Dai Quoc Tran

Ph.D. in Construction Engineering and Management, Sungkyunkwan University, Korea.
Email - LinkedIn - Google Scholar - GitHub

Research Interest

My research is positioned at the intersection of transportation engineering, construction management, and advanced technologies including deep learning and computer vision. I am committed to developing AI-driven solutions that enhance the safety, efficiency, and sustainability of transportation systems and construction sites. My work involves designing innovative deep learning algorithms for real-time hazard detection and multi-camera object tracking, essential for both site and traffic management. I also develop enhancement frameworks for low-light imaging, crucial for the consistent performance of autonomous vehicles and surveillance systems in varying light conditions. Utilizing UAV imagery, CCTV feeds, and meteorological data, I focus on early problem identification and structural health monitoring for infrastructure resilience. Additionally, my research extends to AI-enhanced disaster management, specifically for rapid response and damage assessment in forest fire scenarios. By employing domain adaptation and transfer learning, I ensure the effectiveness and scalability of AI applications, fostering advancements in smart infrastructure management and supporting the evolution towards more intelligent transportation networks.

Experiences

Postdoctoral Researcher

Global Frontiers of Resilient EcoSmart City, Sungkyunkwan University, Suwon, Korea (March 2023 - Present)
Received a three-year project grant from the National Research Foundation of Korea (NRF) to develop an AI-based CCTV system for monitoring construction sites.
Role: Principal Investigator
Budget: 210,000,000 WON
figure

AI Team Lead

SmartInside AI Co., Ltd, Suwon, Korea (February 2021 - Present)

Research Assistant

Smart Construction IT Lab., Sungkyunkwan University, Suwon, Korea (September 2018 - Feb. 2023)

Research Assistant

National Institute of Technology, Gifu College, Gifu, Japan (August 2017)

Awards

Projects

AI-based CCTV System for Monitoring Construction Sites

Project 1

Low-Light Image Enhancement Framework for Improved Object Detection in Fisheye Lens Datasets

Project 2

SSIMS.AI for SmartInside.AI

Project 3

Image-to-Hazard: GPT-based Logic Reasoning for Hazard Identification in Construction Site using CCTV Data

Project 4

Forest-Fire Response System

Project 5

VQA-RESCon

Project 6

Publications

Authors Title Publication/Conference Year Link
Dai Quoc Tran, Yuntae Jeon, Armstrong Aboah, Jinyeong Bak, Minsoo Park, Seunghee Park† Leveraging Semi-Supervised Learning for Domain Adaptation: Enhancing Safety at Construction Sites through Long-Tailed Object Detection ASCE’s Journal of Construction Engineering and Management 2024 Link
Minsoo Park, Almo Senja Kulinan, Dai Quoc Tran, Jinyeong Bak, Seunghee Park† Preventing falls from floor openings using quadrilateral detection and construction worker pose-estimation Automation in Construction 2024 Link
Dai Quoc Tran, Armstrong Aboah, Yuntae Jeon, Maged Shoman, Minsoo Park, Seunghee Park† Low-Light Image Enhancement Framework for Improved Object Detection in Fisheye Lens Datasets CVPRw 2024, Seattle, US 2024 Link
Dai Quoc Tran, Armstrong Aboah, Yuntae Jeon, Minsoo Park, Seunghee Park† Vision-Language Model-based PolyFormer for Recognizing Visual Questions with Multiple Answer Groundings CVPRw 2024, Seattle, US - TOP3 VizWiz Grand Challenge Submission 2024 Link
Dai Quoc Tran, Yun Tae Jeon, Tae Heon Kim, Min Soo Park, Joo Ho Shin, Seunghee Park Image-based object recognition method and system based on learning of environment variable data United States Patent 2024 Link
Dai Quoc Tran, Yuntae Jeon, and Seunghee Park† Automatic Construction Safety Report using Visual Question Answering and Segmentation Model ICCCBE, Montreal, Canada 2024 Link
Dai Quoc Tran, Yuntae Jeon, Minsoo Park, and Seunghee Park† GPT-based Logic Reasoning for Hazard Identification in Construction Site using CCTV Data ISARC 2024, Lille, France 2024 Link
Dai Quoc Tran, Yuntae Jeon, Seongwoo Son, Minsoo Park, and Seunghee Park† Identifying Hazards in Construction Sites Using Deep Learning-Based Multimodal with CCTV Data CONVR2023, Florence, Italy 2023 Link
Daniel Asefa Beyene, Dai Quoc Tran, Michael Bekele Maru, Huangrui, Seunghee Park† Unsupervised Domain Adaptation-Based Crack Segmentation Using Transformer Network Journal of Building Engineering 2023 Link
Minsoo Park, Dai Quoc Tran, Jinyeong Bak, Almo Senja Kuliman, Seunghee Park† Real-time monitoring unsafe behaviours of portable multi-position ladder worker using deep learning based on vision data Journal of Safety Research 2023 Link
Yuntae Jeon, Dai Quoc Tran, Minsoo Park, and Seunghee Park† Leveraging Future Trajectory Prediction for Multi-Camera People Tracking CVPRw 2023, Vancouver, Canada 2023 Link
Minsoo Park, Dai Quoc Tran, Jinyeong Bak, Seunghee Park† Small and overlapping worker detection at construction sites Automation in Construction 2022 Link
Minsoo Park, Dai Quoc Tran, Jinyeong Bak, Seunghee Park† Advanced wildfire detection using GAN-based augmented datasets and weakly supervised object localization International Journal of Applied Earth Observation and Geoinformation 2022 Link
Dai Quoc Tran, Minsoo Park, Yuntae Jeon, Jinyeong Park, and Seunghee Park† Forest-fire response system using deep-learning-based approaches with CCTV images and weather data IEEE Access 2022 Link
Dai Quoc Tran, Minsoo Park, Daekyo Jung, and Seunghee Park† Damage-Map Estimation Using UAV Images and Deep Learning Algorithms for Disaster Management System Remote Sensing 2020 Link
Dai Quoc Tran, Ju-Won Kim, Kassahun Demissie Tola, Wonkyu Kim, and Seunghee Park† AI-Based Bolt Loosening Diagnosis Using Deep Learning Algorithms for Laser Ultrasonic Wave Propagation data Sensors 2020 Link