Show simple item record

AuthorKhan, Muhammad Asif
AuthorMenouar, Hamid
AuthorHamila, Ridha
Available date2024-08-21T09:49:58Z
Publication Date2024
Publication NameProceedings of the International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
ResourceScopus
ISSN21845921
URIhttp://dx.doi.org/10.5220/0012414700003660
URIhttp://hdl.handle.net/10576/57851
AbstractRecent advances in deep learning techniques have achieved remarkable performance in several computer vision problems. A notably intuitive technique called Curriculum Learning (CL) has been introduced recently for training deep learning models. Surprisingly, curriculum learning achieves significantly improved results in some tasks but marginal or no improvement in others. Hence, there is still a debate about its adoption as a standard method to train supervised learning models. In this work, we investigate the impact of curriculum learning in crowd counting using the density estimation method. We performed detailed investigations by conducting 112 experiments using six different CL settings using eight different crowd models. Our experiments show that curriculum learning improves the model learning performance and shortens the convergence time.
SponsorThis publication was made possible by the PDRA award PDRA7-0606-21012 from the Qatar National Research Fund (a member of The Qatar Foundation). The statements made herein are solely the responsibility of the authors.
Languageen
PublisherScience and Technology Publications, Lda
SubjectCNN
Crowd Counting
Curriculum Learning
Density Estimation
TitleCurriculum for Crowd Counting: Is It Worthy?
TypeConference
Pagination583-590
Volume Number3
dc.accessType Abstract Only


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record