About
- Name: Huynh Thai Hoa (Huỳnh Thái Hòa)
- Affiliation: Soongsil University / DeltaX
- Email: [email protected]
- Leetcode: huynhthaihoa
- Kaggle: huynhthaihoa1995
- Linkedin: huynh-thai-hoa-qn
I am a software engineer who loves to do research in Computer Vision and Artificial Intelligence. I am working toward the goal of building a better world for my dearest ones, especially in the fields of transportation, education, and healthcare.
Skills & expertise
- Programming languages: C, C++, Python, C#, MATLAB, Java, Javascript, R.
- Fundamentals: Data Structures & Algorithms, Computer Vision, Deep Learning, Machine Learning.
- Frameworks/Libraries: OpenCV, Tensorflow, Pytorch, Darknet, FFMPEG, DeepStream, WPF, Qt.
- Version control: Git, SVN.
- Databases: Microsoft SQL Server, MySQL, SQLite, Postgres.
- Operating system: Windows, Ubuntu.
- Hardware platform: Jetson Nano, Jetson Xavier NX.
- Languages: Vietnamese (Native), English (Professional), Korean (Basic).
- Others: Microsoft Office, LaTEX, JSON, XML, Batch.
Education
- Master of Engineering in Information Communication - Soongsil University (Republic of Korea) | 08/2018 - 08/2020
- GPA: 4.01/4.05
- Thesis: LED Taillight Matching in Nighttime Vehicle Positioning
- Supervisor: Professor Myungsik Yoo
- Bachelor of Engineering in Software Engineering - University of Information Technology, Vietnam National University in Ho Chi Minh City (Vietnam) | 08/2013 - 06/2018
- GPA: 8.6/10
- Thesis: Coronary Arteries Segmentation from Coronary Angiogram
- Supervisor: M.Sc. Thi-Vuong Pham
Work experience
- AI Researcher - DeltaX (Republic of Korea) | 06/2022 - now
- Research Engineer - Smart Deep (Republic of Korea) | 05/2021 - 05/2022
- Participated in Deep Neural Network training pipeline: data labeling, cleaning, and training.
- Developed tools for preprocessing datasets and evaluating inference model consistency.
- Developed new features and improved client programs’ performance and dynamic-link libraries (DLL).
- Developed a video processing system for interesting object detection.
- Researched and proposed solutions for company projects.
- Research Intern - Ubay Solution (Republic of Korea) | 12/2020 - 04/2021
- Researched to find solutions for Optical Character Recognition (OCR) problem on detecting and recognizing license plate numbers and container codes.
- Evaluated the performance and consistency of several deep learning frameworks on our custom dataset.
- Deployed inference model on NVIDIA Jetson Nano board
- Research Assistant - ANDA Lab (Republic of Korea) | 08/2018 - 08/2020
- Researched to find and solve problems in the vehicle localization process based on detecting vehicle taillights using camera images.
- Read and wrote scientific papers related to the project.
- Trainee - Global Cybersoft JSC (Vietnam) | 03/2017 - 08/2017
- Learned the software development process in practice.
- Supported team to design and develop plugins as required by customers.
- Attended the peer-review process that uses appropriate tools to evaluate code quality.
Publications
- Thai-Hoa Huynh, Myungsik Yoo, Taillight Matching and Pairing Algorithm for Stereo-Vision-based Nighttime Vehicle Positioning, Applied Sciences (2020)
- Thai-Hoa Huynh, Myungsik Yoo, Nighttime LED Taillight Detection Method with Optical Camera Communication System, 한국통신학회논문지 (2020)
- Thai-Hoa Huynh, Tuan-Anh Pham, Myungsik Yoo, Detection Algorithm for Overlapping LEDs in Vehicular Visible Light Communication System, IEEE Access (2019)