A Simple Episodic Linear Probe Improves Visual Recognition in the Wild

Liang, Y; Zhu, L; Wang, X; Yang, Y

A Simple Episodic Linear Probe Improves Visual Recognition in the Wild

Liang, Y

Zhu, L Wang, X Yang, Y

Permalink

Publisher:: IEEE COMPUTER SOC
Publication Type:: Conference Proceeding
Citation:: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2022, 2022-June, pp. 9549-9559
Issue Date:: 2022-01-01

Embargoed

	Filename	Description	Size
	A Simple Episodic Linear Probe Improves Visual Recognition in the Wild.pdf	Accepted version	1.33 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Embargoed
Open Access

This item is currently unavailable due to the publisher's embargo.

The embargo period expires on 27 Sep 2024

Full metadata record

Field	Value	Language
dc.contributor.author	Liang, Y https://orcid.org/0000-0002-3429-9798
dc.contributor.author	Zhu, L
dc.contributor.author	Wang, X
dc.contributor.author	Yang, Y https://orcid.org/0000-0002-0512-880X
dc.date	2022-06-18
dc.date.accessioned	2023-03-02T02:53:27Z
dc.date.available	2023-03-02T02:53:27Z
dc.date.issued	2022-01-01
dc.identifier.citation	Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2022, 2022-June, pp. 9549-9559
dc.identifier.isbn	9781665469463
dc.identifier.issn	1063-6919
dc.identifier.uri	http://hdl.handle.net/10453/166618
dc.description.abstract	Understanding network generalization and feature discrimination is an open research problem in visual recognition. Many studies have been conducted to assess the quality of feature representations. One of the simple strategies is to utilize a linear probing classifier to quantitatively evaluate the class accuracy under the obtained features. The typical linear probe is only applied as a proxy at the inference time, but its efficacy in measuring features' suitability for linear classification is largely neglected in training. In this paper, we propose an episodic linear probing (ELP) classifier to reflect the generalization of visual rep-resentations in an online manner. ELP is trained with detached features from the network and re-initialized episodically. It demonstrates the discriminability of the visual representations in training. Then, an ELP-suitable Regularization term (ELP-SR) is introduced to reflect the distances of probability distributions between the ELP classifier and the main classifier. ELP-SR leverages are-scaling factor to regularize each sample in training, which modulates the loss function adaptively and encourages the features to be discriminative and generalized. We observe significant improvements in three real-world visual recognition tasks: fine-grained visual classification, long-tailed visual recognition, and generic object recognition. The performance gains show the effectiveness of our method in im-proving network generalization and feature discrimination.
dc.language	en
dc.publisher	IEEE COMPUTER SOC
dc.relation.ispartof	Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
dc.relation.ispartof	IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
dc.relation.ispartofseries	IEEE Conference on Computer Vision and Pattern Recognition
dc.relation.isbasedon	10.1109/CVPR52688.2022.00934
dc.rights	info:eu-repo/semantics/embargoedAccess
dc.title	A Simple Episodic Linear Probe Improves Visual Recognition in the Wild
dc.type	Conference Proceeding
utslib.citation.volume	2022-June
utslib.location.activity	New Orleans, LA
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/School of Computer Science
utslib.copyright.status	embargoed	*
utslib.copyright.embargo	2024-09-27T00:00:00+1000Z
dc.date.updated	2023-03-02T02:53:24Z
pubs.finish-date	2022-06-24
pubs.publication-status	Published
pubs.start-date	2022-06-18
pubs.volume	2022-June

Abstract:

Understanding network generalization and feature discrimination is an open research problem in visual recognition. Many studies have been conducted to assess the quality of feature representations. One of the simple strategies is to utilize a linear probing classifier to quantitatively evaluate the class accuracy under the obtained features. The typical linear probe is only applied as a proxy at the inference time, but its efficacy in measuring features' suitability for linear classification is largely neglected in training. In this paper, we propose an episodic linear probing (ELP) classifier to reflect the generalization of visual rep-resentations in an online manner. ELP is trained with detached features from the network and re-initialized episodically. It demonstrates the discriminability of the visual representations in training. Then, an ELP-suitable Regularization term (ELP-SR) is introduced to reflect the distances of probability distributions between the ELP classifier and the main classifier. ELP-SR leverages are-scaling factor to regularize each sample in training, which modulates the loss function adaptively and encourages the features to be discriminative and generalized. We observe significant improvements in three real-world visual recognition tasks: fine-grained visual classification, long-tailed visual recognition, and generic object recognition. The performance gains show the effectiveness of our method in im-proving network generalization and feature discrimination.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/166618