Adaptive Object Detection for Indoor Navigation Assistance: A Performance Evaluation of Real-Time Algorithms

TL;DR

This paper tackles the challenge of reliable and efficient object detection for indoor navigation assistance for visually impaired users. It conducts a systematic, real-time performance comparison of four prominent detectors—YOLO, SSD, Faster R-CNN, and Mask R-CNN—on the Indoor Objects Detection dataset tailored for indoor navigation. The study provides concrete results on accuracy and speed, revealing that YOLOv5 offers the strongest real-time balance, with SSD as a viable alternative, while Faster R-CNN and Mask R-CNN trade some speed for higher precision. The findings offer actionable guidance for selecting detectors in real-time assistive navigation systems and highlight avenues for future integration with additional sensors and outdoor scenarios.

Abstract

This study addresses the need for accurate and efficient object detection in assistive technologies for visually impaired individuals. We evaluate four real-time object detection algorithms YOLO, SSD, Faster R-CNN, and Mask R-CNN within the context of indoor navigation assistance. Using the Indoor Objects Detection dataset, we analyze detection accuracy, processing speed, and adaptability to indoor environments. Our findings highlight the trade-offs between precision and efficiency, offering insights into selecting optimal algorithms for realtime assistive navigation. This research advances adaptive machine learning applications, enhancing indoor navigation solutions for the visually impaired and promoting accessibility.

Figures (2)

• Figure 1: Co-Occurrence Matrix of class object in Dataset
• Figure 2: Single stage YOLO architecture