VR Development
As a research developer specializing in AI and VR, I focus on creating immersive educational and training experiences across various fields. My work involves integrating advanced AI techniques into VR environments to enhance learning and skill development. While I am not a 3D artist, I utilize third-party assets and refine them using Blender to align with specific project requirements. My passion lies in continuous research and innovation, exploring new ways to leverage VR technology for impactful, real-world applications.
Accident Reconstruction
Hit and Run Accident Analysis Simulator
In this project, I developed a simulator for accident reconstruction to analyze the causes of hit-and-run accidents using CCTV footage. The simulator was designed as a foundation for the future development of this project, where we aim to reconstruct 3D environments of accident sites by integrating black box data and CCTV recordings.
The following platforms were used to design or develop the simulator;
-
Blender
-
Unity 3D Game Engine
-
Visual Studio (C#)
Acknowledgment
This work has been done at IKLab Inc., Seoul, South Korea.
Demo Video
Graphical User Interface (GUI)
Hit and Run Accident Analysis System
In this project, I developed a C#--based GUI to analyze the causes of hit-and-run accidents using CCTV footage. The GUI was built with various analytical features, including graphical representations of hit-and-run incidents and the ability to visualize critical accident parameters such as time, speed, and location. This system provides a structured approach to accident analysis, supporting future developments in forensic accident reconstruction.
The following platforms were used to develop the GUI;
-
Visual Studio (C#)
Acknowledgment
This work has been done at IKLab Inc., Seoul, South Korea.
Demo Video
iSafeTrainer
Personalized construction safety training system using conversational AI in virtual reality
-
Integration of VR and AI (LLMs) for personalized safety training in construction, focusing on trades like electrical, scaffolding, and welding.
-
iSafeTrainer enables conversational interactions tailored to users' knowledge levels (novice, beginner, expert).
-
An experimental study with surveys evaluated the system on user satisfaction, experience, training effectiveness, guidance, and impact on confidence.
-
Results indicate the advantages of iSafeTrainer over traditional methods, including tailored content, immersive environments, and real-time AI guidance.
-
Research demonstrates the potential of VR and AI in enhancing construction safety training by providing dynamic, personalized, and engaging learning experiences.
-
This prototype supports both text and voice inputs.
The following platforms were used to design or develop the iSafeCom;
-
Blender
-
Unity 3D Game Engine
-
Visual Studio (C#)
-
Ready Player Me
-
Conversational AI
Acknowledgment
This work has been done to fulfill my Master's thesis requirements under Professor Chansik Park's supervision at ConTI Lab, Chung Ang University, Seoul, South Korea, and is also published in a prestigious journal, Automation in Construction, DOI: https://doi.org/10.1016/j.autcon.2025.106207. Furthermore, this work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. NRF-2022R1A2B5B02002553), the “National R&D Project for Smart Construction Technology (No. RS-2020-KA156291)” funded by the Korea Agency for Infrastructure Technology Advancement under the Ministry of Land, Infrastructure and Transport, and managed by the Korea Expressway Corporation, and the Chung Ang University Young Scientist Scholarship (CAYSS) 2022.
Thesis Committee members:
-
Prof. Hyun Ki Hong (Committee Chair)
-
Prof. Chung-Gun Lee
-
Prof. Chansik Park
Demo Video
Gangform Training Platform
This project focused on developing a prototype safety training system tailored for construction workers engaged in gangform operations in South Korea. The key features of the training system include
-
Virtual environment created using Gaussian Splatting.
-
Integration of Virtual Reality (VR) and Blockchain technologies to enhance transparency, traceability, and immersion in gangform safety training.
-
Step-by-step instructional modules, guiding workers through each phase of the gangform process to ensure procedural clarity and compliance.
-
Simulated task execution, designed to expose trainees to realistic on-site hazards and improve their preparedness for high-risk environments.
The following platforms were used to design or develop the prototype:
-
Blender
-
Unity 3D Game Engine
-
Visual Studio (C#)
-
Gussain Spletting
Acknowledgment
This work has been done at ConTI Lab, Chung Ang University, Seoul, South Korea.
Demo Video
iSafeCom
-
Inadequate communication compromises the effectiveness of safety training for workers, notably migrants.
-
The study introduced iSafeCom, an AI-based virtual training system for effective communication to enhance learning outcomes.
-
iSafeCom shows equal efficacy across diverse groups and yields a 23% increase in participants' scores after intervention.
-
Global organizations can integrate iSafeCom into diverse safety training with minimal technical challenges to adoption.
The following platforms were used to design or develop the iSafeCom;
-
Blender
-
Unity 3D Game Engine
-
Visual Studio (C#)
-
Ready Player Me
-
Conversational AI
Acknowledgment
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. NRF-2022R1A2B5B02002553) and the “National R&D Project for Smart Construction Technology (No.23SMIP-A158708-04)” funded by the Korea Agency for Infrastructure Technology Advancement under the Ministry of Land, Infrastructure and Transport, and managed by the Korea Expressway Corporation. This study was conducted with equal contributions from Rahat Hussain and Aqsa Sabir. Therefore, they both have the right to share the first authorship of the research paper in their resumes. DOI: https://doi.org/10.1016/j.autcon.2024.105315