Deeper, Broader and Artier Domain Generalization

Li, Da; Yang, Yongxin; Song, Yi-Zhe; Hospedales, Timothy M.

doi:10.1109/iccv.2017.591

Cited by 951 publications

(876 citation statements)

References 20 publications

Supporting

Mentioning

834

Contrasting

Unclassified

Order By: Relevance

“…Datasets To evaluate the performance of JiGen when training over multiple sources we considered three domain generalization datasets. PACS [27] covers 7 object categories and 4 domains (Photo, Art Paintings, Cartoon and Sketches). We followed the experimental protocol in [27] and trained our model considering three domains as source datasets and the remaining one as target.…”

Section: Methodsmentioning

confidence: 99%

Domain Generalization by Solving Jigsaw Puzzles

Carlucci

D’Innocente

Bucci

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

626

424

View full text Add to dashboard Cite

Human adaptability relies crucially on the ability to learn and merge knowledge both from supervised and unsupervised learning: the parents point out few important concepts, but then the children fill in the gaps on their own. This is particularly effective, because supervised learning can never be exhaustive and thus learning autonomously allows to discover invariances and regularities that help to generalize. In this paper we propose to apply a similar approach to the task of object recognition across domains: our model learns the semantic labels in a supervised fashion, and broadens its understanding of the data by learning from self-supervised signals how to solve a jigsaw puzzle on the same images. This secondary task helps the network to learn the concepts of spatial correlation while acting as a regularizer for the classification task. Multiple experiments on the PACS, VLCS, Office-Home and digits datasets confirm our intuition and show that this simple method outperforms previous domain generalization and adaptation solutions. An ablation study further illustrates the inner workings of our approach. * This work was done while at

show abstract

Section: Methodsmentioning

confidence: 99%

Domain Generalization by Solving Jigsaw Puzzles

Carlucci

D’Innocente

Bucci

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

626

424

View full text Add to dashboard Cite

show abstract

“…[28] imposed Maximum Mean Discrepancy measure to align the distributions among different domains and train the network with adversarial feature learning. [26] assigned a separate network duplication to each training domain during training and used the shared parameter for inference. [27] improved generalization performance by using a meta-learning approach on the split training sets.…”

Section: Related Workmentioning

confidence: 99%

Domain Randomization and Pyramid Consistency: Simulation-to-Real Generalization Without Accessing Target Domain Data

Yue

Zhang

Zhao

et al. 2019

2019 IEEE/CVF International Conference on Computer Vision (ICCV)

300

251

View full text Add to dashboard Cite

We propose to harness the potential of simulation for the semantic segmentation of real-world self-driving scenes in a domain generalization fashion. The segmentation network is trained without any data of target domains and tested on the unseen target domains. To this end, we propose a new approach of domain randomization and pyramid consistency to learn a model with high generalizability. First, we propose to randomize the synthetic images with the styles of real images in terms of visual appearances using auxiliary datasets, in order to effectively learn domain-invariant representations. Second, we further enforce pyramid consistency across different "stylized" images and within an image, in order to learn domaininvariant and scale-invariant features, respectively. Extensive experiments are conducted on the generalization from GTA and SYNTHIA to Cityscapes, BDDS and Mapillary; and our method achieves superior results over the stateof-the-art techniques. Remarkably, our generalization results are on par with or even better than those obtained by state-of-the-art simulation-to-real domain adaptation methods, which access the target domain data at training time.

show abstract

“…Office-Caltech, we compare our method on the DG scenario with the state-of-the-art DG methods: learned -support vector machine (L-SVM) [47], kernel fisher discriminant analysis (KDA) [49], domain-invariant component analysis (DICA) [36], multi-task auto-encoder (MTAE) [20], domain separation network (DSN) [48], deeper, broader and artier domain generalization (DBADG) [21], conditional invariant deep domain generalization (CIDDG) [38], undoing the damage of dataset bias (Undo-Bias) [19], unbiased metric learning (UML) [46], multi-task autoencoders (MTAE) [20] and deep domain generalization with structured low-rank constraint (DGLRC) [22].…”

Section: Alexnetmentioning

confidence: 99%

“…This approach significantly enhances the ability of prior adversarial adaptation approaches through our additional proposed domain discrepancy module. The proposed method was evaluated on five benchmark datasets (Office-31 [25], Office-Home [26], ImageCLEF-DA 1 , Office-Caltech [27] and PACS [21]) on standard unsupervised DA and DG settings. Experiments prove that our proposed model for unsupervised DA and DG yields state-of-the-art results.…”

Section: Introductionmentioning

confidence: 99%

Correlation-aware adversarial domain adaptation and generalization

Rahman

Fookes

Baktashmotlagh

et al. 2020

Pattern Recognition

114

View full text Add to dashboard Cite

Domain adaptation (DA) and domain generalization (DG) have emerged as a solution to the domain shift problem where the distribution of the source and target data is different. The task of DG is more challenging than DA as the target data is totally unseen during the training phase in DG scenarios. The current state-of-the-art employs adversarial techniques, however, these are rarely considered for the DG problem. Furthermore, these approaches do not consider correlation alignment which has been proven highly beneficial for minimizing domain discrepancy. In this paper, we propose a correlation-aware adversarial DA and DG framework where the features of the source and target data are minimized using correlation alignment along with adversarial learning.Incorporating the correlation alignment module along with adversarial learning helps to achieve a more domain agnostic model due to the improved ability to reduce domain discrepancy with unlabeled target data more effectively. Experiments on benchmark datasets serve as evidence that our proposed method yields improved state-of-the-art performance.

show abstract

Deeper, Broader and Artier Domain Generalization

Cited by 951 publications

References 20 publications

Domain Generalization by Solving Jigsaw Puzzles

Domain Generalization by Solving Jigsaw Puzzles

Domain Randomization and Pyramid Consistency: Simulation-to-Real Generalization Without Accessing Target Domain Data

Correlation-aware adversarial domain adaptation and generalization

Contact Info

Product

Resources

About