MUMBAI, India, Oct. 5 -- Intellectual Property India has published a patent application (202641115766 A) filed by Srm Institute Of Science And Technology, Ramapuram Campus; and Easwari Engineering College on September 28, 2026, for Third Eye : An Intelligent Vision Assistant For The Blind.
Inventors include V Krishnakumar; Manish Stephen K; Ranesh Karthik B; Dr. R. Angeline; Dr. A. Senthilselvi; and Dr. V. Jananee.
The application for the patent was published on October 02, 2026, under issue no. 40/2026.
Abstract: Description: Background of the Invention: The model relates to an offline, AI enabled smart glasses system for facilitating a visually impaired user. It enables real-time navigation, obstacle deduction, and OCR implementation of edge computing excluding reliability on cloud services. Summary of the Invention: The current invention focuses to provide a complete offline, wearable facilitative system for visually impaired user that overcomes latency, privacy, and connectivity issues of cloud-based solutions. The system is configured to use independently of internet access while ensuring real-time solution and power efficiency via on-device AI. The model includes a standalone wearable device, ideally implemented as smart glasses, along with a forward facing camera, ultrasonic sensors, a local processing unit, tactile control buttons, and a bone-conduction audio output mechanism. All sensing and processing functions are performed locally on the device using edge computing. A key aspect of the invention is a dual-pipeline processing model. The always-on safety pipeline operates continuously in a low-energy mode to recognize hazards like obstacles and moving vehicles using light weight models and sensor fusion. By ensuring resource intensive processes dormant until accepted, the system automatically decreases power utilization while ensuring faster response for safety-critical moments. To sum up, the invention facilitates user privacy by mitigating data transmission to external servers and provides situational knowledge via bone-conduction audio that upholds ambient sound perception. The outcome is trustworthy, energy-efficient, and privacy-centric assistive solution usable in all environments, including areas that lacks network availability.
Disclaimer: Curated by HT Syndication.
