Duc Tai Phan

phantaiduc2005@gmail.com

AI Researcher

Short Bio: My name is Phan Tai Duc (Phan Tài Đức - Vietnamese). I am a person who is deeply enthusiastic about exploring new AI research and mathematical expression. My goal in the next few years is to become an AI Engineer role, focusing on advancements in Computer Vision and NLP, especially on time series forecasting. I am currently a third-year student and an AI Researcher Assistant with 1 year of experience specializing in Computer Vision at the AiTA Lab - Artificial Intelligence of Technologies and Application - at FPT University HCMC. My expertise includes image classification, object detection, and face recognition.

Research Interests: Artificial Intelligence, Deep Learning, Machine Learning, Computer Vision, NLP, GenAI. My main research field is applying machine learning techniques to a specific use case in CV and NLP.

Inspirational Quotes:

ABC - Always Be Curious - Josh Starmer

📰News📰

Aug 10, 2025	🎓 2nd Prize in Student Research Competition at FPT University!
Aug 02, 2025	📰 02 papers is accepted at The 25th Asia-Pacific Network Operations and Management Symposium (APNOMS 2025), Kaohsiung, Taiwan.
Nov 05, 2024	📰 01 paper is accepted at The 18th IEEE-RIVF International Conference on Computing & Communication Technologies (RIVF 2024), Da Nang, Vietnam.
Aug 13, 2024	📰 01 paper is accepted at The 15th International Conference on ICT Convergence (ICTC 2024), Jeju, Korea.

📝Selected Publications📝

Improving Face Attendance Checking System with Ensemble Learning

Duc Tai Phan, Nam Phuong Tran, and Duc Ngoc Minh Dang

In 18th IEEE-RIVF International Conference on Computing and Communication Technologies , Oct 2024

Abs Code

In many industrialized countries, implementing an efficient face attendance checking system is crucial for effective staff management. The face attendance system nowadays encounters such problems. The biggest challenge to be concerned with is the various light and angle conditions, which can damage the performance of these systems. In this paper, we present an innovative approach that combines the results of GoogLeNet, VGGFace, and FaceNet to achieve higher accuracy through Ensemble Learning. With this solution, newer systems can achieve better results. Our system encompasses image capture, face detection, database management, and feature extraction for face recognition. All related code is available at https://anonymous.4open.science/r/Face-Attendance-Checking-2640/
Towards Real-time Vietnamese Traffic Sign Recognition on Embedded Systems

Nam Phuong Tran, Nhat Truong Pham, Nam Van Hai Phan, Duc Tai Phan, Tuan Cuong Nguyen, and Duc Ngoc Minh Dang

In 2024 15th International Conference on Information and Communication Technology Convergence (ICTC) (ICTC 2024) , Oct 2024

Abs

In recent years, AI development has brought many significant changes in various aspects of our daily lives. Integrating AI technology into various applications has revolutionized multiple domains, and one particularly vital area is traffic sign recognition, which significantly enhances driver safety. This paper presents an approach to traffic sign recognition specifically designed for the Jetson Nano 2GB device. By utilizing the YOLOv8 Nano model, the proposed approach achieves a remarkable frame rate of up to 32 frames per second (FPS). To optimize inference speed on Jetson with limited memory, the approach incorporates TensorRT and quantization techniques. In addition, this paper introduces a dataset called the Vietnamese Traffic Sign Detection Database 100 (VTSDB100). This dataset is an extension of the VTSDB46 dataset and encompasses a comprehensive collection of 100 different classes of traffic signs. These signs were captured in diverse locations within Ho Chi Minh City, Vietnam, providing a rich and diverse dataset for training and evaluating traffic sign recognition models. An extensive experiment and analysis are also conducted using various object detection methods on the VTSDB100 dataset. The findings highlight the potential of deploying the proposed approach on resource-constrained devices and provide valuable insights for further research and development in the field of AI-powered driver safety systems.