Real-time UAV Detection System
Leader
The proliferation of small Unmanned Aerial Vehicles (UAVs) poses significant security challenges to critical infrastructure. This project presents a computer vision system using a YOLOv8-based architecture enhanced by:
• Multiscale Processing - 5-crop inference mechanism with Non-Maximum Weighted (NMW) fusion
• Cross-Head Knowledge Distillation (CrossKD) - Efficient transfer of detection-sensitive features from Teacher to Student model
Github repository →
View Report Slides →
Python
PyTorch
Yolo
Computer Vision