Supervised and Self-Supervised Learning for Assembly Line Action Recognition
The safety and efficiency of assembly lines are critical to manufacturing, but human supervisors cannot oversee all activities simultaneously. This study addresses this challenge by performing a comparative study to construct an initial real-time, semi-supervised temporal action recognition setup fo...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Journal of Imaging |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2313-433X/11/1/17 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832588254628544512 |
|---|---|
| author | Christopher Indris Fady Ibrahim Hatem Ibrahem Götz Bramesfeld Jie Huo Hafiz Mughees Ahmad Syed Khizer Hayat Guanghui Wang |
| author_facet | Christopher Indris Fady Ibrahim Hatem Ibrahem Götz Bramesfeld Jie Huo Hafiz Mughees Ahmad Syed Khizer Hayat Guanghui Wang |
| author_sort | Christopher Indris |
| collection | DOAJ |
| description | The safety and efficiency of assembly lines are critical to manufacturing, but human supervisors cannot oversee all activities simultaneously. This study addresses this challenge by performing a comparative study to construct an initial real-time, semi-supervised temporal action recognition setup for monitoring worker actions on assembly lines. Various feature extractors and localization models were benchmarked using a new assembly dataset, with the I3D model achieving an average mAP@IoU=0.1:0.7 of 85% without optical flow or fine-tuning. The comparative study was extended to self-supervised learning via a modified SPOT model, which achieved a mAP@IoU=0.1:0.7 of 65% with just 10% of the data labeled using extractor architectures from the fully-supervised portion. Milestones include high scores for both fully and semi-supervised learning on this dataset and improved SPOT performance on ANet1.3. This study identified the particularities of the problem, which were leveraged and referenced to explain the results observed in semi-supervised scenarios. The findings highlight the potential for developing a scalable solution in the future, providing labour efficiency and safety compliance for manufacturers. |
| format | Article |
| id | doaj-art-12ae3d3b446743539b26163c2ede0764 |
| institution | Kabale University |
| issn | 2313-433X |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Imaging |
| spelling | doaj-art-12ae3d3b446743539b26163c2ede07642025-01-24T13:36:17ZengMDPI AGJournal of Imaging2313-433X2025-01-011111710.3390/jimaging11010017Supervised and Self-Supervised Learning for Assembly Line Action RecognitionChristopher Indris0Fady Ibrahim1Hatem Ibrahem2Götz Bramesfeld3Jie Huo4Hafiz Mughees Ahmad5Syed Khizer Hayat6Guanghui Wang7Department of Computer Science, Toronto Metropolitan University, Toronto, ON M5B 2K3, CanadaDepartment of Computer Science, Toronto Metropolitan University, Toronto, ON M5B 2K3, CanadaDepartment of Computer Science, Toronto Metropolitan University, Toronto, ON M5B 2K3, CanadaDepartment of Aerospace Engineering, Toronto Metropolitan University, Toronto, ON M5B 2K3, CanadaIFIVEO Canada Inc., Windsor, ON N8W 0A6, CanadaIFIVEO Canada Inc., Windsor, ON N8W 0A6, CanadaIFIVEO Canada Inc., Windsor, ON N8W 0A6, CanadaDepartment of Computer Science, Toronto Metropolitan University, Toronto, ON M5B 2K3, CanadaThe safety and efficiency of assembly lines are critical to manufacturing, but human supervisors cannot oversee all activities simultaneously. This study addresses this challenge by performing a comparative study to construct an initial real-time, semi-supervised temporal action recognition setup for monitoring worker actions on assembly lines. Various feature extractors and localization models were benchmarked using a new assembly dataset, with the I3D model achieving an average mAP@IoU=0.1:0.7 of 85% without optical flow or fine-tuning. The comparative study was extended to self-supervised learning via a modified SPOT model, which achieved a mAP@IoU=0.1:0.7 of 65% with just 10% of the data labeled using extractor architectures from the fully-supervised portion. Milestones include high scores for both fully and semi-supervised learning on this dataset and improved SPOT performance on ANet1.3. This study identified the particularities of the problem, which were leveraged and referenced to explain the results observed in semi-supervised scenarios. The findings highlight the potential for developing a scalable solution in the future, providing labour efficiency and safety compliance for manufacturers.https://www.mdpi.com/2313-433X/11/1/17computer visionaction recognitiontemporal action localizationsemi-supervised learningsupervised learningreal-time feature extraction |
| spellingShingle | Christopher Indris Fady Ibrahim Hatem Ibrahem Götz Bramesfeld Jie Huo Hafiz Mughees Ahmad Syed Khizer Hayat Guanghui Wang Supervised and Self-Supervised Learning for Assembly Line Action Recognition Journal of Imaging computer vision action recognition temporal action localization semi-supervised learning supervised learning real-time feature extraction |
| title | Supervised and Self-Supervised Learning for Assembly Line Action Recognition |
| title_full | Supervised and Self-Supervised Learning for Assembly Line Action Recognition |
| title_fullStr | Supervised and Self-Supervised Learning for Assembly Line Action Recognition |
| title_full_unstemmed | Supervised and Self-Supervised Learning for Assembly Line Action Recognition |
| title_short | Supervised and Self-Supervised Learning for Assembly Line Action Recognition |
| title_sort | supervised and self supervised learning for assembly line action recognition |
| topic | computer vision action recognition temporal action localization semi-supervised learning supervised learning real-time feature extraction |
| url | https://www.mdpi.com/2313-433X/11/1/17 |
| work_keys_str_mv | AT christopherindris supervisedandselfsupervisedlearningforassemblylineactionrecognition AT fadyibrahim supervisedandselfsupervisedlearningforassemblylineactionrecognition AT hatemibrahem supervisedandselfsupervisedlearningforassemblylineactionrecognition AT gotzbramesfeld supervisedandselfsupervisedlearningforassemblylineactionrecognition AT jiehuo supervisedandselfsupervisedlearningforassemblylineactionrecognition AT hafizmugheesahmad supervisedandselfsupervisedlearningforassemblylineactionrecognition AT syedkhizerhayat supervisedandselfsupervisedlearningforassemblylineactionrecognition AT guanghuiwang supervisedandselfsupervisedlearningforassemblylineactionrecognition |