MUMBAI, India, Aug. 24 -- Intellectual Property India has published a patent application (202641098059 A) filed by Vignans Nirula Institute Of Technology And Science For Women on August 13, 2026, for A Machine Learning Approach For Hand Gesture Recognition Using Mediapipe And Opencv.
Inventors include B. Aruna Kumari; V. Pavani; K. V. S. S. Rama Krishna; P. Sandhya Krishna; I. Naga Varshitha; R. Vasavi; M. Jeevana Jyothi; and P Prasanthi.
The application for the patent was published on August 21, 2026, under issue no. 34/2026.
Abstract: The present invention is related to the field of machine learning applications for hand gesture recognition using MediaPipe and OpenCV for facilitating touchless human–computer interaction in a real-time, accurate manner. The proposed system mainly consists of the following tasks: Image acquisition, Image pre-processing, Hand detection and Landmark detection. The first step of the image pre-processing is to improve the captured image using open-cv, and then find twenty-one (21) 3-D hand landmarks with the help of MediaPipe hands module. The landmark positions are subsequently transformed to feature vectors consisting of distance, relative position and angles between fingers to be used as input for training and testing machine learning classification models. Video images can be captured by a standard RGB camera for gesture recognition, and recognized gestures can be used for controlling home appliances, computer applications, multimedia systems, robots, healthcare devices and IoT. The proposed invention is also different from the current gesture recognition system as it does not involve special purpose hardware such as a data glove or a depth camera to recognize gestures, and can be implemented using an ordinary RGB camera to capture video images. Furthermore, the present invention provides more accurate, faster computation, illumination and background invariance and robustness to ensure reliable real-time performance. In this way, the invention proposes an improved solution to the above- mentioned problems, which is both efficient and economic, in an advanced human-computer interaction and home automation system, for real-time applications.
Disclaimer: Curated by HT Syndication.
