Attribute recognition for person re-identification using federated learning at all-in-edge
- Shini Girija,
- Thar Baker(corresponding author),
- Naveed Ahmed,
- Ahmed M. Khedr,
- Zaher Al Aghbari,
- Ashish Jha
- University of Sharjah,
- University of Brighton,
- Skolkovo Institute of Science and Technology
Research Output:
Contribution to journal
Article
Research
Peer-reviewOpen access
Abstract
<p>The advancement in person re-identification using attribute recognition is constrained by the increasingly strict data privacy standards since it necessitates the centralization of vast amounts of data containing sensitive personal data in the cloud. Cloud-based person re-identification requires the transfer of original video information to the servers, causing increased communication costs because of the need for significant bandwidth, resulting in unpredictable timing. This work presents an all-in-edge architecture for attribute-based person re-identification, which deploys training data in edge nodes that support distributed inference. Edge nodes independently learn but collaborate with specific neighboring nodes by sharing information to minimize communication and computational costs through the utilization of federated learning and transfer learning methods. Furthermore, this paper proposes a federated aggregation strategy-FedTransferLoss to obtain optimal global accuracy by using transfer learning to re-train the low-quality local models. Extensive experiments on two prominent pedestrian datasets-PETA and RAP show that FedTransferLoss achieves higher accuracy, recall and precision values compared to the traditional FedAvg algorithm.</p>
Research Output Information
Output type
Research Output:
Contribution to journal
Article
Research
Peer-reviewOriginal language
EnglishArticle number
100793Journal (Volume, Issue Number)
Internet of Things (Netherlands) (Volume 22)Publication milestones
- Published - 07/2023
Publication status
Published - 07/2023
Publication IDs
- Scopus: 85154037812
Documents and links
Publication metrics
Metrics
SciVal
FWCI
0.46
PlumX, opens in new tab
Captures
17
Citation count
7
