Welcome to my GitHub profile! I am a Ph.D. candidate in Computer Science at the University of Tennessee at Chattanooga, specializing in computer vision. My research focuses on vehicle tracking, federated learning, sensor fusion and trajectory prediction.
I am currently passionate about advancing the field of computer vision and applying my expertise to real-world problems. My key interests include:
- Federated Learning
- Vision Language Models (VLM)
- Vision Foundation Models (VFM)
- 3D Scene Understanding
- Autonomous Systems
Developed and implemented real-time multi-camera vehicle tracking systems for various intersection scenarios using data from:
- Roadside cameras
- Fisheye cameras
- Fusion of camera and LiDAR data
Designed a federated learning-based trajectory prediction algorithm to detect potential near-crash scenarios and ensure the safety of Vulnerable Road Users (VRUs) like pedestrians, motorcyclists, and cyclists.
Developed a privacy-preserving federated learning approach for multi-camera vehicle tracking, enabling decentralized training on distributed datasets.
Developed an application for real-time detection of available parking spaces using high-mounted cameras within parking lots, leveraging computer vision techniques to analyze camera footage.
Created an application for comprehensive traffic video analysis utilizing object segmentation, detection, and classification techniques to detect and classify various road users.
-
LaMMOn: Language Model Combined Graph Neural Network for Multi-Target Multi-Camera Tracking in Online Scenarios
- Machine Learning Journal (2024) (Accepted)
- (This Journal is highly regarded, with a Q1 ranking and an Impact Factor 5.8 in Artificial Intelligence and Software)
-
Smart Camera Parking System With Auto Parking Spot Detection
- Transportation Research Board (2024)
-
Real-time Multi-Vehicle Multi-Camera Tracking with Graph-Based Tracklet Features
- Transportation Research Record 2678.1 (2024): 296-308
-
Multi-Vehicle Multi-Camera Tracking with Graph-Based Tracklet Features
- IEEE Transactions on Multimedia (2023)
- (This Journal is highly regarded, with a Q1 ranking and an Impact Factor 8.4 in Computer science and Technologies)
- Programming Languages: Python
- Frameworks and Libraries: PyTorch, OpenCV, Flower
- Techniques and Models: Federated Learning, Deep Learning, Machine Learning, Multi-Object Tracking
- Tools: Git, Jupyter Notebook
- Email: Xwz778@mocs.utc.edu
- LinkedIn: linkedin.com/in/elituan/
- Google Scholar: scholar.google.com/citations?user=mIMXTkUAAAAJ&hl=en
- GitHub : github.com/elituan
Feel free to connect with me via email or LinkedIn. I am always open to discussing new projects, collaborations, or opportunities in the field of computer vision and AI.
Thank you for visiting my GitHub profile!