Hyeonseong Jeon
Personal Data
Research Interests
Deep Learning, Machine Learning, Computer Vision, Pattern Recognition, and Artificial Intelligence
Specific Research Interests:
Self-supervised learning with hyper-scale AI
Generative models (GAN, Diffusion models)
Multi-modal learning (text & image)
AI vision inspection & AutoML (easy-to-use AI)
Education
Mar. 2019 - Feb. 2021 : M.S. in Artificial Intelligence, SKKU
Thesis: Learning Extra Representation by Knowledge Distillation and Adversarial Training without Extra Training Data
GPA: 4.38 / 4.5
Mar. 2015 - Feb. 2018 : B.S. in Mass Communication and Journalism, SKKU
GPA: 3.48 / 4.5
Publication
Junhyung Kang*, Hyeonseong Jeon*, Youngoh Bang, Simon S. Woo "Sarod: Efficient End-To-End Object Detection On SAR Images With Reinforcement Learning." 2021 IEEE International Conference on Image Processing (ICIP). 2021. (*equal contribution) [paper]
Hyeonseong Jeon, Siho Han, Sangwon Lee, and Simon S. Woo. "Compensating for the Lack of Extra Training Data by Learning Extra Representation." 15th
Asian Conference on Computer Vision (ACCV). 2020. [paper] [code] [video]Hyeonseong Jeon, Youngoh Bang, Junyaup Kim, and Simon S. Woo. "T-GD: Transferable GAN-generated Images Detection Framework." Thirty-seventh International Conference on Machine Learning (ICML). 2020. [paper] [code] [video]
Hyeonseong Jeon, Youngoh Bang, and Simon S. Woo. "FDFtNet: Facing Off Fake Images using Fake Detection Fine-tuning Network." 35th International Conference on ICT Systems Security and Privacy Protection (IFIP-SEC). 2020. [paper] [code]
Hyeonseong Jeon, Youngoh Bang, and Simon S. Woo. "FakeTalkerDetect: Effective and Practical Realistic Neural Talking Head Detection with a Highly Unbalanced Dataset." Proceedings of the IEEE International Conference on Computer Vision Workshops (ICCVW). 2019. [paper] [code] [poster]
Work Experience
LG Energy Solution (Mar.2023 - Present)
AI Research Engineer (Scientist) on Computer Vision
Nov.2021 – present
Anomaly Detection of Battery formation process for Smart factory (ESGM & ESWA)
AIRS Company, Hyundai Motor Company (Dec.2020 - Mar.2023)
AI Research Engineer (Scientist) on Computer Vision
Nov.2021 – Mar.2023
Hyper-scale vision-AI with self-supervised learning and text-image multimodal learning on NVIDIA Super pods.
AutoML about serving hyper-scale vision models. (research & service products)
May.2021 –Mar.2023
Vehicle vision inspection with deep anomaly detection algorithms. (In Operation in Hyundai Motor Ulsan factory 5)
Mar.2021 –Mar.2023
Wheel design generator with image-to-image translation (In Operation)
Car design generation with StyleGANs and text-to-image translation Dall-e for car design. (PoC).
Jan.2021 – Mar.2023
AI platform team helper - building team docker image, data collection and maintenance, easy-to-use vision-AI development environment, and maintenance with GPU. (MLOps)
Jan.2021 – Feb.2022
Driver status monitoring with a deep gaze estimation. (PoC)
Jan.2021 – Mar.2021
Battery Management System analysis in BEV based on deep learning.
Research Project Experience
Mar.2023 - : Anomaly detection with AutoML
Anomaly Detection using image retrieval with semi auto-labeling
Jan.2021 - Mar.2023 : Self-supervision and representation learning
Image-Text multi-modal learning with hyper-scale AI
June.2020 - Jan.2021 : Satellite and Synthetic Aperture Radar (SAR) image processing based on deep learning with DASH lab.
Efficient Object Detection with Reinforcement Learning (Policy gradient) on SAR images
Single shot Super Resolution on Satellite images
Mar.2020 - Feb.2021 : AI Generated Deepfake Multimedia Detection Research funded by National Research Foundation of Korea (NRF).
Face & Human Body Pose Generative Adversarial Networks
GAN-generated images detection
Jun. 2019 - Dec.2019 : Smartphone Continuous Touch-Dynamics Authentication based on deep learning with Samsung Research (SR).
Sequential Sensor data processing
One-class classification
Auto Encoder & Siamese Networks
Apr. 2019 - Jun. 2019 : Augmented Reality with Object Detection via Intel Real-Sense in A.I. Robot Computing class.
Single Shot Detection (SSD) & MobileNet
OpenCV & OpenGL
Windows Real-Sense API
Feb. 2019 - Apr. 2019 : Real-time Object Door Detection in Hallway with CnR lab.
tiny You Look Only Once (YOLO)
Robot Operating System (ROS)
Linux Ubuntu Real-Sense API
Sep. 2018 - Dec. 2018 : Searching for Optimal Geographical conditions of Government-established Pre-kinder in Seoul Bigdata campus Contest.
Geographical Data
K-Nearest Neighbor (KNN) & K-means clustering
LSTM
Professional Services
Reviewer - WACV (2021, 2022, 2023)
Awards and Grants
Dec. 2018 : The grand prize of big data education course from the Korea Chamber of Commerce & Industry (KCCI).
Dec. 2018 : The Excellent Award from Seoul big data campus contest (Searching for optimal geographical conditions of government-established pre-kinder in Seoul).
Dec. 2014 : The Citizenship Award in Seoul public service worker from the mayor of Seoul.
Technical Skills
Programming Language : Python (proficient), C++ (experienced)
Deep learning frameworks : Tensorflow, Keras, Pytorch
Machine learning frameworks : Scikit-learn