The objective of this project is to improve object detection for items like pens and books in various environments. We trained the YOLOv8 model using annotated images, tested its accuracy, and applied it to real-time video footage.
Detecting objects such as pens and books in real-time can be highly beneficial for numerous applications including inventory management, automated checkout systems, and organizing academic resources. Manual detection is prone to errors and inefficiencies. By leveraging advanced AI techniques, this project aims to automate and improve the accuracy of object detection, ensuring efficiency and reliability.
We used Python version 3.8 along with packages such as ultralytics, opencv, numpy, torch, matplotlib, and roboflow. For more details on YOLOv8, refer to the YOLOv8 Documentation. Detailed steps and findings are available in the provided notebook and additional documentation.