Implement object detection and recognition models using YOLO and deep learning frameworks

Learning Objectives

At the end of this theory note, trainees must be able to:

1. Distinguish object detection from image classification and localisation.
2. Prepare, annotate, and organise a dataset for object detection using recognised annotation formats.
3. Explain the architecture and working principles of YOLO models.
4. Configure and train a YOLO model, and run inference on images and video.
5. Evaluate detection performance using IoU, precision, recall, and mAP.
6. Apply transfer learning and hyperparameter tuning to refine a detection model.