EAR: An Edge-Assisted and Energy-Efficient Mobile Augmented Reality Framework

Abstract

Mobile Augmented Reality (MAR) apps may cause short battery life due to high-quality virtual objects rendered in the augmented environment. Given the limited ability of the human eye to recognize small details at larger distances, using (lower-quality) decimated virtual objects with a lower triangle count can still provide high user experience but at lower energy consumption. Current state-of-the-art solutions may impose a high burden on the MAR app developer, limited energy savings, and high storage overhead. In this paper, we propose eAR, an edge-assisted autonomous and energy-efficient framework for MAR apps designed to solve the limitations of state-of-the-art solutions. eAR features an offline software running on an edge server that uses Image Quality Assessment (IQA) to model user-perceived quality for each virtual object in terms of triangle count and user-object distance without requiring any effort from the MAR developer. In addition, eAR features a runtime lightweight optimization algorithm that minimizes storage overhead by dynamically deciding the most energy-efficient virtual object decimation ratios to request from the edge server. Our results show that eAR can help reduce energy consumption by up to 16.5% while reducing storage overhead by almost 60% compared to existing schemes, with minimal impact on user-perceived quality.