FedDAD: Federated Domain Adaptation for Object Detection

Peggy Joy Lu*, Chia Yung Jui, Jen Hui Chuang

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

Training an object detection model often requires numerous annotated images on a centralized host, which may violate user privacy and data confidentiality. Federated learning (FL) resolves this issue by allowing multiple clients, e.g., cameras, to collaboratively train a model while protecting user privacy. However, models trained with FL may fail to be generalized for new target domain due to domain shift when the data between source and target domains are statistically different. In this work, we formulate a real-world object detection problem as a source-free multi-domain adaptation problem in FL architecture. Moreover, we propose an adaptive FL algorithm, called FedDAD (Federated Domain Adaptive Detector), which aggregates models with dynamic attention targeting the unsupervised domain on server, and utilize instance-level alignment to alleviate the effects of scene variation on clients. Experimental results show that FedDAD improves the average precision (AP) by up to 10.05% and 19.15% compared to the popular FedAvg for specific object classes in the KAIST and MI3 datasets, respectively.

Original languageEnglish
Pages (from-to)51320-51330
Number of pages11
JournalIEEE Access
Volume11
DOIs
StatePublished - 2023

Keywords

  • federated learning
  • multi-source domain adaptation
  • object detection
  • Unsupervised domain adaptation

Fingerprint

Dive into the research topics of 'FedDAD: Federated Domain Adaptation for Object Detection'. Together they form a unique fingerprint.

Cite this