Object tracking enhanced interactive pinball display

• Main Author: Tan, Bryan Zhen Ye
• Other Authors: Chua Hock Chuan
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2021
• Subjects: Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence; Engineering::Electrical and electronic engineering::Applications of electronics
• Online Access: https://hdl.handle.net/10356/149199
• Institution: Nanyang Technological University

With the recent rapid advances in machine learning and artificial intelligence, it is not difficult to leverage these technologies to create something new or give new life to an old object. We are all used to interacting with screens and their graphical interfaces and some of them do have artificial intelligence capabilities. However, it is uncommon to see interactive displays which do not use a screen. The most common place for such displays to be found would be at the Singapore Science Centre as part of an exhibition. Thus, the objective of this project is to create an interactive display that is not just a screen, but a physical device that people are able to feel and interact with in the real world. The interactive display should also encompass some modes of artificial intelligence to showcase how AI is able to breathe new life into what is considered an outdated object. In line with the objectives above, a 2-player pinball game has been created. The opponent, a computer, would also be able to see the ball and know where it is through the use of object recognition and object-tracking. Through the usage of object recognition, the player is able to rearrange the obstacles on the board anywhere he/she wants and the opponent will still be able to know where the ball is. Something new in something old. An old-school pinball board with the necessary components like flippers, obstacles and a controller is reborn.