Projects
Natural Rhythm: Real-Time EEG/EDA Biofeedback for XR Mindfulness.
Supervised by Prof. Rolf Nordal and Stefania Serafin, Aalborg University Copenhagen, Denmark.
Unity
3D Spatial Sound
Machine Learning
C#
1. Design & built the spatial 3d sound in Unity, explored how nature and musical rhythm can create a
soothing effect in the brain and contributed to the experiment design, including user testing setup and
data collection for system evaluation. (In preparation for Conference).
2. Implementation part is Awarded 3rd Prize in Programming & Artistic Project at the G.Tec BCI &
Neurotechnology Spring School 2025 Hackathon (Copenhagen, Denmark/Worldwide event)
Aseptic Training in Virtual Reality
Unity
VR
Machine Learning
C#
1. Conducted a comprehensive literature review to align the VR application with industry regulations and training needs.
2. Contributed to the design and development of immersive 3D virtual environments simulating cleanrooms.
3. Implemented machine learning using InteractML to dynamically adapt the VR environment based on user position, enhancing personalized training.
Position-Based Interaction Using InteractML in Unity
Unity
Machine Learning
InteractML
C#
1. Designed and implemented a real-time ML classification system in Unity using InteractML to map avatar position
data to dynamic environmental changes.
2. Recorded and prepared a dataset of live position data to train the classification model with distinct movement states.
3. Collected and labeled live position data to train a KNN model for binary movement state classification.
Mood-Based Chord Progression App – Compose with Emotion
Android
Java
Music Theory
Real-Time Audio
1. Developed an Android application in Java that generates chord progressions based on selected key and emotional tone (e.g., Happy, Sad, Energetic, Romantic, Mysterious).
2. Designed a music theory–based algorithm to drive emotionally contextual chord sequences and enabled real-time audio playback for intuitive composition.
Object Detection using Vision Transformer.
Python
Keras
TensorFlow
Jupyter Notebook
1. Implemented an advanced object detection system leveraging Vision Transformer architecture.
2. Developed a robust solution for detecting and classifying objects in images.
3. Utilized cutting-edge technologies to achieve state-of-the-art performance in computer vision tasks.
Image Quality Metrics Comparison.
Python
OpenCV
scikit-image
Jupyter Notebook
1. Conducted a comprehensive analysis of image quality using various metrics, including Mean Squared Error (MSE), Structural Similarity Index Measure (SSIM), and Peak Signal-to-Noise Ratio (PSNR).
2. Evaluated and compared the performance of these metrics across different sets of images.
3. Provided insights into the strengths and limitations of each metric in assessing image quality.
Find Device Address with Geocoding API.
Android
Java
Geocoding API
1. Implemented a location-based service using the Android Location Manager to obtain accurate longitude and latitude coordinates.
2. Integrated the Geocoding API to translate obtained coordinates into human-readable addresses.
3. Developed a seamless solution for efficiently retrieving and displaying device locations on-demand.
News App.
Android
Java
News Api
Picasso
1. Integrated the News API to fetch and display real-time news articles from diverse sources.
2. Successfully developed a feature-rich news app with a focus on user experience and accessibility.
3. Overcame challenges related to API integration, offline functionality, and error handling to deliver a seamless news
browsing experience.
