Domain Generalization with MixStyle
- URL: http://arxiv.org/abs/2104.02008v1
- Date: Mon, 5 Apr 2021 16:58:09 GMT
- Title: Domain Generalization with MixStyle
- Authors: Kaiyang Zhou and Yongxin Yang and Yu Qiao and Tao Xiang
- Abstract summary: Domain generalization aims to address this problem by learning from a set of source domains a model that is generalizable to any unseen domain.
Our method, termed MixStyle, is motivated by the observation that visual domain is closely related to image style.
MixStyle fits into mini-batch training perfectly and is extremely easy to implement.
- Score: 120.52367818581608
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Though convolutional neural networks (CNNs) have demonstrated remarkable
ability in learning discriminative features, they often generalize poorly to
unseen domains. Domain generalization aims to address this problem by learning
from a set of source domains a model that is generalizable to any unseen
domain. In this paper, a novel approach is proposed based on probabilistically
mixing instance-level feature statistics of training samples across source
domains. Our method, termed MixStyle, is motivated by the observation that
visual domain is closely related to image style (e.g., photo vs.~sketch
images). Such style information is captured by the bottom layers of a CNN where
our proposed style-mixing takes place. Mixing styles of training instances
results in novel domains being synthesized implicitly, which increase the
domain diversity of the source domains, and hence the generalizability of the
trained model. MixStyle fits into mini-batch training perfectly and is
extremely easy to implement. The effectiveness of MixStyle is demonstrated on a
wide range of tasks including category classification, instance retrieval and
reinforcement learning.
Related papers
- CycleMix: Mixing Source Domains for Domain Generalization in Style-Dependent Data [5.124256074746721]
In the case of image classification, one frequent reason that algorithms fail to generalize is that they rely on spurious correlations present in training data.
These associations may not be present in the unseen test data, leading to significant degradation of their effectiveness.
In this work, we attempt to mitigate this Domain Generalization problem by training a robust feature extractor which disregards features attributed to image-style but infers based on style-invariant image representations.
arXiv Detail & Related papers (2024-07-18T11:43:26Z) - Bidirectional Domain Mixup for Domain Adaptive Semantic Segmentation [73.3083304858763]
This paper systematically studies the impact of mixup under the domain adaptaive semantic segmentation task.
In specific, we achieve domain mixup in two-step: cut and paste.
We provide extensive ablation experiments to empirically verify our main components of the framework.
arXiv Detail & Related papers (2023-03-17T05:22:44Z) - Normalization Perturbation: A Simple Domain Generalization Method for
Real-World Domain Shifts [133.99270341855728]
Real-world domain styles can vary substantially due to environment changes and sensor noises.
Deep models only know the training domain style.
We propose Normalization Perturbation to overcome this domain style overfitting problem.
arXiv Detail & Related papers (2022-11-08T17:36:49Z) - FIXED: Frustratingly Easy Domain Generalization with Mixup [53.782029033068675]
Domain generalization (DG) aims to learn a generalizable model from multiple training domains such that it can perform well on unseen target domains.
A popular strategy is to augment training data to benefit generalization through methods such as Mixupcitezhang 2018mixup.
We propose a simple yet effective enhancement for Mixup-based DG, namely domain-invariant Feature mIXup (FIX)
Our approach significantly outperforms nine state-of-the-art related methods, beating the best performing baseline by 6.5% on average in terms of test accuracy.
arXiv Detail & Related papers (2022-11-07T09:38:34Z) - Adversarial Style Augmentation for Domain Generalized Urban-Scene
Segmentation [120.96012935286913]
We propose a novel adversarial style augmentation approach, which can generate hard stylized images during training.
Experiments on two synthetic-to-real semantic segmentation benchmarks demonstrate that AdvStyle can significantly improve the model performance on unseen real domains.
arXiv Detail & Related papers (2022-07-11T14:01:25Z) - Domain Generalization via Gradient Surgery [5.38147998080533]
In real-life applications, machine learning models often face scenarios where there is a change in data distribution between training and test domains.
In this work, we characterize the conflicting gradients emerging in domain shift scenarios and devise novel gradient agreement strategies.
arXiv Detail & Related papers (2021-08-03T16:49:25Z) - MixStyle Neural Networks for Domain Generalization and Adaptation [122.36901703868321]
MixStyle is a plug-and-play module that can improve domain generalization performance without the need to collect more data or increase model capacity.
Our experiments show that MixStyle can significantly boost out-of-distribution generalization performance across a wide range of tasks including image recognition, instance retrieval and reinforcement learning.
arXiv Detail & Related papers (2021-07-05T14:29:19Z) - Heterogeneous Domain Generalization via Domain Mixup [0.0]
One of the main drawbacks of deep Convolutional Neural Networks (DCNN) is that they lack generalization capability.
We propose a novel heterogeneous domain generalization method by mixing up samples across multiple source domains.
Our experimental results based on the Visual Decathlon benchmark demonstrates the effectiveness of our proposed method.
arXiv Detail & Related papers (2020-09-11T13:53:56Z) - Frustratingly Simple Domain Generalization via Image Stylization [27.239024949033496]
Convolutional Neural Networks (CNNs) show impressive performance in the standard classification setting.
CNNs do not readily generalize to new domains with different statistics.
We demonstrate an extremely simple yet effective method, namely correcting this bias by augmenting the dataset with stylized images.
arXiv Detail & Related papers (2020-06-19T16:20:40Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.