TY - JOUR
T1 - Smooth-Guided Implicit Data Augmentation for Domain Generalization
AU - Wang, Mengzhu
AU - Liu, Junze
AU - Luo, Ge
AU - Wang, Shanshan
AU - Wang, Wei
AU - Lan, Long
AU - Wang, Ye
AU - Nie, Feiping
N1 - Publisher Copyright:
IEEE
PY - 2024
Y1 - 2024
N2 - The training process of a domain generalization (DG) model involves utilizing one or more interrelated source domains to attain optimal performance on an unseen target domain. Existing DG methods often use auxiliary networks or require high computational costs to improve the model’s generalization ability by incorporating a diverse set of source domains. In contrast, this work proposes a method called Smooth-Guided Implicit Data Augmentation (SGIDA) that operates in the feature space to capture the diversity of source domains. To amplify the model’s generalization capacity, a distance metric learning (DML) loss function is incorporated. Additionally, rather than depending on deep features, the suggested approach employs logits produced from cross entropy (CE) losses with infinite augmentations. A theoretical analysis shows that logits are effective in estimating distances defined on original features, and the proposed approach is thoroughly analyzed to provide a better understanding of why logits are beneficial for DG. Moreover, to increase the diversity of the source domain, a sampling-based method called smooth is introduced to obtain semantic directions from interclass relations. The effectiveness of the proposed approach is demonstrated through extensive experiments on widely used DG, object detection, and remote sensing datasets, where it achieves significant improvements over existing state-of-the-art methods across various backbone networks.
AB - The training process of a domain generalization (DG) model involves utilizing one or more interrelated source domains to attain optimal performance on an unseen target domain. Existing DG methods often use auxiliary networks or require high computational costs to improve the model’s generalization ability by incorporating a diverse set of source domains. In contrast, this work proposes a method called Smooth-Guided Implicit Data Augmentation (SGIDA) that operates in the feature space to capture the diversity of source domains. To amplify the model’s generalization capacity, a distance metric learning (DML) loss function is incorporated. Additionally, rather than depending on deep features, the suggested approach employs logits produced from cross entropy (CE) losses with infinite augmentations. A theoretical analysis shows that logits are effective in estimating distances defined on original features, and the proposed approach is thoroughly analyzed to provide a better understanding of why logits are beneficial for DG. Moreover, to increase the diversity of the source domain, a sampling-based method called smooth is introduced to obtain semantic directions from interclass relations. The effectiveness of the proposed approach is demonstrated through extensive experiments on widely used DG, object detection, and remote sensing datasets, where it achieves significant improvements over existing state-of-the-art methods across various backbone networks.
KW - Computational modeling
KW - Data augmentation
KW - Data models
KW - Distance metric learning (DML)
KW - Germanium
KW - Semantics
KW - Task analysis
KW - Training
KW - domain generalization (DG)
KW - implicit data augmentation
KW - smooth
UR - http://www.scopus.com/inward/record.url?scp=85192203344&partnerID=8YFLogxK
U2 - 10.1109/TNNLS.2024.3377439
DO - 10.1109/TNNLS.2024.3377439
M3 - 文章
AN - SCOPUS:85192203344
SN - 2162-237X
SP - 1
EP - 12
JO - IEEE Transactions on Neural Networks and Learning Systems
JF - IEEE Transactions on Neural Networks and Learning Systems
ER -