Related Experiment Video
Updated: Jun 21, 2025
08:25
Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
Published on: May 7, 2019
9.0K
ARAware: Assisting Visually Impaired People with Real-Time Critical Moving Object Identification
Hadeel Surougi1, Cong Zhao2, Julie A McCann1
1Department of Computing, Imperial College London, London SW7 2AZ, UK.
Sensors (Basel, Switzerland)
|July 13, 2024
Summary
This study introduces ARAware, a camera-based system for visually impaired people (VIPs) that identifies critical moving objects (CMOs) in real-time. ARAware enhances VIP safety by accurately detecting and prioritizing high-risk threats, improving mobility assistance.
Area of Science:
- Computer Vision
- Assistive Technology
- Robotics
Background:
- Visually impaired people (VIPs) face significant safety risks from autonomous outdoor moving objects.
- Existing camera-based mobility solutions struggle to detect high-speed threats, termed Critical Moving Objects (CMOs).
Purpose of the Study:
- To develop and evaluate ARAware, a novel camera-based system for real-time identification and risk assessment of CMOs for VIPs.
- To improve the safety and independence of visually impaired individuals by providing timely warnings of potential dangers.
Main Methods:
- The ARAware system integrates CMO identification, real-time risk level evaluation and classification, and prioritized warning notifications.
- A real-world prototype was used for experimental validation, processing video at 32 fps.
Main Results:
- ARAware achieved high accuracy in CMO identification (97.26% mAR, 88.20% mAP) and risk classification (100% mAR, 91.69% mAP).
- The system provides timely warnings for high-risk CMOs while minimizing false alarms.
- ARAware demonstrated superior performance over the DEEP-SEE approach, with 42.62% higher mAR and 10.88% higher mAP in CMO identification, and a 93% faster processing speed.
Conclusions:
- ARAware represents the first practical camera-based mobility assistant scheme that effectively addresses the challenge of CMO detection for VIPs.
- The system significantly enhances safety by providing accurate, real-time threat assessment and prioritized warnings, outperforming existing methods.
