A Simple Feature Augmentation for Domain Generalization

Pan Li, Da Li, Wei Li, Shaogang Gong, Yanwei Fu*, Timothy M Hospedales

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract / Description of output

The topical domain generalization (DG) problem asks trained models to perform well on an unseen target domain with different data statistics from the source training domains. In computer vision, data augmentation has proven one of the most effective ways of better exploiting the source data to improve domain generalization. However, existing approaches primarily rely on image-space data augmentation, which requires careful augmentation design, and provides limited diversity of augmented data. We argue that feature augmentation is a more promising direction for DG. We find that an extremely simple technique of perturbing the feature embedding with Gaussian noise during training leads to a classifier with domain-generalization performance comparable to existing state of the art. To model more meaningful statistics reflective of cross-domain variability, we further estimate the full class-conditional feature covariance matrix iteratively during training. Subsequent joint stochastic feature augmentation provides an effective domain randomization method, perturbing features in the directions of intra-class/cross-domain variability. We verify our proposed method on three standard domain generalization benchmarks, Digit-DG, VLCS and PACS, and show it is outperforming or comparable to the state of the art in all setups, together with experimental analysis to illustrate how our method works towards training a robust generalisable model.
Original languageEnglish
Title of host publicationProceedings of 2021 IEEE/CVF International Conference on Computer Vision ICCV 2021
PublisherInstitute of Electrical and Electronics Engineers
Pages8866-8875
Number of pages10
ISBN (Electronic)978-1-6654-2812-5
ISBN (Print)978-1-6654-2813-2
DOIs
Publication statusPublished - 28 Feb 2022
EventInternational Conference on Computer Vision 2021 - Online
Duration: 11 Oct 202117 Oct 2021
https://iccv2021.thecvf.com/home

Publication series

Name2021 IEEE/CVF International Conference on Computer Vision (ICCV)
PublisherIEEE
ISSN (Print)1550-5499
ISSN (Electronic)2380-7504

Conference

ConferenceInternational Conference on Computer Vision 2021
Abbreviated titleICCV 2021
Period11/10/2117/10/21
Internet address

Fingerprint

Dive into the research topics of 'A Simple Feature Augmentation for Domain Generalization'. Together they form a unique fingerprint.

Cite this