MetaGON: A Lightweight Pedestrian Re-Identification Domain Generalization Model Adapted to Edge Devices
Pedestrian re-identification (Re-ID) leverages cross-camera data acquired by the Internet of Things (IoT) devices and sensors to identify, monitor, and analyze pedestrians, allowing IoT applications to provide more intelligent, secure, and tailored services. Current pedestrian Re-ID research faces m...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Open Journal of the Communications Society |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10380789/ |
| _version_ | 1797346316262572032 |
|---|---|
| author | Xiting Peng Huaxuan Zhao Xiaoyu Zhang Chaofeng Zhang Caijuan Chen |
| author_facet | Xiting Peng Huaxuan Zhao Xiaoyu Zhang Chaofeng Zhang Caijuan Chen |
| author_sort | Xiting Peng |
| collection | DOAJ |
| description | Pedestrian re-identification (Re-ID) leverages cross-camera data acquired by the Internet of Things (IoT) devices and sensors to identify, monitor, and analyze pedestrians, allowing IoT applications to provide more intelligent, secure, and tailored services. Current pedestrian Re-ID research faces many challenges, such as low image resolution, perspective changes, posture changes, light changes, and occlusions, resulting in models trained on other datasets being unable to be directly applied and showing poor generalization capabilities. In addition, IoT edge devices are often limited by processing power and memory capacity and cannot withstand complex and large deep learning models. Therefore, designing a lightweight and generalizable pedestrian Re-ID model is better suited for implementation on edge devices. Considering these issues, this study presents MetaGON, a lightweight model with cross-domain generalization capabilities, which combines the lightweight omni-scale network (OSNet) with the meta-learning method and Cycle Generative Adversarial Network (CycleGAN) to perform domain generalization. The model’s generalization is enhanced through the simulation of the two stages of domain generalization in the meta-learning pipeline, where the obtained losses from meta-training and meta-testing are utilized for model optimization. Moreover, CycleGAN is employed to enhance and introduce style variations to the source data. The proposed MetaGON model is tested on a railway station re-identification dataset, and the model is deployed to edge devices for evaluation, which verifies the effectiveness of the algorithm. |
| first_indexed | 2024-03-08T11:30:19Z |
| format | Article |
| id | doaj.art-ad6b11731e1b43ad83e9dcaac0d21760 |
| institution | Directory Open Access Journal |
| issn | 2644-125X |
| language | English |
| last_indexed | 2024-03-08T11:30:19Z |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Open Journal of the Communications Society |
| spelling | doaj.art-ad6b11731e1b43ad83e9dcaac0d217602024-01-26T00:02:03ZengIEEEIEEE Open Journal of the Communications Society2644-125X2024-01-01569069910.1109/OJCOMS.2024.334959710380789MetaGON: A Lightweight Pedestrian Re-Identification Domain Generalization Model Adapted to Edge DevicesXiting Peng0https://orcid.org/0000-0002-3230-0329Huaxuan Zhao1Xiaoyu Zhang2https://orcid.org/0009-0000-7728-007XChaofeng Zhang3https://orcid.org/0000-0002-1042-1541Caijuan Chen4https://orcid.org/0000-0002-1967-7092School of Information Science and Engineering, Shenyang University of Technology, Shenyang, ChinaSchool of Information Science and Engineering, Shenyang University of Technology, Shenyang, ChinaSchool of Artificial Intelligence, Shenyang University of Technology, Shenyang, ChinaSchool of Information and Electronic Engineering, Advanced Institute of Industrial Technology, Tokyo, JapanThe Information Systems Architecture Science Research Division, National Institute of Informatics, Tokyo, JapanPedestrian re-identification (Re-ID) leverages cross-camera data acquired by the Internet of Things (IoT) devices and sensors to identify, monitor, and analyze pedestrians, allowing IoT applications to provide more intelligent, secure, and tailored services. Current pedestrian Re-ID research faces many challenges, such as low image resolution, perspective changes, posture changes, light changes, and occlusions, resulting in models trained on other datasets being unable to be directly applied and showing poor generalization capabilities. In addition, IoT edge devices are often limited by processing power and memory capacity and cannot withstand complex and large deep learning models. Therefore, designing a lightweight and generalizable pedestrian Re-ID model is better suited for implementation on edge devices. Considering these issues, this study presents MetaGON, a lightweight model with cross-domain generalization capabilities, which combines the lightweight omni-scale network (OSNet) with the meta-learning method and Cycle Generative Adversarial Network (CycleGAN) to perform domain generalization. The model’s generalization is enhanced through the simulation of the two stages of domain generalization in the meta-learning pipeline, where the obtained losses from meta-training and meta-testing are utilized for model optimization. Moreover, CycleGAN is employed to enhance and introduce style variations to the source data. The proposed MetaGON model is tested on a railway station re-identification dataset, and the model is deployed to edge devices for evaluation, which verifies the effectiveness of the algorithm.https://ieeexplore.ieee.org/document/10380789/meta-learningpedestrian re-identificationInternet of Things |
| spellingShingle | Xiting Peng Huaxuan Zhao Xiaoyu Zhang Chaofeng Zhang Caijuan Chen MetaGON: A Lightweight Pedestrian Re-Identification Domain Generalization Model Adapted to Edge Devices IEEE Open Journal of the Communications Society meta-learning pedestrian re-identification Internet of Things |
| title | MetaGON: A Lightweight Pedestrian Re-Identification Domain Generalization Model Adapted to Edge Devices |
| title_full | MetaGON: A Lightweight Pedestrian Re-Identification Domain Generalization Model Adapted to Edge Devices |
| title_fullStr | MetaGON: A Lightweight Pedestrian Re-Identification Domain Generalization Model Adapted to Edge Devices |
| title_full_unstemmed | MetaGON: A Lightweight Pedestrian Re-Identification Domain Generalization Model Adapted to Edge Devices |
| title_short | MetaGON: A Lightweight Pedestrian Re-Identification Domain Generalization Model Adapted to Edge Devices |
| title_sort | metagon a lightweight pedestrian re identification domain generalization model adapted to edge devices |
| topic | meta-learning pedestrian re-identification Internet of Things |
| url | https://ieeexplore.ieee.org/document/10380789/ |
| work_keys_str_mv | AT xitingpeng metagonalightweightpedestrianreidentificationdomaingeneralizationmodeladaptedtoedgedevices AT huaxuanzhao metagonalightweightpedestrianreidentificationdomaingeneralizationmodeladaptedtoedgedevices AT xiaoyuzhang metagonalightweightpedestrianreidentificationdomaingeneralizationmodeladaptedtoedgedevices AT chaofengzhang metagonalightweightpedestrianreidentificationdomaingeneralizationmodeladaptedtoedgedevices AT caijuanchen metagonalightweightpedestrianreidentificationdomaingeneralizationmodeladaptedtoedgedevices |