A Privacy-Preserving Approach for Multi-Source Domain Adaptive Object Detection

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

*此作品的通信作者

研究成果: Conference contribution同行評審

摘要

A new research topic, multi-source domain adaptive object detection (MSDAOD) under privacy-preserving constraint is explored in this paper, where the clients can only access their own source data while the server can only access unlabeled target data. Accordingly, a novel MSDAOD framework is proposed wherein the clients employ a source-only Probabilistic Faster R-CNN (PFRCNN) to generate models with localization uncertainty, while a Multi-teacher Pseudo-label Ensemble Network (MPEN) is developed on the server side. In MPEN, FedMA-based algorithm aggregates the above models to a domain-invariant backbone while a novel pseudo-label ensemble (PLE) scheme is employed to reduce false positives arising from domain specific parts, and enhance the overall system performance using target domain information. Experiments demonstrate that our method outperforms other state-of-the-art MSDAOD and privacy-preserving methods by 10%~16% in average precision (AP).

原文English
主出版物標題2023 IEEE International Conference on Image Processing, ICIP 2023 - Proceedings
發行者IEEE Computer Society
頁面1075-1079
頁數5
ISBN(電子)9781728198354
DOIs
出版狀態Published - 2023
事件30th IEEE International Conference on Image Processing, ICIP 2023 - Kuala Lumpur, Malaysia
持續時間: 8 10月 202311 10月 2023

出版系列

名字Proceedings - International Conference on Image Processing, ICIP
ISSN(列印)1522-4880

Conference

Conference30th IEEE International Conference on Image Processing, ICIP 2023
國家/地區Malaysia
城市Kuala Lumpur
期間8/10/2311/10/23

指紋

深入研究「A Privacy-Preserving Approach for Multi-Source Domain Adaptive Object Detection」主題。共同形成了獨特的指紋。

引用此