<italic>FedVision:</italic> Federated Video Analytics With Edge Computing
Widely deployed smart cameras are generating a large amount of video data and capable of processing frames on devices. Empowered by edge computing, the video data can also be offloaded to edge servers for processing. By leveraging the on-device processing and computation offloading, we propose a fed...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Open Journal of the Computer Society |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9097917/ |
| _version_ | 1818619318803365888 |
|---|---|
| author | Yang Deng Tao Han Nirwan Ansari |
| author_facet | Yang Deng Tao Han Nirwan Ansari |
| author_sort | Yang Deng |
| collection | DOAJ |
| description | Widely deployed smart cameras are generating a large amount of video data and capable of processing frames on devices. Empowered by edge computing, the video data can also be offloaded to edge servers for processing. By leveraging the on-device processing and computation offloading, we propose a federated video analytics system named FedVision to efficiently provision video analytics across devices and servers. The challenge of designing FedVision is to optimally use the computing and networking resources for video analytics. Since there is no closed-form expression of the system performance, black-box optimization is employed to optimize the system performance. However, using black-box optimization directly incurs excessive system queries that lead to very poor system performance. To solve this problem, we design a new optimization method that integrates black-box optimization with Neural Processes (NPs) as a system performance approximator. This method allows black-box optimizer to query NPs instead of the real system. We validate the performance of FedVision and the new optimization method using both numerical results and experiments with a testbed. |
| first_indexed | 2024-12-16T17:35:35Z |
| format | Article |
| id | doaj.art-7d2d17ce78f14e1a9769cfa7fc7ad366 |
| institution | Directory Open Access Journal |
| issn | 2644-1268 |
| language | English |
| last_indexed | 2024-12-16T17:35:35Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Open Journal of the Computer Society |
| spelling | doaj.art-7d2d17ce78f14e1a9769cfa7fc7ad3662022-12-21T22:22:47ZengIEEEIEEE Open Journal of the Computer Society2644-12682020-01-011627210.1109/OJCS.2020.29961849097917<italic>FedVision:</italic> Federated Video Analytics With Edge ComputingYang Deng0https://orcid.org/0000-0002-4077-409XTao Han1https://orcid.org/0000-0002-6626-1305Nirwan Ansari2https://orcid.org/0000-0001-8541-3565Department of Electrical and Computer Engineering, The University of North Carolina at Charlotte, Charlotte, NC, USADepartment of Electrical and Computer Engineering, The University of North Carolina at Charlotte, Charlotte, NC, USAHelen and John C. Hartmann Department of Electrical and Computer Engineering, New Jersey Institute of Technology, Newark, NJ, USAWidely deployed smart cameras are generating a large amount of video data and capable of processing frames on devices. Empowered by edge computing, the video data can also be offloaded to edge servers for processing. By leveraging the on-device processing and computation offloading, we propose a federated video analytics system named FedVision to efficiently provision video analytics across devices and servers. The challenge of designing FedVision is to optimally use the computing and networking resources for video analytics. Since there is no closed-form expression of the system performance, black-box optimization is employed to optimize the system performance. However, using black-box optimization directly incurs excessive system queries that lead to very poor system performance. To solve this problem, we design a new optimization method that integrates black-box optimization with Neural Processes (NPs) as a system performance approximator. This method allows black-box optimizer to query NPs instead of the real system. We validate the performance of FedVision and the new optimization method using both numerical results and experiments with a testbed.https://ieeexplore.ieee.org/document/9097917/Edge computingblack-box optimizationneural processmachine learningvideo analytics |
| spellingShingle | Yang Deng Tao Han Nirwan Ansari <italic>FedVision:</italic> Federated Video Analytics With Edge Computing IEEE Open Journal of the Computer Society Edge computing black-box optimization neural process machine learning video analytics |
| title | <italic>FedVision:</italic> Federated Video Analytics With Edge Computing |
| title_full | <italic>FedVision:</italic> Federated Video Analytics With Edge Computing |
| title_fullStr | <italic>FedVision:</italic> Federated Video Analytics With Edge Computing |
| title_full_unstemmed | <italic>FedVision:</italic> Federated Video Analytics With Edge Computing |
| title_short | <italic>FedVision:</italic> Federated Video Analytics With Edge Computing |
| title_sort | italic fedvision italic federated video analytics with edge computing |
| topic | Edge computing black-box optimization neural process machine learning video analytics |
| url | https://ieeexplore.ieee.org/document/9097917/ |
| work_keys_str_mv | AT yangdeng italicfedvisionitalicfederatedvideoanalyticswithedgecomputing AT taohan italicfedvisionitalicfederatedvideoanalyticswithedgecomputing AT nirwanansari italicfedvisionitalicfederatedvideoanalyticswithedgecomputing |