EfMAR: An Outdoor Mobile Augmented Reality Framework for Geospatial Measurements.

Area of Science:

• Computer Vision
• Robotics
• Geospatial Technology

Background:

• Accurate outdoor distance measurement is a significant challenge for mobile augmented reality (AR) systems.
• Existing AR solutions often lack robustness in large-scale or varied outdoor environments.

Purpose of the Study:

• To present EfMAR, an adaptive framework for robust outdoor mobile AR-based geospatial measurements.
• To formalize a reusable architectural model for adaptive multi-sensor fusion in AR.

Main Methods:

• EfMAR integrates visual SLAM, inertial sensing, depth information, and global positioning cues.
• A structured sensor fusion architecture is employed for adaptive multi-modal integration.
• A dedicated dataset with 584 real-world evaluation instances was collected.

Main Results:

• Hybrid sensor fusion strategies demonstrate more stable and consistent performance than single-modality solutions.
• Performance is particularly enhanced in challenging urban contexts.
• Environmental factors like lighting and spatial scale significantly influence measurement variability.

Conclusions:

• EfMAR provides a flexible and adaptive framework to enhance robustness in outdoor AR geospatial measurement.
• The framework supports reproducibility and future comparative research in mobile AR sensing.
• This work lays a practical foundation for real-world outdoor AR applications.