PIT: A Plug-and-Play Image Translator for Making Off-the-Shelf Models Adapt to Corruptions

doi:10.1109/TPAMI.2025.3598147

CSpace

	PIT: A Plug-and-Play Image Translator for Making Off-the-Shelf Models Adapt to Corruptions
	Li, Yinqi 1,2; Chang, Hong 1,2; Shan, Shiguang 1,2; Chen, Xilin 1,2
	2025-12-01
发表期刊	IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
ISSN	0162-8828
卷号	47 期号:12 页码:11644-11661
摘要	Visual recognition models pretrained on clean images usually do not perform well in the presence of image corruptions, such as blurring or noise, which limits their applicability in real-world scenarios. To solve this problem, existing approaches usually design complex data augmentations to train a robust model from scratch or adapt a pretrained model to corrupted scenarios. These approaches ignore the existence of the large number of deployed models in our community, causing extensive computation and storage costs for making deployed models adapted. Based on this consideration, this paper focuses on solving a practical problem of making many clean-image-pretrained models adapt to unlabeled corrupted images through one training procedure. To this end, we aim to learn a Plug-and-play Image Translator (PIT) that can be directly combined with recognition models after training. Existing approaches, such as vanilla image translation and restoration, are not proper for solving this problem, as they are mostly based on supervised training and are not recognition-oriented. To address this issue, we propose a recognition-oriented unsupervised image translation framework to make PIT produce images with indistinguishable recognition predictions from the clean ones. We verify the effectiveness of PIT on several recognition tasks and show that PIT boosts the performance of clean-image-pretrained models significantly in the presence of image corruptions.
关键词	Adaptation models Image recognition Computational modeling Translation Training Data models Image restoration Data augmentation Head Flowering plants Corrupted image recognition image-to-image translation generative adversarial network
DOI	10.1109/TPAMI.2025.3598147
收录类别	SCI
语种	英语
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS记录号	WOS:001609561600050
出版者	IEEE COMPUTER SOC
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/43100
专题	中国科学院计算技术研究所
通讯作者	Chang, Hong
作者单位	1.Chinese Acad Sci, State Key Lab AI Safety, Inst Comp Technol, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Beijing 100049, Peoples R China
推荐引用方式 GB/T 7714	Li, Yinqi,Chang, Hong,Shan, Shiguang,et al. PIT: A Plug-and-Play Image Translator for Making Off-the-Shelf Models Adapt to Corruptions[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,2025,47(12):11644-11661.
APA	Li, Yinqi,Chang, Hong,Shan, Shiguang,&Chen, Xilin.(2025).PIT: A Plug-and-Play Image Translator for Making Off-the-Shelf Models Adapt to Corruptions.IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,47(12),11644-11661.
MLA	Li, Yinqi,et al."PIT: A Plug-and-Play Image Translator for Making Off-the-Shelf Models Adapt to Corruptions".IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 47.12(2025):11644-11661.