File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Learning from Extrinsic and Intrinsic Supervisions for Domain Generalization

TitleLearning from Extrinsic and Intrinsic Supervisions for Domain Generalization
Authors
KeywordsSelf-supervision
Unsupervised learning
Metric learning
Domain generalization
Issue Date2020
PublisherSpringer.
Citation
16th European Conference on Computer Vision (ECCV 2020), Glasgow, UK, 23-28 August 2020. In Vedaldi, A, Bischof, H, Brox, T, Frahm, J (Eds.), Computer Vision – ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part IX, p. 159-176. Cham, Switzerland: Springer, 2020 How to Cite?
AbstractThe generalization capability of neural networks across domains is crucial for real-world applications. We argue that a generalized object recognition system should well understand the relationships among different images and also the images themselves at the same time. To this end, we present a new domain generalization framework (called EISNet) that learns how to generalize across domains simultaneously from extrinsic relationship supervision and intrinsic self-supervision for images from multi-source domains. To be specific, we formulate our framework with feature embedding using a multi-task learning paradigm. Besides conducting the common supervised recognition task, we seamlessly integrate a momentum metric learning task and a self-supervised auxiliary task to collectively integrate the extrinsic and intrinsic supervisions. Also, we develop an effective momentum metric learning scheme with the K-hard negative mining to boost the network generalization ability. We demonstrate the effectiveness of our approach on two standard object recognition benchmarks VLCS and PACS, and show that our EISNet achieves state-of-the-art performance.
Persistent Identifierhttp://hdl.handle.net/10722/299482
ISBN
ISSN
2023 SCImago Journal Rankings: 0.606
Series/Report no.Lecture Notes in Computer Science ; 12354

 

DC FieldValueLanguage
dc.contributor.authorWang, Shujun-
dc.contributor.authorYu, Lequan-
dc.contributor.authorLi, Caizi-
dc.contributor.authorFu, Chi Wing-
dc.contributor.authorHeng, Pheng Ann-
dc.date.accessioned2021-05-21T03:34:30Z-
dc.date.available2021-05-21T03:34:30Z-
dc.date.issued2020-
dc.identifier.citation16th European Conference on Computer Vision (ECCV 2020), Glasgow, UK, 23-28 August 2020. In Vedaldi, A, Bischof, H, Brox, T, Frahm, J (Eds.), Computer Vision – ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part IX, p. 159-176. Cham, Switzerland: Springer, 2020-
dc.identifier.isbn9783030585440-
dc.identifier.issn0302-9743-
dc.identifier.urihttp://hdl.handle.net/10722/299482-
dc.description.abstractThe generalization capability of neural networks across domains is crucial for real-world applications. We argue that a generalized object recognition system should well understand the relationships among different images and also the images themselves at the same time. To this end, we present a new domain generalization framework (called EISNet) that learns how to generalize across domains simultaneously from extrinsic relationship supervision and intrinsic self-supervision for images from multi-source domains. To be specific, we formulate our framework with feature embedding using a multi-task learning paradigm. Besides conducting the common supervised recognition task, we seamlessly integrate a momentum metric learning task and a self-supervised auxiliary task to collectively integrate the extrinsic and intrinsic supervisions. Also, we develop an effective momentum metric learning scheme with the K-hard negative mining to boost the network generalization ability. We demonstrate the effectiveness of our approach on two standard object recognition benchmarks VLCS and PACS, and show that our EISNet achieves state-of-the-art performance.-
dc.languageeng-
dc.publisherSpringer.-
dc.relation.ispartofComputer Vision – ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part IX-
dc.relation.ispartofseriesLecture Notes in Computer Science ; 12354-
dc.subjectSelf-supervision-
dc.subjectUnsupervised learning-
dc.subjectMetric learning-
dc.subjectDomain generalization-
dc.titleLearning from Extrinsic and Intrinsic Supervisions for Domain Generalization-
dc.typeConference_Paper-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1007/978-3-030-58545-7_10-
dc.identifier.scopuseid_2-s2.0-85097091172-
dc.identifier.spage159-
dc.identifier.epage176-
dc.identifier.eissn1611-3349-
dc.publisher.placeCham, Switzerland-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats