A Multi-Modal Deep Learning Framework for Assistive Vision: Software-Based Design and Real-Time Validation

Author(s)	Mr. Sri Saagar G N, Dr. Manjula R Chougala
Country	India
Abstract	Visual impairment greatly affects a person’s ability to perceive and navigate their surroundings. The conventional white cane is a basic mobility aid that does not provide any context and lacks intelligent scene understanding. This research paper outlines the development of an intelligent visual aid incorporating object detection, face recognition, distance estimation, and text-to-speech audio feedback. Deep learning models such as YOLOv5n and MobileFaceNet are implemented to ensure efficient performance on devices with limited computational resources. A combination of monocular and ultrasonic distance estimation techniques provides additional awareness of the environment. The system provides structured information about the scene and guidance through navigation using natural language processing and off-line text-to-speech conversion technology. Our experiments yielded performance at 15-20 FPS and accurate detections of objects in several cases.
Keywords	Assistive vision, Edge AI, Object Detection, Depth Estimation, Face Recognition, NLP, Real-time Systems
Field	Engineering
Published In	Volume 8, Issue 4, July-August 2026
Published On	2026-07-29

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research