Box-based Refinement for Weakly Supervised and Unsupervised Localization Tasks

Eyal Gomel; Tal Shaharbany; Lior Wolf

doi:10.1109/ICCV51070.2023.01470

Box-based Refinement for Weakly Supervised and Unsupervised Localization Tasks

Eyal Gomel^*, Tal Shaharbany, Lior Wolf

^*Corresponding author for this work

Tel Aviv University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Scopus citations

Abstract

It has been established that training a box-based detector network can enhance the localization performance of weakly supervised and unsupervised methods. Moreover, we extend this understanding by demonstrating that these detectors can be utilized to improve the original network, paving the way for further advancements. To accomplish this, we train the detectors on top of the network output instead of the image data and apply suitable loss backpropagation. Our findings reveal a significant improvement in phrase grounding for the "what is where by looking"task, as well as various methods of unsupervised object discovery. Our code is available at https://github.com/eyalgomel/box-based-refinement.

Original language	English
Title of host publication	Proceedings - 2023 IEEE/CVF International Conference on Computer Vision, ICCV 2023
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	15998-16008
Number of pages	11
ISBN (Electronic)	9798350307184
DOIs	https://doi.org/10.1109/ICCV51070.2023.01470
State	Published - 2023
Event	2023 IEEE/CVF International Conference on Computer Vision, ICCV 2023 - Paris, France Duration: 2 Oct 2023 → 6 Oct 2023

Publication series

Name	Proceedings of the IEEE International Conference on Computer Vision
ISSN (Print)	1550-5499

Conference

Conference	2023 IEEE/CVF International Conference on Computer Vision, ICCV 2023
Country/Territory	France
City	Paris
Period	2/10/23 → 6/10/23

Access to Document

10.1109/ICCV51070.2023.01470

Cite this

Gomel, E., Shaharbany, T., & Wolf, L. (2023). Box-based Refinement for Weakly Supervised and Unsupervised Localization Tasks. In Proceedings - 2023 IEEE/CVF International Conference on Computer Vision, ICCV 2023 (pp. 15998-16008). (Proceedings of the IEEE International Conference on Computer Vision). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICCV51070.2023.01470

@inproceedings{a6304806dc0c4367b981740029aba12e,

title = "Box-based Refinement for Weakly Supervised and Unsupervised Localization Tasks",

abstract = "It has been established that training a box-based detector network can enhance the localization performance of weakly supervised and unsupervised methods. Moreover, we extend this understanding by demonstrating that these detectors can be utilized to improve the original network, paving the way for further advancements. To accomplish this, we train the detectors on top of the network output instead of the image data and apply suitable loss backpropagation. Our findings reveal a significant improvement in phrase grounding for the {"}what is where by looking{"}task, as well as various methods of unsupervised object discovery. Our code is available at https://github.com/eyalgomel/box-based-refinement.",

author = "Eyal Gomel and Tal Shaharbany and Lior Wolf",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 2023 IEEE/CVF International Conference on Computer Vision, ICCV 2023 ; Conference date: 02-10-2023 Through 06-10-2023",

year = "2023",

doi = "10.1109/ICCV51070.2023.01470",

language = "אנגלית",

series = "Proceedings of the IEEE International Conference on Computer Vision",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "15998--16008",

booktitle = "Proceedings - 2023 IEEE/CVF International Conference on Computer Vision, ICCV 2023",

address = "ארצות הברית",

}

Gomel, E, Shaharbany, T & Wolf, L 2023, Box-based Refinement for Weakly Supervised and Unsupervised Localization Tasks. in Proceedings - 2023 IEEE/CVF International Conference on Computer Vision, ICCV 2023. Proceedings of the IEEE International Conference on Computer Vision, Institute of Electrical and Electronics Engineers Inc., pp. 15998-16008, 2023 IEEE/CVF International Conference on Computer Vision, ICCV 2023, Paris, France, 2/10/23. https://doi.org/10.1109/ICCV51070.2023.01470

Box-based Refinement for Weakly Supervised and Unsupervised Localization Tasks. / Gomel, Eyal; Shaharbany, Tal; Wolf, Lior.
Proceedings - 2023 IEEE/CVF International Conference on Computer Vision, ICCV 2023. Institute of Electrical and Electronics Engineers Inc., 2023. p. 15998-16008 (Proceedings of the IEEE International Conference on Computer Vision).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Box-based Refinement for Weakly Supervised and Unsupervised Localization Tasks

AU - Gomel, Eyal

AU - Shaharbany, Tal

AU - Wolf, Lior

PY - 2023

Y1 - 2023

N2 - It has been established that training a box-based detector network can enhance the localization performance of weakly supervised and unsupervised methods. Moreover, we extend this understanding by demonstrating that these detectors can be utilized to improve the original network, paving the way for further advancements. To accomplish this, we train the detectors on top of the network output instead of the image data and apply suitable loss backpropagation. Our findings reveal a significant improvement in phrase grounding for the "what is where by looking"task, as well as various methods of unsupervised object discovery. Our code is available at https://github.com/eyalgomel/box-based-refinement.

AB - It has been established that training a box-based detector network can enhance the localization performance of weakly supervised and unsupervised methods. Moreover, we extend this understanding by demonstrating that these detectors can be utilized to improve the original network, paving the way for further advancements. To accomplish this, we train the detectors on top of the network output instead of the image data and apply suitable loss backpropagation. Our findings reveal a significant improvement in phrase grounding for the "what is where by looking"task, as well as various methods of unsupervised object discovery. Our code is available at https://github.com/eyalgomel/box-based-refinement.

UR - http://www.scopus.com/inward/record.url?scp=85175228762&partnerID=8YFLogxK

U2 - 10.1109/ICCV51070.2023.01470

DO - 10.1109/ICCV51070.2023.01470

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:85175228762

T3 - Proceedings of the IEEE International Conference on Computer Vision

SP - 15998

EP - 16008

BT - Proceedings - 2023 IEEE/CVF International Conference on Computer Vision, ICCV 2023

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2023 IEEE/CVF International Conference on Computer Vision, ICCV 2023

Y2 - 2 October 2023 through 6 October 2023

ER -

Gomel E, Shaharbany T, Wolf L. Box-based Refinement for Weakly Supervised and Unsupervised Localization Tasks. In Proceedings - 2023 IEEE/CVF International Conference on Computer Vision, ICCV 2023. Institute of Electrical and Electronics Engineers Inc. 2023. p. 15998-16008. (Proceedings of the IEEE International Conference on Computer Vision). doi: 10.1109/ICCV51070.2023.01470

Box-based Refinement for Weakly Supervised and Unsupervised Localization Tasks

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this