 6 - 2022. 9] Perception intern, Autonomous Driving Team at NVIDIA (worked with Heeseok Lee, Junghyun Kwon, and Sangmin Oh)
[2018. 3 - 2018. 9] Research intern, Autonomous Driving Team at Qualcomm (worked with Heesoo Myeong and Duck Hoon Kim)
[2014. 3 - 2016. 6] HR Officer (Battalion S1), Lieutenant of Republic of Korea Army
Academic Services
Computer vision conference reviewer: CVPR, ECCV, ICCV, BMVC, ACCV, WACV
Machine learning conference reviewer: ICLR, NeurIPS, AAAI
Journal reviewer: TPAMI, TIP, TNNLS, TCSVT, SPL
Emergency reviewer: ICCV 2021, CVPR 2021, CVPR 2020, ECCV 2020, ECCV 2024
Academic Activities
[2024.04 - present] HCI(한국HCI학회) 프로그램 위원장-기술
[2024.03 - 2024.08] KCCV(한국컴퓨터비전학술대회) 프로그램 위원(Industry Chair)
[2024.03 - 2024.05] ASK(한국정보처리학회) 학술부위원장
Co-organizer: Mobile AI workshop and challenge at CVPR 2021
Invited Talks
[2024.08.01] 홍익대학교, 영상커뮤니케이션대학원
NeRF의 개념과 최신 연구 트렌드
[2024.05.24] 한국정보처리학회 ASK, 신진연구자 세션
Adaptive Deep Image Signal Processor for Practical Applications
Learning Controllable ISP for Image Enhancement
[2024.05.09] 서울대학교, 식품영양 커뮤니케이션 세미나
인공지능의 기초와 식품영양 산업을 위한 혁신과 전망
[2024.02.01] IPIU (영상처리 및 이해에 관한 워크샵), 신진연구자 세션
Adaptive Deep Image Signal Processor for Practical Applications
NERDS: A General Framework to Train Camera Denoisers from Raw-RGB Noisy Image Pairs
[2023.09.26] Kyung Hee University Medical Center (경희의료원)
AI가 주는 새로운 가능성: 의료 분야에서 생성형 AI의 역할과 도전
[2023.01.30] LG AI Research
Searching for Controllable Image Restoration Networks
[2022.09-11] Fast Campus
33개 프로젝트로 완성하는 컴퓨터비전 딥러닝 심화 과정
[2019.08.13] NAVER LABS
Task-Aware Image Downscaling
[2019.01.17] KECFT (한국미래기술교육연구원)
딥러닝 기반 초해상도 영상복원 기술과 적용 방안
[2018-2021] NEO Convergence
Recent Trends in Image Restoration and Generative AI
[2018.11.02] VTT Research Highlights
Information Hiding within Images Using Neural Networks
[2017.9] NAVER
Enhanced Deep Residual Networks for Single Image Super-Resolution
YouTube view: 4,000+ (link: https://youtu.be/OMIqkn2DCUk)
Research Project
[2022. 5 - 2023. 2] Deep Camera Denoisers from Raw-RGB Noisy Image Pairs with Samsung Advanced Institute of Technology
[2021. 5 - 2022. 4] Tunable Image Signal Processor for Image Enhancement with Samsung Advanced Institute of Technology
[2020. 4 - 2021. 3] Searching for Controllable Image Restoration Networks with Samsung Advanced Institute of Technology
[2019. 4 - 2020. 5] Continuous domain generalization for object recognition with Samsung Advanced Institute of Technology
[2018. 9 - 2019. 4] Night pedestrian synthesis using image-to-image translation with Samsung Advanced Institute of Technology
[2018. 9 - 2019. 4] Unsupervised image super-resolution using continual learning with Samsung Research
Selected Publications
Learning to Learn Task-Adaptive Hyperparameters for Few-Shot Learning
Sungyong Baik, Myungsub Choi, Janghoon Choi, Heewon Kim, and Kyoung Mu Lee.
TPAMI, Accepted 2023 (IF: 24.314)
NERDS: A General Framework to Train Camera Denoisers from Raw-RGB Noisy Image Pairs
Heewon Kim and Kyoung Mu Lee.
ICLR 2023
Fine-Grained Neural Architecture Search for Image Super-Resolution
Heewon Kim, Seokil Hong, Bohyung Han, Heesoo Myeong, and Kyoung Mu Lee.
JVCI, Accepted 2022
Machine Learning-Based Predictive Modeling of Depression in Hypertensive Populations
Chiyoung LeeI and Heewon Kim
PLOS ONE, Accepted 2022
CADyQ: Content-Aware Dynamic Quantization for Image Super-Resolution
Cheeun Hong, Sungyong Baik, Heewon Kim, Seungjun Nah, and Kyoung Mu Lee.
ECCV 2022.
Attentive Fine-Grained Structured Sparsity for Image Restoration
Junghun Oh, Heewon Kim, Seungjun Nah, Cheeun Hong, Jonghyun Choi, and Kyoung Mu Lee.
CVPR 2022.
DAQ: Channel-Wise Distribution-Aware Quantization for Deep Image Super-Resolution Networks
Cheeun Hong*, Heewon Kim*, Sungyong Baik, Junghun Oh, and Kyoung Mu Lee.
(* equal contribution)
WACV 2022.
Batch Normalization Tells You Which Filter is Important
Junghun Oh, Heewon Kim, Sungyong Baik, Cheeun Hong, and Kyoung Mu Lee.
WACV 2022.
Channel Attention is All You Need for Video Frame Interpolation
Myungsub Choi, Heewon Kim, Bohyung Han, Ning Xu, and Kyoung Mu Lee.
AAAI 2020 (Spotlight presentation), 2nd place in AIM 2019 ICCV Workshop - Video Temporal Super-Resolution Challenge.
Fine-Grained Neural Architecture Search
Heewon Kim, Seokil Hong, Bohyung Han, Heesoo Myeong, and Kyoung Mu Lee.
arXiv 2019 (Weekly paper at Medium)
Enhanced Deep Residual Networks for Single Image Super-Resolution
Bee Lim, Sanghyun Son, Heewon Kim, Seungjun Nah, and Kyoung Mu Lee.
CVPRW 2017 (Best paper award), 1st place in NTIRE 2017 Workshop and Challenge, 4000+ citations.