THE DESIGN OF HAND GESTURE CONTROLLED VIRTUAL MOUSE USING CONVOLUTIONAL NEURAL NETWORK

THE DESIGN OF HAND GESTURE CONTROLLED VIRTUAL MOUSE USING CONVOLUTIONAL NEURAL NETWORK

Niharika K¹, K Bhuvanesh², Mohan M³, Muhammad Zidan K M⁴ and Nagaraj M. Lutimath ⁵

¹²³⁴ Students of Computer Science and Engineering, Dayananda Sagar Academy of Technology and Management

⁵ Associate professor Computer Science and Engineering, Dayananda Sagar Academy of Technology and Management

---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract - This exploratory paper proposes a way to implement the functionalities of the cursor based on hand gestures with the use of computer vision and deep learning models which aim to emulate the swaying motion of human hand gestures hold the key to further improving the performance of such computer vision solutions. Many technologies are continually evolving in today's technological environment. The human-machine interface is one such promising concept. The idea is to make use of hand gestures to implement the functionalities of the mouse on the screen without using any hardware, and merely by exploiting finger motions, known as gesture recognition. We present a unique Human-Computer Interaction (HCI) approach in this study. The python dependencies that will be used for implementing this system are OpenCV, MediaPipe, and the latest packages such as PyAutoGUI.

Key Words:  Computer Vision, Deep Learning, Human-Computer Interaction (HCI), MediaPipe,  PyAutoGUI