通过模拟域偏移并重聚焦注意力，提升模型在未知域上的泛化能力。
Simulate, Refocus and Ensemble: An Attention-Refocusing Scheme for Domain Generalization
- 用数据增强模拟目标域，生成伪目标数据以模拟域偏移。
- 通过注意力重聚焦使模型关注跨域不变特征，显著提升性能。
- 集成学习策略帮助捕捉更稳定的域不变注意力图，适合图像分类任务。
领域泛化（DG）旨在从源域学习模型，并应用于包含分布外数据的未见目标域。由于CLIP强大的语义编码能力，其在领域泛化中备受关注。然而，CLIP常难以在不同域间聚焦于任务相关区域（即域不变区域），导致在未见目标域上表现不佳。为此，本文提出一种注意力重聚焦方案SRE（Simulate, Refocus and Ensemble），通过注意力重聚焦对齐CLIP中的注意力图以减小域差异。SRE首先通过源数据增强生成模拟目标域；然后在源域与模拟目标域间学习注意力重聚焦以减少域偏移；最后利用集成学习增强捕捉源数据与模拟目标数据间域不变注意力图的能力。在多个数据集上的大量实验表明，SRE普遍优于现有最先进方法。代码已开源：https://github.com/bitPrincy/SRE-DG。
原文摘要 · Abstract (English)
Domain generalization (DG) aims to learn a model from source domains and apply it to unseen target domains with out-of-distribution data. Owing to CLIP's strong ability to encode semantic concepts, it has attracted increasing interest in domain generalization. However, CLIP often struggles to focus on task-relevant regions across domains, i.e., domain-invariant regions, resulting in suboptimal performance on unseen target domains. To address this challenge, we propose an attention-refocusing scheme, called Simulate, Refocus and Ensemble (SRE), which learns to reduce the domain shift by aligning the attention maps in CLIP via attention refocusing. SRE first simulates domain shifts by performing augmentation on the source data to generate simulated target domains. SRE then learns to reduce the domain shifts by refocusing the attention in CLIP between the source and simulated target domains. Finally, SRE utilizes ensemble learning to enhance the ability to capture domain-invariant attention maps between the source data and the simulated target data. Extensive experimental results on several datasets demonstrate that SRE generally achieves better results than state-of-the-art methods. The code is available at: https://github.com/bitPrincy/SRE-DG.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。