eduzhai > Applied Sciences > Engineering >

An Edge Computing-based Photo Crowdsourcing Framework for Real-time 3D Reconstruction

  • king
  • (0) Download
  • 20210505
  • Save

... pages left unread,continue reading

Document pages: 12 pages

Abstract: Image-based three-dimensional (3D) reconstruction utilizes a set of photos tobuild 3D model and can be widely used in many emerging applications such asaugmented reality (AR) and disaster recovery. Most of existing 3Dreconstruction methods require a mobile user to walk around the target area andreconstruct objectives with a hand-held camera, which is inefficient andtime-consuming. To meet the requirements of delay intensive and resource hungryapplications in 5G, we propose an edge computing-based photo crowdsourcing(EC-PCS) framework in this paper. The main objective is to collect a set ofrepresentative photos from ubiquitous mobile and Internet of Things (IoT)devices at the network edge for real-time 3D model reconstruction, with networkresource and monetary cost considerations. Specifically, we first propose aphoto pricing mechanism by jointly considering their freshness, resolution anddata size. Then, we design a novel photo selection scheme to dynamically selecta set of photos with the required target coverage and the minimum monetarycost. We prove the NP-hardness of such problem, and develop an efficientgreedy-based approximation algorithm to obtain a near-optimal solution.Moreover, an optimal network resource allocation scheme is presented, in orderto minimize the maximum uploading delay of the selected photos to the edgeserver. Finally, a 3D reconstruction algorithm and a 3D model caching schemeare performed by the edge server in real time. Extensive experimental resultsbased on real-world datasets demonstrate the superior performance of our EC-PCSsystem over the existing mechanisms.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×