This project combines real-time object detection using YOLOv8 with concurrent neural network training visualization. It processes a video stream, detects objects, collects training data, and simultaneously displays object counts and neural network training progress.
- Real-time object detection using YOLOv8
- Concurrent neural network training on detected objects
- Live visualization of object counts and training loss
- Multithreaded processing for smooth performance
- Python 3.7+
- OpenCV
- NumPy
- Matplotlib
- Ultralytics YOLO
- TensorFlow/Keras
pip install -r requirements.txt