Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

ALOFT: A Lightweight MLP-like Architecture with Dynamic Low-frequency Transform for Domain Generalization

About

Domain generalization (DG) aims to learn a model that generalizes well to unseen target domains utilizing multiple source domains without re-training. Most existing DG works are based on convolutional neural networks (CNNs). However, the local operation of the convolution kernel makes the model focus too much on local representations (e.g., texture), which inherently causes the model more prone to overfit to the source domains and hampers its generalization ability. Recently, several MLP-based methods have achieved promising results in supervised learning tasks by learning global interactions among different patches of the image. Inspired by this, in this paper, we first analyze the difference between CNN and MLP methods in DG and find that MLP methods exhibit a better generalization ability because they can better capture the global representations (e.g., structure) than CNN methods. Then, based on a recent lightweight MLP method, we obtain a strong baseline that outperforms most state-of-the-art CNN-based methods. The baseline can learn global structure representations with a filter to suppress structure irrelevant information in the frequency space. Moreover, we propose a dynAmic LOw-Frequency spectrum Transform (ALOFT) that can perturb local texture features while preserving global structure features, thus enabling the filter to remove structure-irrelevant information sufficiently. Extensive experiments on four benchmarks have demonstrated that our method can achieve great performance improvement with a small number of parameters compared to SOTA CNN-based DG methods. Our code is available at https://github.com/lingeringlight/ALOFT/.

Jintao Guo, Na Wang, Lei Qi, Yinghuan Shi• 2023

Related benchmarks

TaskDatasetResultRank
Image ClassificationPACS (test)
Average Accuracy91.58
254
Image ClassificationDigits-DG leave-one-domain-out
Average Accuracy90.93
81
Domain GeneralizationOfficeHome (leave-one-domain-out)
Art Accuracy73.3
59
Domain GeneralizationVLCS (leave-one-domain-out)
Avg Acc81.31
22
Showing 4 of 4 rows

Other info

Code

Follow for update