Cross-Domain Object Detection through Consistent and Contrastive Teacher with Fourier Transform

The teacher–student framework has been employed in unsupervised domain adaptation, which transfers knowledge learned from a labeled source domain to an unlabeled target domain. However, this framework suffers from two serious challenges: the domain gap, causing performance degradation, and noisy tea...

Full description

Saved in:
Bibliographic Details
Published in:Electronics (Basel) 2024-08, Vol.13 (16), p.3292
Main Authors: Jia, Longfei, Tian, Xianlong, Jing, Mengmeng, Zuo, Lin, Li, Wen
Format: Article
Language:English
Subjects:
Adaptation
Algorithms
Classification
Fourier transforms
Knowledge management
Labels
Learning
Localization
Methods
Noise control
Noise prediction
Object recognition
Performance degradation
Target detection
Teachers
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The teacher–student framework has been employed in unsupervised domain adaptation, which transfers knowledge learned from a labeled source domain to an unlabeled target domain. However, this framework suffers from two serious challenges: the domain gap, causing performance degradation, and noisy teacher pseudo-labels, which tend to mislead students. In this paper, we propose a Consistent and Contrastive Teacher with Fourier Transform (CCTF) method to address these challenges for high-performance cross-domain object detection. To mitigate the negative impact of domain shifts, we use the Fourier transform to exchange the low-frequency components of the source and target domain images, replacing the source domain inputs with the transformed image, thereby reducing domain gaps. In addition, we encourage the localization and classification branches of the teacher to make consistent predictions to minimize the noise in the generated pseudo-labels. Finally, contrastive learning is employed to resist the impact of residual noise in pseudo-labels. After extensive experiments, we show that our method achieves the best performance. For example, our model outperforms previous methods by 3.0% on FoggyCityscapes.
ISSN:2079-9292
2079-9292
DOI:10.3390/electronics13163292