MUMBAI, India, July 7 -- Intellectual Property India has published a patent application (202641079765 A) filed by Sona College Of Technology on June 29, 2026, for Ai-Based Multimodal Object Detection And Contextual Guidance Assistant For Visually Impaired Users.
Inventors include Sasirekha K; Shanthini K; and Divyaraj M.
The application for the patent was published on July 03, 2026, under issue no. 27/2026.
Abstract: ABSTRACT The present invention relates to an artificial intelligence-based multimodal assistive vision system for visually impaired users. The system comprises a camera module (102), a preprocessing module (104), an object perception module (106), a depth estimation module (108), a spatial reasoning module (110), a text extraction module (112), a social context analysis module (114), a multimodal contextual fusion engine (116), and an audio guidance module (118). The system captures real-time environmental images and simultaneously performs object detection, distance estimation, directional analysis, text recognition, gesture identification, and emotion recognition. The multimodal contextual fusion engine integrates outputs from the various perception modules to generate a unified contextual scene representation. The generated information is converted into multilingual spoken guidance through the audio guidance module. The invention enhances environmental awareness, navigation safety, social interaction, and independent mobility by providing context-aware real-time auditory assistance to visually impaired users using a unified multimodal perception framework.
Disclaimer: Curated by HT Syndication.
