MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202641048611 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on April 16, for 'system and method for real-time indoor navigation and hazard detection using multi-modal vision.'
Inventor(s) include Archana Tamizharasan; and Siddharth Y.
The application for the patent was published on May 1, under issue no. 18/2026.
According to the abstract released by the Intellectual Property India: "The present disclosure proposes a system (100) for real-time indoor navigation and hazard detection for visually impaired users. The system (100) comprises a computing device (102) coupled with a capturing unit (108) for acquiring real-time visual data. An input module (112) generates processed image data, which is analyzed by a perception module (114) using integrated multi-stage perception comprises instance segmentation, depth estimation, and structural scene analysis to obtain perception data. A hazard detection and risk evaluation module (116) computes hazard risk based on object, depth, and structural features. A mapping module (118) generates a navigation grid identifying safe and unsafe regions. A decision module (120) determines navigation directions, and an instruction generation module (122) produces context-aware navigation instructions. An output module (124) delivers real-time voice guidance, thereby improving navigation safety, environmental awareness, and response efficiency."
Disclaimer: Curated by HT Syndication.
