Hello! :)
I am Truc.
I am a research intern at Sony Europe, specializing in developing computer vision algorithms for smartphone cameras. I possess a deep understanding of image processing and a passion for training deep neural networks and optimizing them to make them lightweight.
Previously, I was a Graduate Student Researcher in Colourlab at Norwegian University of Science and Technology, where I pursued an M.S. in Computational Colour and Spectral Imaging. My research interests include deep learning, computer vision, and image processing. I also worked with Viettel AI as a Computer Vision Researcher.
I enjoy videos of 3Blue1Brown and
Veritasium.
I love hamsters. I own a YouTube channel with 283K subscribers.
Connect with me on LinkedIn!
Research Interests
- Deep Learning
- Computer Vision
- Image Processing
Education
Experience
Research Intern
Sony Europe
- Improving the performance of multiple illumination estimation from multispectral image.
- Utilizing the estimated spectral illumination to perform auto white balance on RGB image.
- Developing deep learning models for multispectral reconstruction from optical system that 90% faster than the original model.
- Working in the optical lab for data capturing and calibrating.
- Manual data labeling and developing auto labeling.
Graduate Student Researcher
ColourLab
- Working on multispectral/hyperspectral image analysis for cultural heritage.
- Worked on underwater surveillance project in segmenting known and unknown parts in ship hull images.
- Supported an eyeglasses company by assessing the visual quality of the glasses in different conditions.
- Developed a renderer for multispectral 3D scenes from spectral reflectance, spectral power distribution and camera models.
- Analyzed the correlation of light scattering with visual opacity in translucent colorants in printing industry.
Teaching Assistant
Norwegian University of Science and Technology
- Conducted lab sessions and provided individual guidance to students.
- Assisted with Matlab-based assignments, clarifying complex image processing concepts.
- Provided feedback on assignments to reinforce understanding of key topics.
- Mentored students through coding challenges, encouraging independent problem-solving.
Graduate Student Researcher
Laboratoire Hubert Curien
- Implemented multiple approaches for 3D reconstruction of a static scene using multi-view images.
- Developed a pipeline for inserting virtual objects into a 2D image realistically.
- Improved SVM classification performance by 5% on Credit Card Approvals dataset.
- Implemented Phong reflection model for real-time colored lighting using depth and normal maps.
- Reviewed deep learning techniques for complex indoor image relighting tasks.
Research Assistant
University of Alberta
- Awarded Canada Scholarships and Educational Exchanges for Development for excellence undergraduate student researcher.
- Conducted research in Medical Image Processing and Artificial Intelligence.
- Implemented solutions to support the 2030 Agenda for Sustainable Development and address poverty and development gaps in ASEAN region.
Computer Vision Researcher
Viettel AI
- Worked with a team of 5 engineers to design and build an eKYC system for ViettelPay app, reaching 98% accuracy and reducing processing speed 16x compared to previous technology.
- Developed a general text recognition model achieving 95% accuracy across diverse text types.
- Implemented error correction algorithms, reducing recognition output errors by 90%.
Accomplishments
- Course 1: Supervised Machine Learning: Regression and Classification
- Course 2: Advanced Learning Algorithms
- Course 3: Unsupervised Learning, Recommenders, Reinforcement Learning
- Granted CAD$15400 for research works in Department of Radiology and Diagnostic Imaging of University of Alberta, Canada.
- Research project: Segmentation of dental issues in intraoral ultrasound with Deep Learning.
- Exchange experience for undergraduates from ASEAN and other countries to engage in meaningful academic, cultural, and social exchanges.
- Conducted thoughtful analysis of public administration in Vietnam and wrote the paper: "Why OCR is the key for public administration optimization in Vietnam?"
The International English Language Testing System (IELTS) is an international standardized test of English language proficiency for non-native speakers.
See certificate
- Selected among top 50 out of 1000 students.
- Studied React Native intensively for 10 weeks to build production-ready mobile apps.
- Collaborated with students from different tracks to solve a company project.
- Final winning team after two demo days with a Fraud Detection application.
- Promoted women in tech by joining Shecodes NPO.
- Built a desktop app during a 12-hour hackathon supporting work-from-home employees.
- Course 1: How Google does Machine Learning
- Course 2: Launching into Machine Learning
- Course 3: Build, Train, and Deploy ML Models with Keras on Google Cloud
- Course 4: Feature Engineering
- Course 5: Machine Learning in the Enterprise
Publications
This work presents a hyperspectral imaging-based approach to digitally reconstruct ancient textile textures, enabling preservation and study of historical artifacts.
Accepted - AIC 16th Congress, 2025
This poster introduces a method for lighting estimation in augmented reality, enhancing object realism by ensuring spatial lighting coherence through deep learning.
CVCS 2024, Norway
This paper proposes a deep transformer-based framework for text-guided image retrieval that refines search results based on iterative natural language feedback.
DOI
NICS 2021, IEEE