Human body gesture-controlled gaming application

Gaming has always held the requirement of peripheral devices such as keyboards, mouse, joystick, etc. But with computer technology evolving every single day, human-body gestures are found to be a potential replacement for the primary accessory of Human-Computer Interaction (HCI). In this paper, an idea of a human-body gesture controlled gaming desktop application has been proposed. With a simple web camera, body gestures have been detected and image processing techniques have been performed with the help of mediapipe - a prebuilt computer vision neural network, which enables a user to directly interact with the gaming console and with the use of OpenCV, a library designed to solve computer vision problems and which provides many useful programming functions used for features like face, edge, object detection, etc. So the input images can be processed and communicated with the gaming commands virtually. The activity of the said user is recorded and stored to MongoDB, an open source No-SQL document based database program. With this desktop application, a user can play any of the various games offered on the portal.