MUMBAI, India, Jan. 23 -- Intellectual Property India has published a patent application (202521123094 A) filed by Mr. Kiran C. Kulkarni; and Dr. Manoj A. Wakchaure, Pune, Maharashtra, on Dec. 6, 2025, for 'system and method for an efficient multi-model framework for sign language recognition.'
Inventor(s) include Mr. Kiran C. Kulkarni; and Dr. Manoj A. Wakchaure.
The application for the patent was published on Jan. 23, under issue no. 04/2026.
According to the abstract released by the Intellectual Property India: "System and Method For an Efficient Multi-Model Framework For Sign Language Recognition This is a new technology that allows for the recognition of different sign languages using several different things, including gestures of hands; the manner in which a gesture is made, how to move objects from the hand to the lips when signing, and other points in the hand(s) as they sign (where it is pointed and held on the body); where the hand is when you are signing, within the environment where the signing is supposed to take place; and to what degree the fingers are bent to indicate what letters are being signed. This has been accomplished by creating a unique type of sensor that captures the physical characteristics of a signer. Then, that information is processed and combined through the use of an adaptive neural network (a neural model) to produce data that can be recognised through various forms of input, either by the use of cameras or by human interpretation of the completed video. After this process has been completed, the signing system will then have an intelligence engine that evaluates whether a given gesture is a word and what type of word it represents. The intelligence engine will also determine the relationship between time and the connection between a signed gesture and the visual appearance of the object being signed and that the signed gesture and its visual appearance are aligned."
Disclaimer: Curated by HT Syndication.
