Intellectual Property India Publishes Patent Application for A System And Method For Real-Time Sign Language Translation Using Media Pipe’s Landmark-Based Gesture Recognition, Lstm Deep Learning And Llm Api Integration

MUMBAI, India, Aug. 10 -- Intellectual Property India has published a patent application (202611074471 A) filed by Manipal University Jaipur on June 16, 2026, for A System And Method For Real-Time Sign Language Translation Using Media Pipe’s Landmark-Based Gesture Recognition, Lstm Deep Learning And Llm Api Integration.

Inventors include Aryaman Sharma; Adya Kumari; and Dr. Rahul Sharma.

The application for the patent was published on July 31, 2026, under issue no. 31/2026.

Abstract: The present invention relates to a real-time sign language translation system for converting sign language gestures into natural-language sentences. The system comprises a camera, a landmark extraction module, a sequence processing module, a gesture recognition module, a confidence evaluation module, a language refinement module, and an output module. The camera captures image frames containing hand gestures. The landmark extraction module identifies hand landmarks and generates landmark data for each frame. The gesture recognition module employ long short-term memory neural network to recognise dynamic sign language gestures and generate gesture tokens. The confidence evaluation module removes low-confidence predictions. The language refinement module processes the gesture tokens and generates grammatically correct natural-language sentences. The system provides an accessible and cost-effective sign language communication solution.

Disclaimer: Curated by HT Syndication.