Deep learning framework for congestion detection at public places via learning from synthetic data
Congestion in public places is one of the major problems in public transportation systems and causes a high level of discomfort for the commuters. Traditionally, overcrowding is detected by manually monitoring and analyzing the video streams from the surveillance cameras, which might lead to errors...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-01-01
|
| Series: | Journal of King Saud University: Computer and Information Sciences |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1319157822004037 |
| _version_ | 1797942126941569024 |
|---|---|
| author | Saleh Basalamah Sultan Daud Khan Emad Felemban Atif Naseer Faizan Ur Rehman |
| author_facet | Saleh Basalamah Sultan Daud Khan Emad Felemban Atif Naseer Faizan Ur Rehman |
| author_sort | Saleh Basalamah |
| collection | DOAJ |
| description | Congestion in public places is one of the major problems in public transportation systems and causes a high level of discomfort for the commuters. Traditionally, overcrowding is detected by manually monitoring and analyzing the video streams from the surveillance cameras, which might lead to errors due to limited human activity. On the other hand, current machine learning models for automatic congestion detection require a massive amount of labeled data to train the network. These models suffer from the over-fitting problem and cannot be generalized to novel scenes. First, we propose a novel synthetic dataset for congestion detection in public places to address these problems. Secondly, we propose a Bidirectional Long-short-term-memory (Bi-LSTM) model that exploits synthetic datasets to boost the performance of congestion detection in the wild. We adopt a domain adaptation strategy to bridge the gap between the real and synthetic data by pre-train the model on the synthetic dataset and then fine-tuning the model on real data. From experiment results, we observe that the proposed framework achieves a significant performance boost on the real datasets after training on the synthetic dataset. |
| first_indexed | 2024-04-10T20:02:26Z |
| format | Article |
| id | doaj.art-50436a7bd55f43c38e5e86871b698ec3 |
| institution | Directory Open Access Journal |
| issn | 1319-1578 |
| language | English |
| last_indexed | 2024-04-10T20:02:26Z |
| publishDate | 2023-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Journal of King Saud University: Computer and Information Sciences |
| spelling | doaj.art-50436a7bd55f43c38e5e86871b698ec32023-01-27T04:18:41ZengElsevierJournal of King Saud University: Computer and Information Sciences1319-15782023-01-01351102114Deep learning framework for congestion detection at public places via learning from synthetic dataSaleh Basalamah0Sultan Daud Khan1Emad Felemban2Atif Naseer3Faizan Ur Rehman4Umm Al-Qura University, Makkah, Saudi Arabia; King Abdulaziz University, Jeddah, Saudi ArabiaKing Abdulaziz University, Jeddah, Saudi Arabia; National University of Technology, Islamabad, PakistanUmm Al-Qura University, Makkah, Saudi Arabia; King Abdulaziz University, Jeddah, Saudi ArabiaUmm Al-Qura University, Makkah, Saudi Arabia; King Abdulaziz University, Jeddah, Saudi ArabiaKing Abdulaziz University, Jeddah, Saudi Arabia; FirstCity, Makkah, Saudi Arabia; Corresponding author at: FirstCity, Makkah, Saudi Arabia.Congestion in public places is one of the major problems in public transportation systems and causes a high level of discomfort for the commuters. Traditionally, overcrowding is detected by manually monitoring and analyzing the video streams from the surveillance cameras, which might lead to errors due to limited human activity. On the other hand, current machine learning models for automatic congestion detection require a massive amount of labeled data to train the network. These models suffer from the over-fitting problem and cannot be generalized to novel scenes. First, we propose a novel synthetic dataset for congestion detection in public places to address these problems. Secondly, we propose a Bidirectional Long-short-term-memory (Bi-LSTM) model that exploits synthetic datasets to boost the performance of congestion detection in the wild. We adopt a domain adaptation strategy to bridge the gap between the real and synthetic data by pre-train the model on the synthetic dataset and then fine-tuning the model on real data. From experiment results, we observe that the proposed framework achieves a significant performance boost on the real datasets after training on the synthetic dataset.http://www.sciencedirect.com/science/article/pii/S1319157822004037Congestion detectionSynthetic trajectoriesDomain adaptationCrowd analysis |
| spellingShingle | Saleh Basalamah Sultan Daud Khan Emad Felemban Atif Naseer Faizan Ur Rehman Deep learning framework for congestion detection at public places via learning from synthetic data Journal of King Saud University: Computer and Information Sciences Congestion detection Synthetic trajectories Domain adaptation Crowd analysis |
| title | Deep learning framework for congestion detection at public places via learning from synthetic data |
| title_full | Deep learning framework for congestion detection at public places via learning from synthetic data |
| title_fullStr | Deep learning framework for congestion detection at public places via learning from synthetic data |
| title_full_unstemmed | Deep learning framework for congestion detection at public places via learning from synthetic data |
| title_short | Deep learning framework for congestion detection at public places via learning from synthetic data |
| title_sort | deep learning framework for congestion detection at public places via learning from synthetic data |
| topic | Congestion detection Synthetic trajectories Domain adaptation Crowd analysis |
| url | http://www.sciencedirect.com/science/article/pii/S1319157822004037 |
| work_keys_str_mv | AT salehbasalamah deeplearningframeworkforcongestiondetectionatpublicplacesvialearningfromsyntheticdata AT sultandaudkhan deeplearningframeworkforcongestiondetectionatpublicplacesvialearningfromsyntheticdata AT emadfelemban deeplearningframeworkforcongestiondetectionatpublicplacesvialearningfromsyntheticdata AT atifnaseer deeplearningframeworkforcongestiondetectionatpublicplacesvialearningfromsyntheticdata AT faizanurrehman deeplearningframeworkforcongestiondetectionatpublicplacesvialearningfromsyntheticdata |