Fourier-based augmentation with applications to domain generalization

Xu, Qinwei; Zhang, Yingze; Fan, Ziqing; Wang, Yanfeng; Wu, Yiyan; Zhang, Ya

doi:10.1016/j.patcog.2023.109474

Cited by 23 publications

(2 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…At the same time, the core domain-invariant semantics information remains unchanged in the generated image. Instead of swapping low-frequency amplitude components, Fourier augmented co-teacher (FACT) [22] and AmpMix [44] proposes to mix the whole amplitude spectrum with the MixUp [37] technique and achieves better generalization ability. By assigning pixelwise significance with Gaussian distribution and introducing pixel-wise disturbance in the amplitude spectrum, HCDG [23] proposes to highlight the core information in the center area of the image than the marginal area.…”

Section: A Fourier Transform For Domain Adaptationmentioning

confidence: 99%

Curriculum-Based Augmented Fourier Domain Adaptation for Robust Medical Image Segmentation

Wang

Islam

et al. 2024

IEEE Trans. Automat. Sci. Eng.

View full text Add to dashboard Cite

Accurate and robust medical image segmentation is fundamental and crucial for enhancing the autonomy of computer-aided diagnosis and intervention systems. Medical data collection normally involves different scanners, protocols, and populations, making domain adaptation (DA) a highly demanding research field to alleviate model degradation in the deployment site. To preserve the model performance across multiple testing domains, this work proposes the Curriculum-based Augmented Fourier Domain Adaptation (Curri-AFDA) for robust medical image segmentation. In particular, our curriculum learning strategy is based on the causal relationship of a model under different levels of data shift in the deployment phase, where the higher the shift is, the harder to recognize the variance. Considering this, we progressively introduce more amplitude information from the target domain to the source domain in the frequency space during the curriculum-style training to smoothly schedule the semantic knowledge transfer in an easier-to-harder manner. Besides, we incorporate the training-time chained augmentation mixing to help expand the data distributions while preserving the domain-invariant semantics, which is beneficial for the acquired model to be more robust and generalize better to unseen domains. Extensive experiments on two segmentation tasks of Retina and Nuclei collected from multiple sites and scanners suggest that our proposed method yields superior adaptation and generalization performance. Meanwhile, our approach proves to be more robust under various corruption types and increasing severity levels. In addition, we show our method is also beneficial in the domainadaptive classification task with skin lesion datasets. The code is available at https://github.com/lofrienger/Curri-AFDA.

show abstract

Section: A Fourier Transform For Domain Adaptationmentioning

confidence: 99%

Curriculum-Based Augmented Fourier Domain Adaptation for Robust Medical Image Segmentation

Wang

Islam

et al. 2024

IEEE Trans. Automat. Sci. Eng.

View full text Add to dashboard Cite

show abstract

“…Data augmentation-whether through synthetic images or augmented images-is a source of invariance and stochasticity in the data, 22 where the randomization of domain-dependent features is known to improve generalization to out-of-distribution instances. 23 While the task of training on synthetic data and predicting on real data is framed as a domain generalization or domain shift task, 24 mixed sample data augmentation (MSDA) is typically understood in the context of regularization. 8,14 The type of regularization contributed by patch-level (e.g.…”

Section: Data Augmentation As Regularization Techniquementioning

confidence: 99%

Mind the (domain) gap: metrics for the differences in synthetic and real data distributions

Dale,

Reindl,

Sanchez

et al. 2024

Synthetic Data for Artificial Intelligence and Machine Learning: Tools, Techniques, and Applications II

View full text Add to dashboard Cite

Synthetic data are frequently used to supplement a small set of real images and create a dataset with diverse features, but this may not improve the equivariance of a computer vision model. Our work answers the following questions: First, what metrics are useful for measuring a domain gap between real and synthetic data distributions? Second, is there an effective method for bridging an observed domain gap? We explore these questions by presenting a pathological case where the inclusion of synthetic data did not improve model performance, then presenting measurements of the difference between the real and synthetic distributions in the image space, latent space, and model prediction space. We find that augmenting the dataset with pixel-level augmentation effectively reduced the observed domain gap, and improves the model F1 score to 0.95 compared to 0.43 for un-augmented data. We also observe that an increase in the average cross entropy of the latent space feature vectors is positively correlated with increased model equivariance and the closing of the domain gap. The results are explained using a framework of model regularization effects.

show abstract

Semi-hard constraint augmentation of triplet learning to improve image corruption classification

Zhang,

Xiong,

Sun

et al. 2024

Vis Comput

View full text Add to dashboard Cite

Fourier-based augmentation with applications to domain generalization

Cited by 23 publications

References 9 publications

Curriculum-Based Augmented Fourier Domain Adaptation for Robust Medical Image Segmentation

Curriculum-Based Augmented Fourier Domain Adaptation for Robust Medical Image Segmentation

Mind the (domain) gap: metrics for the differences in synthetic and real data distributions

Semi-hard constraint augmentation of triplet learning to improve image corruption classification

Contact Info

Product

Resources

About