A Smart System for Tracking and Analyzing Human Hand Movements using MediaPipe Technology and TensorFlow

Preview this chapter:

A Smart System for Tracking and Analyzing Human Hand Movements using MediaPipe Technology and TensorFlow, Page 1 of 1

< Previous page | Next page > /docserver/preview/fulltext/9789815324099/chapter-19-1.gif

Gesture recognition is the latest and the most popular technology nowadays. The main aim of this technology is to recognize human body parts using mathematical algorithms MediaPipe and TensorFlow. The hand is a very important part of the human body for performing any activity. The detection and analysis of body language have recently gained a lot of attention. In this paper, we look at the skeleton poses of a person. It is easy to grasp and have images with low dimensionality statistics. Underfed interpretations generalize a person's appearance and background, allowing them to be identified. This paper describes a real human chasing channel capable of anticipating the structure of both hands and the location of the fingers, focusing on motion recognition, and creating virtual hand brushes that can be very beneficial and easing activities like the selection of colors and paintings in combination with paint art. This paper, which uses hand gesture detection and has a 95% confidence accuracy rate, was built using MediaPipe, a deep learning framework, in addition to assessing numerous static or dynamic hand motion detection methods.