File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Contrastive Learning based Hybrid Networks for Long-Tailed Image Classification

TitleContrastive Learning based Hybrid Networks for Long-Tailed Image Classification
Authors
Issue Date2021
Citation
Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2021, p. 943-952 How to Cite?
AbstractLearning discriminative image representations plays a vital role in long-tailed image classification because it can ease the classifier learning in imbalanced cases. Given the promising performance contrastive learning has shown recently in representation learning, in this work, we explore effective supervised contrastive learning strategies and tailor them to learn better image representations from imbalanced data in order to boost the classification accuracy thereon. Specifically, we propose a novel hybrid network structure being composed of a supervised contrastive loss to learn image representations and a cross-entropy loss to learn classifiers, where the learning is progressively transited from feature learning to the classifier learning to embody the idea that better features make better classifiers. We explore two variants of contrastive loss for feature learning, which vary in the forms but share a common idea of pulling the samples from the same class together in the normalized embedding space and pushing the samples from different classes apart. One of them is the recently proposed supervised contrastive (SC) loss, which is designed on top of the state-of-the-art unsupervised contrastive loss by incorporating positive samples from the same class. The other is a prototypical supervised contrastive (PSC) learning strategy which addresses the intensive memory consumption in standard SC loss and thus shows more promise under limited memory budget. Extensive experiments on three long-tailed classification datasets demonstrate the advantage of the proposed contrastive learning based hybrid networks in long-tailed classification.
Persistent Identifierhttp://hdl.handle.net/10722/311557
ISSN
2023 SCImago Journal Rankings: 10.331
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorWang, Peng-
dc.contributor.authorHan, Kai-
dc.contributor.authorWei, Xiu Shen-
dc.contributor.authorZhang, Lei-
dc.contributor.authorWang, Lei-
dc.date.accessioned2022-03-22T11:54:13Z-
dc.date.available2022-03-22T11:54:13Z-
dc.date.issued2021-
dc.identifier.citationProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2021, p. 943-952-
dc.identifier.issn1063-6919-
dc.identifier.urihttp://hdl.handle.net/10722/311557-
dc.description.abstractLearning discriminative image representations plays a vital role in long-tailed image classification because it can ease the classifier learning in imbalanced cases. Given the promising performance contrastive learning has shown recently in representation learning, in this work, we explore effective supervised contrastive learning strategies and tailor them to learn better image representations from imbalanced data in order to boost the classification accuracy thereon. Specifically, we propose a novel hybrid network structure being composed of a supervised contrastive loss to learn image representations and a cross-entropy loss to learn classifiers, where the learning is progressively transited from feature learning to the classifier learning to embody the idea that better features make better classifiers. We explore two variants of contrastive loss for feature learning, which vary in the forms but share a common idea of pulling the samples from the same class together in the normalized embedding space and pushing the samples from different classes apart. One of them is the recently proposed supervised contrastive (SC) loss, which is designed on top of the state-of-the-art unsupervised contrastive loss by incorporating positive samples from the same class. The other is a prototypical supervised contrastive (PSC) learning strategy which addresses the intensive memory consumption in standard SC loss and thus shows more promise under limited memory budget. Extensive experiments on three long-tailed classification datasets demonstrate the advantage of the proposed contrastive learning based hybrid networks in long-tailed classification.-
dc.languageeng-
dc.relation.ispartofProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition-
dc.titleContrastive Learning based Hybrid Networks for Long-Tailed Image Classification-
dc.typeConference_Paper-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1109/CVPR46437.2021.00100-
dc.identifier.scopuseid_2-s2.0-85123182834-
dc.identifier.spage943-
dc.identifier.epage952-
dc.identifier.isiWOS:000739917301014-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats