Computer games are a popular consumer electronics item. The game players find it captivating to interact with games via joysticks, buttons, trackballs, or wired gloves. They may find it even more engaging to interact through natural, unencumbered hand or body motions. A computer vision-based user interface could provide these capabilities. Computer games represent a possible mass-market application for computer vision. So in this project we exploit this mass-market application and try to build and interactive game which can be controlled using a particular object and also build an AI which can play the game on its own.
Using OpenCV to control actions in the game using a particular object for which the model is trained
Using PyGame, which is a good library to make simple 2-D games like space shooter etc.
Using Behaviour Tree which is a mathematical model of plan execution which describes switching between a finite set of tasks in a modular fashion & Reinforcement Learningn using which Q and DQ-Network models were applied to game
Using Intel RealSense camera which tracks the 3-D movements and helps play game using our gestures
People who worked hard together to make this project a success