Overview:
This project demonstrates object blurring in computer vision, using the YOLOv8l model to detect and blur specific objects within images. It was trained on a dataset of 4680 images across 10 vehicle classes (e.g., cars, trucks) to apply blur to detected objects in real time. The system achieved a mean average precision (mAP) of 0.764 after 50 training epochs.
Key Features:
1. YOLOv8l for Object Detection:
The system uses the YOLOv8l model to detect objects, particularly vehicles, with higher accuracy and speed. The model identifies vehicles in real-time, allowing it to quickly locate objects that need to be blurred.
2. Real-Time Object Blurring:
Once objects are detected, the system applies blurring to conceal specific elements within the image, such as vehicle plates, ensuring privacy.
3. Diverse Dataset:
The system was trained on 4680 images across 10 vehicle classes, enabling it to handle a variety of vehicles effectively in different scenarios.
4. Performance: The model achieved a mean average precision (mAP) of 0.764, with precision of 0.757 and recall of 0.687, ensuring reliable detection and object blurring.
Category:
Tags:
Links
Portfolio 
Leave a comment