期刊
NEUROCOMPUTING
卷 474, 期 -, 页码 48-59出版社
ELSEVIER
DOI: 10.1016/j.neucom.2021.12.009
关键词
Object detection; Domain adaptation; Domain adaptive object detection
资金
- National Key R&D Program of China [2018YFC 0309400]
- National Natural Science Foundation of China [61901160, 61871188, 61801133]
- Guang zhou city science and technology research projects [201902020008]
Domain adaptive object detection achieves performance improvement by constructing a transferable model for unlabeled target images and utilizing well-labeled source images with different distributions. However, most current methods overlook two crucial factors: 1) different areas of an image should not be equally aligned because some areas may contribute more to distribution alignment if they contain more discriminative information for classifying the objects; and 2) the objectives of feature alignment and classification should not be independently optimized as it fails to capture the discriminative information of data. To address these issues, a new domain adaptive object detection model is proposed.
Domain adaptive object detection has achieved appealing performance by constructing an effective transferable model for unlabeled target images, which takes advantage of the well-labeled source images with different distributions. However, two crucial factors are overlooked by most current methods: 1) different areas of an image should not be equally aligned since some areas may contribute more to distribution alignment if they contain more discriminative information for classifying the objects; and 2) the objectives of feature alignment and classification should not be independently optimized since it will fail to capture the discriminative information of data. To address these issues, we propose a new domain adaptive object detection model, referred to as discriminative distribution alignment domain adaptive detector. To be specific, the proposed method first makes the model focus on the areas that are quantified with high localization probability at the image level to enhance discrimination between foregrounds and backgrounds. Then the source and target images are aligned at the category level to learn class-invariant features by two adversarial regions-of-interest classifiers. Comprehensive experiments on several visual tasks verify that the proposed method outperforms the competitive domain adaptive object detection methods significantly in unsupervised domain adaptation setting. (c) 2021 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据