Curriculum for Crowd Counting: Is It Worthy?
Author | Khan, Muhammad Asif |
Author | Menouar, Hamid |
Author | Hamila, Ridha |
Available date | 2024-08-21T09:49:58Z |
Publication Date | 2024 |
Publication Name | Proceedings of the International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications |
Resource | Scopus |
ISSN | 21845921 |
Abstract | Recent 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. |
Sponsor | This 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. |
Language | en |
Publisher | Science and Technology Publications, Lda |
Subject | CNN Crowd Counting Curriculum Learning Density Estimation |
Type | Conference Paper |
Pagination | 583-590 |
Volume Number | 3 |
Files in this item
Files | Size | Format | View |
---|---|---|---|
There are no files associated with this item. |
This item appears in the following Collection(s)
-
Electrical Engineering [2616 items ]