Experience
Industry
AtomBeam Technologies Inc
Jun 2024 – Jan 2025
Intern Research Scientist/Consultant, Sep 2024 – Jan 2025
Moraga, CA, USA (Remote)
Consultant, Jun 2024 – Aug 2024
- Phase Unwrapping (InSAR): Worked on developing deep learning-based phase unwrapping for Interferometric Synthetic Aperture Radar (InSAR).
Thales
Jun 2023 – Aug 2023
PhD Intern
Pasadena, CA, USA
Digital Identity & Security
- Contactless Fingerprint Recognition: Worked on contactless fingerprint recognition for mobile devices. Improved fingertip detection performance and added custom image denoising to replace the built-in noise reduction available in mobile devices.
Academic
University of Missouri–Kansas City
Sept 2021 – May 2024
Graduate Research Assistant, Sept 2021 – May 2024
Kansas City, MO, USA
Multimedia Computing & Communication Lab
- Neuromorphic SIFT: Developed a system that detects SIFT keypoints directly from event stream captured via event camera.
- Privacy-Preserving Face Recognition with Lensless Camera: Developed a face recognition system that maintained user’s privacy. It used sensor measurements from a lensless camera which are incomprehensible to humans.
- Video Deduplication: Developed a location-aware video deduplication system that retrieves duplicate videos along with their precise locations.
- Aerial Image Classification: Developed a frequency domain deep learning-based aerial image classification system. It directly learns from lensless camera’s sensor measurements
Teaching Assistant, Sept 2023 – Dec 2023
- Course name: CS/ECE 5582 Computer Vision
Gachon University
Feb 2017 – Aug 2019
Graduate Research Assistant
Seongnam, South Korea
Pattern Recognition & Machine Learning Laboratory
- Multinational License Plate Recognition (LPR): Developed a deep learning based LPR system applicable to multinational license plates. The system worked in real-time consuming about 42 ms per image on average.
- Automatic Trimap Generation: Designed and implemented a fully automatic trimap generation algorithm for image matting. The system was based on image saliency, graph cut segmentation, & fuzzy c-means clustering.
- Drone PV Module Inspection: A drone-based solution was developed to automatically detect and estimate the exact location of faulty solar panels in a solar power station.
- Allergic Rhinitis Prediction: A system based on AI was developed to predict the likelihood of allergic rhinitis. The system used ANN and RNN to learn from weather & user data for a personalized prediction.