Training a Regression-Based Model for Crowd Counting in Transit Cars Using Ranked Image Pairs and Triplets
Accurately measuring the level of crowding in transit cars is crucial for ensuring passenger safety and efficient operation. However, applying object detection algorithms to crowd counting in transit cars poses difficulties due to the low viewpoint of the cameras and the labor-intensive task of imag...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2024-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/10403893/ |
_version_ | 1797339212138151936 |
---|---|
author | Hojun Lee Kyeongjun Lee Jiwon Kang Keemin Sohn |
author_facet | Hojun Lee Kyeongjun Lee Jiwon Kang Keemin Sohn |
author_sort | Hojun Lee |
collection | DOAJ |
description | Accurately measuring the level of crowding in transit cars is crucial for ensuring passenger safety and efficient operation. However, applying object detection algorithms to crowd counting in transit cars poses difficulties due to the low viewpoint of the cameras and the labor-intensive task of image labeling. Although some researchers have explored regression-based crowd counting methods without labeling with bounding boxes, their approaches still necessitate manual counting of passengers for image labeling. To overcome these challenges, we propose a novel calibration method for regression-based models that minimizes the number of labeled images required for training. Our approach employs image pairs and triplets with ranks for reinforcing the model training. Subsequently, the training task requires a minimal number of images labeled with exact passenger counts. Experimental results demonstrate that our proposed calibration approach considerably enhances the crowd counting performance of the conventional regression-based model. Specifically, our method reduces the mean absolute error (MAE) by 76.5% and 34.3% for conventional detection- and regression-based calibration methods, respectively. |
first_indexed | 2024-03-08T09:42:40Z |
format | Article |
id | doaj.art-5bd0b1324956483f9f2db3f1c4cc2f1b |
institution | Directory Open Access Journal |
issn | 2169-3536 |
language | English |
last_indexed | 2024-03-08T09:42:40Z |
publishDate | 2024-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Access |
spelling | doaj.art-5bd0b1324956483f9f2db3f1c4cc2f1b2024-01-30T00:04:02ZengIEEEIEEE Access2169-35362024-01-0112128181282610.1109/ACCESS.2024.335544210403893Training a Regression-Based Model for Crowd Counting in Transit Cars Using Ranked Image Pairs and TripletsHojun Lee0Kyeongjun Lee1Jiwon Kang2https://orcid.org/0009-0007-1228-5049Keemin Sohn3https://orcid.org/0000-0002-7270-7094Department of Smart City, Chung-Ang University, Seoul, South KoreaDepartment of Urban Engineering, Chung-Ang University, Seoul, South KoreaDepartment of Urban Engineering, Chung-Ang University, Seoul, South KoreaDepartment of Smart City, Chung-Ang University, Seoul, South KoreaAccurately measuring the level of crowding in transit cars is crucial for ensuring passenger safety and efficient operation. However, applying object detection algorithms to crowd counting in transit cars poses difficulties due to the low viewpoint of the cameras and the labor-intensive task of image labeling. Although some researchers have explored regression-based crowd counting methods without labeling with bounding boxes, their approaches still necessitate manual counting of passengers for image labeling. To overcome these challenges, we propose a novel calibration method for regression-based models that minimizes the number of labeled images required for training. Our approach employs image pairs and triplets with ranks for reinforcing the model training. Subsequently, the training task requires a minimal number of images labeled with exact passenger counts. Experimental results demonstrate that our proposed calibration approach considerably enhances the crowd counting performance of the conventional regression-based model. Specifically, our method reduces the mean absolute error (MAE) by 76.5% and 34.3% for conventional detection- and regression-based calibration methods, respectively.https://ieeexplore.ieee.org/document/10403893/Passenger load in transitcrowd countingcomputer visionregression-based modelranking model |
spellingShingle | Hojun Lee Kyeongjun Lee Jiwon Kang Keemin Sohn Training a Regression-Based Model for Crowd Counting in Transit Cars Using Ranked Image Pairs and Triplets IEEE Access Passenger load in transit crowd counting computer vision regression-based model ranking model |
title | Training a Regression-Based Model for Crowd Counting in Transit Cars Using Ranked Image Pairs and Triplets |
title_full | Training a Regression-Based Model for Crowd Counting in Transit Cars Using Ranked Image Pairs and Triplets |
title_fullStr | Training a Regression-Based Model for Crowd Counting in Transit Cars Using Ranked Image Pairs and Triplets |
title_full_unstemmed | Training a Regression-Based Model for Crowd Counting in Transit Cars Using Ranked Image Pairs and Triplets |
title_short | Training a Regression-Based Model for Crowd Counting in Transit Cars Using Ranked Image Pairs and Triplets |
title_sort | training a regression based model for crowd counting in transit cars using ranked image pairs and triplets |
topic | Passenger load in transit crowd counting computer vision regression-based model ranking model |
url | https://ieeexplore.ieee.org/document/10403893/ |
work_keys_str_mv | AT hojunlee trainingaregressionbasedmodelforcrowdcountingintransitcarsusingrankedimagepairsandtriplets AT kyeongjunlee trainingaregressionbasedmodelforcrowdcountingintransitcarsusingrankedimagepairsandtriplets AT jiwonkang trainingaregressionbasedmodelforcrowdcountingintransitcarsusingrankedimagepairsandtriplets AT keeminsohn trainingaregressionbasedmodelforcrowdcountingintransitcarsusingrankedimagepairsandtriplets |