Robust object counting through distribution uncertainty matching and optimal transport

Abstract Object counting can be formulated as a density estimation task using point-annotated images. Although such labeling is cost-effective, trained models can be sensitive to annotation noise. In this paper, we propose a method called DUMLO (Distribution Uncertainty Matching for Loss Optimizatio...

Full description

Saved in:
Bibliographic Details
Main Authors: Sabri Boughorbel, Fethi Jarray, Rachida Zegour, Nauman Ullah Gilal, Khaled Al Thelaya, Marco Agus, Jens Schneider
Format: Article
Language:English
Published: Nature Portfolio 2025-08-01
Series:Scientific Reports
Online Access:https://doi.org/10.1038/s41598-025-14056-2
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849226268689563648
author Sabri Boughorbel
Fethi Jarray
Rachida Zegour
Nauman Ullah Gilal
Khaled Al Thelaya
Marco Agus
Jens Schneider
author_facet Sabri Boughorbel
Fethi Jarray
Rachida Zegour
Nauman Ullah Gilal
Khaled Al Thelaya
Marco Agus
Jens Schneider
author_sort Sabri Boughorbel
collection DOAJ
description Abstract Object counting can be formulated as a density estimation task using point-annotated images. Although such labeling is cost-effective, trained models can be sensitive to annotation noise. In this paper, we propose a method called DUMLO (Distribution Uncertainty Matching for Loss Optimization) that defines a loss function between a ground-truth density map and a target density map by modeling uncertainty over an augmented set of points. DUMLO formulates the loss function as a coupling between two optimal transport problems, which involves an unknown density map defined over the augmented points. To solve the problem, we propose a new algorithm, called Trihorn, which jointly estimates the loss function and the density map of the augmentation set. The latter can be interpreted as a measure of the uncertainty associated with the annotations. We provide a theoretical analysis and show that the generalization error bound of the proposed loss is tight. We extensively evaluate our model on benchmark datasets from three real-world applications: pathology cell counting, crowd counting and Vehicle Images Datasets. Our results demonstrate that the proposed model achieves good performance in terms of Mean Absolute Error and is robust to annotation noise while exhibiting a fast convergence property.
format Article
id doaj-art-46fa3ed693414c7ebc5b69e827d2945b
institution Kabale University
issn 2045-2322
language English
publishDate 2025-08-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-46fa3ed693414c7ebc5b69e827d2945b2025-08-24T11:30:32ZengNature PortfolioScientific Reports2045-23222025-08-0115111310.1038/s41598-025-14056-2Robust object counting through distribution uncertainty matching and optimal transportSabri Boughorbel0Fethi Jarray1Rachida Zegour2Nauman Ullah Gilal3Khaled Al Thelaya4Marco Agus5Jens Schneider6Qatar Computing Research Institute, Hamad Bin Khalifa UniversityLIMTIC Laboratory, UTM UniversityQatar Computing Research Institute, Hamad Bin Khalifa UniversityQatar Computing Research Institute, Hamad Bin Khalifa UniversityCollege of Science and Engineering, Hamad Bin Khalifa UniversityCollege of Science and Engineering, Hamad Bin Khalifa UniversityCollege of Science and Engineering, Hamad Bin Khalifa UniversityAbstract Object counting can be formulated as a density estimation task using point-annotated images. Although such labeling is cost-effective, trained models can be sensitive to annotation noise. In this paper, we propose a method called DUMLO (Distribution Uncertainty Matching for Loss Optimization) that defines a loss function between a ground-truth density map and a target density map by modeling uncertainty over an augmented set of points. DUMLO formulates the loss function as a coupling between two optimal transport problems, which involves an unknown density map defined over the augmented points. To solve the problem, we propose a new algorithm, called Trihorn, which jointly estimates the loss function and the density map of the augmentation set. The latter can be interpreted as a measure of the uncertainty associated with the annotations. We provide a theoretical analysis and show that the generalization error bound of the proposed loss is tight. We extensively evaluate our model on benchmark datasets from three real-world applications: pathology cell counting, crowd counting and Vehicle Images Datasets. Our results demonstrate that the proposed model achieves good performance in terms of Mean Absolute Error and is robust to annotation noise while exhibiting a fast convergence property.https://doi.org/10.1038/s41598-025-14056-2
spellingShingle Sabri Boughorbel
Fethi Jarray
Rachida Zegour
Nauman Ullah Gilal
Khaled Al Thelaya
Marco Agus
Jens Schneider
Robust object counting through distribution uncertainty matching and optimal transport
Scientific Reports
title Robust object counting through distribution uncertainty matching and optimal transport
title_full Robust object counting through distribution uncertainty matching and optimal transport
title_fullStr Robust object counting through distribution uncertainty matching and optimal transport
title_full_unstemmed Robust object counting through distribution uncertainty matching and optimal transport
title_short Robust object counting through distribution uncertainty matching and optimal transport
title_sort robust object counting through distribution uncertainty matching and optimal transport
url https://doi.org/10.1038/s41598-025-14056-2
work_keys_str_mv AT sabriboughorbel robustobjectcountingthroughdistributionuncertaintymatchingandoptimaltransport
AT fethijarray robustobjectcountingthroughdistributionuncertaintymatchingandoptimaltransport
AT rachidazegour robustobjectcountingthroughdistributionuncertaintymatchingandoptimaltransport
AT naumanullahgilal robustobjectcountingthroughdistributionuncertaintymatchingandoptimaltransport
AT khaledalthelaya robustobjectcountingthroughdistributionuncertaintymatchingandoptimaltransport
AT marcoagus robustobjectcountingthroughdistributionuncertaintymatchingandoptimaltransport
AT jensschneider robustobjectcountingthroughdistributionuncertaintymatchingandoptimaltransport