Improved object detection method for autonomous driving based on DETR
Object detection is a critical component in the development of autonomous driving technology and has demonstrated significant growth potential. To address the limitations of current techniques, this paper presents an improved object detection method for autonomous driving based on a detection transf...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-01-01
|
| Series: | Frontiers in Neurorobotics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fnbot.2024.1484276/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832593903147024384 |
|---|---|
| author | Huaqi Zhao Songnan Zhang Xiang Peng Zhengguang Lu Guojing Li |
| author_facet | Huaqi Zhao Songnan Zhang Xiang Peng Zhengguang Lu Guojing Li |
| author_sort | Huaqi Zhao |
| collection | DOAJ |
| description | Object detection is a critical component in the development of autonomous driving technology and has demonstrated significant growth potential. To address the limitations of current techniques, this paper presents an improved object detection method for autonomous driving based on a detection transformer (DETR). First, we introduce a multi-scale feature and location information extraction method, which solves the inadequacy of the model for multi-scale object localization and detection. In addition, we developed a transformer encoder based on the group axial attention mechanism. This allows for efficient attention range control in the horizontal and vertical directions while reducing computation, ultimately enhancing the inference speed. Furthermore, we propose a novel dynamic hyperparameter tuning training method based on Pareto efficiency, which coordinates the training state of the loss functions through dynamic weights, overcoming issues associated with manually setting fixed weights and enhancing model convergence speed and accuracy. Experimental results demonstrate that the proposed method surpasses others, with improvements of 3.3%, 4.5%, and 3% in average precision on the COCO, PASCAL VOC, and KITTI datasets, respectively, and an 84% increase in FPS. |
| format | Article |
| id | doaj-art-003ac97d0f3f4f6fa76838e7f2a37565 |
| institution | Kabale University |
| issn | 1662-5218 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neurorobotics |
| spelling | doaj-art-003ac97d0f3f4f6fa76838e7f2a375652025-01-20T07:20:10ZengFrontiers Media S.A.Frontiers in Neurorobotics1662-52182025-01-011810.3389/fnbot.2024.14842761484276Improved object detection method for autonomous driving based on DETRHuaqi Zhao0Songnan Zhang1Xiang Peng2Zhengguang Lu3Guojing Li4The Heilongjiang Provincial Key Laboratory of Autonomous Intelligence and Information Processing, School of Information and Electronic Technology, Jiamusi University, Jiamusi, ChinaThe Heilongjiang Provincial Key Laboratory of Autonomous Intelligence and Information Processing, School of Information and Electronic Technology, Jiamusi University, Jiamusi, ChinaThe Heilongjiang Provincial Key Laboratory of Autonomous Intelligence and Information Processing, School of Information and Electronic Technology, Jiamusi University, Jiamusi, ChinaThe Heilongjiang Provincial Key Laboratory of Autonomous Intelligence and Information Processing, School of Information and Electronic Technology, Jiamusi University, Jiamusi, ChinaSchool of Materials Science and Engineering, Jiamusi University, Jiamusi, ChinaObject detection is a critical component in the development of autonomous driving technology and has demonstrated significant growth potential. To address the limitations of current techniques, this paper presents an improved object detection method for autonomous driving based on a detection transformer (DETR). First, we introduce a multi-scale feature and location information extraction method, which solves the inadequacy of the model for multi-scale object localization and detection. In addition, we developed a transformer encoder based on the group axial attention mechanism. This allows for efficient attention range control in the horizontal and vertical directions while reducing computation, ultimately enhancing the inference speed. Furthermore, we propose a novel dynamic hyperparameter tuning training method based on Pareto efficiency, which coordinates the training state of the loss functions through dynamic weights, overcoming issues associated with manually setting fixed weights and enhancing model convergence speed and accuracy. Experimental results demonstrate that the proposed method surpasses others, with improvements of 3.3%, 4.5%, and 3% in average precision on the COCO, PASCAL VOC, and KITTI datasets, respectively, and an 84% increase in FPS.https://www.frontiersin.org/articles/10.3389/fnbot.2024.1484276/fullobject detectionfeature extractiontransformer encoderloss functionparameter tuning |
| spellingShingle | Huaqi Zhao Songnan Zhang Xiang Peng Zhengguang Lu Guojing Li Improved object detection method for autonomous driving based on DETR Frontiers in Neurorobotics object detection feature extraction transformer encoder loss function parameter tuning |
| title | Improved object detection method for autonomous driving based on DETR |
| title_full | Improved object detection method for autonomous driving based on DETR |
| title_fullStr | Improved object detection method for autonomous driving based on DETR |
| title_full_unstemmed | Improved object detection method for autonomous driving based on DETR |
| title_short | Improved object detection method for autonomous driving based on DETR |
| title_sort | improved object detection method for autonomous driving based on detr |
| topic | object detection feature extraction transformer encoder loss function parameter tuning |
| url | https://www.frontiersin.org/articles/10.3389/fnbot.2024.1484276/full |
| work_keys_str_mv | AT huaqizhao improvedobjectdetectionmethodforautonomousdrivingbasedondetr AT songnanzhang improvedobjectdetectionmethodforautonomousdrivingbasedondetr AT xiangpeng improvedobjectdetectionmethodforautonomousdrivingbasedondetr AT zhengguanglu improvedobjectdetectionmethodforautonomousdrivingbasedondetr AT guojingli improvedobjectdetectionmethodforautonomousdrivingbasedondetr |