MUMBAI, India, June 24 -- Intellectual Property India has published a patent application (202441098019 A) filed by Rv Institute Of Technology And Management on December 11, 2024, for A Method And System For Identifying Gesture Of One Or More Disabled Persons.
Inventors include Krishnamurthy H; Rahul S; and Sameer.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: ABSTRACT A METHOD AND SYSTEM FOR IDENTIFYING GESTURE OF ONE OR MORE DISABLED PERSONS The present invention relates to a method and system for identifying gesture of one or more disabled persons. The method processing and analysing various types of input data, including audio, images, and text in language used by disabled people. The method starts with data collection, followed by preprocessing steps tailored input data to each data modules. Audio data is converted to text using the Pyaudio package, while images undergo resizing and splitting into training and testing sets. Text data is pre-processed using natural language processing techniques. Once the data is prepared, large learned modules LLM based on machine learning models are employed for analysis. Dictionary-based machine translation is used for text data, and convolutional neural networks (CNNs) are applied for image classification. The final output is a prediction or classification, which is visualized using a GIF file. FIG. 3
Disclaimer: Curated by HT Syndication.
