An application performance optimization model of mobile augmented reality based on hd restoration
Authors:
- 1 December 2020
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 201-206
Abstract
In order to realize the energy consumption of video transmission in mobile augmented reality, a content adaptive image uninstall model is proposed. Taking into account the impact of limited uplink bandwidth resources on transmission latency, the model based on the complexity of shooting content and user expectations, low-resolution video sampling and upload to the cloud, and in the cloud using ultra-resolution algorithm for high-definition restoration of key frame images to ensure the accuracy of target recognition and tracking, and provide users with a high-quality viewing experience. In this paper, 289 videos are analyzed for complexity, machine learning models are constructed to select suitable image sample configurations, and low-resolution images are uploaded to reduce upload delays. The experimental results show that the model reduces upload by 77% to 82%.Keywords
Funding Information
- Natural Science Basic Research Program of Shaanxi Province
This publication has 6 references indexed in Scilit:
- Camel: Smart, Adaptive Energy Optimization for Mobile Web Interactions2020
- EDVR: Video Restoration With Enhanced Deformable Convolutional Networks2019
- Proteus2018
- To Compress, or Not to Compress: Characterizing Deep Learning Model Compression for Embedded Inference2018
- Understanding the Characteristics of Mobile Augmented Reality Applications2018
- On the Networking Challenges of Mobile Augmented Reality2017