Light-Net: Lightweight Object Detector
Currently, object detectors based on CNN, such as RetinaNet, Faster-RCNN, CornerNet series, can achieve good performance, but have some common drawbacks, like large calculation cost, high model complexity and slow detection speed. In this paper, a new lightweight object detector is proposed, which a...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9217510/ |
| _version_ | 1818459289190137856 |
|---|---|
| author | Yunbo Rao Guo Yi Junmin Xue Jiansu Pu Jianping Gou Qiujie Wang Qifei Wang |
| author_facet | Yunbo Rao Guo Yi Junmin Xue Jiansu Pu Jianping Gou Qiujie Wang Qifei Wang |
| author_sort | Yunbo Rao |
| collection | DOAJ |
| description | Currently, object detectors based on CNN, such as RetinaNet, Faster-RCNN, CornerNet series, can achieve good performance, but have some common drawbacks, like large calculation cost, high model complexity and slow detection speed. In this paper, a new lightweight object detector is proposed, which adopted a density-based approach to merge the real boxes. To reduce calculation cost and improve detection speed, the tactic of multi-scale output is adopted to predict objects of different sizes with features of different scales. Furthermore, a new lightweight network model is proposed, which can show better performance in computation, FPS, and model complexity. Meanwhile, the separation of convolution is used to improve the basic convolution layer, which can achieve better results under the same number of filters. In the experiments, we verified the capability of our methods based on ablation experiment and model evaluation, which demonstrates the superiority of our method. Moreover, we have also conducted deep network and multichannel experiments on MS-COCO2014 datasets and achieved 20.9% mAP performance. |
| first_indexed | 2024-12-14T23:11:59Z |
| format | Article |
| id | doaj.art-d3edfae310964fdb85e26db3898899b2 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-14T23:11:59Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-d3edfae310964fdb85e26db3898899b22022-12-21T22:44:10ZengIEEEIEEE Access2169-35362020-01-01820170020171210.1109/ACCESS.2020.30295929217510Light-Net: Lightweight Object DetectorYunbo Rao0https://orcid.org/0000-0001-5433-7379Guo Yi1https://orcid.org/0000-0002-5228-1567Junmin Xue2Jiansu Pu3https://orcid.org/0000-0002-4284-6958Jianping Gou4https://orcid.org/0000-0002-8438-7286Qiujie Wang5Qifei Wang6School of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu, ChinaSchool of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu, ChinaSchool of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu, ChinaSchool of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu, ChinaSchool of Computer Science and Telecommunication Engineering, Jiangsu University, Zhenjiang, ChinaSchool of Computer Science, Guangdong University of Technology, Guangzhou, ChinaDepartment of Electrical Engineering and Computer Sciences (EECS), University of California at Berkeley, Berkeley, CA, USACurrently, object detectors based on CNN, such as RetinaNet, Faster-RCNN, CornerNet series, can achieve good performance, but have some common drawbacks, like large calculation cost, high model complexity and slow detection speed. In this paper, a new lightweight object detector is proposed, which adopted a density-based approach to merge the real boxes. To reduce calculation cost and improve detection speed, the tactic of multi-scale output is adopted to predict objects of different sizes with features of different scales. Furthermore, a new lightweight network model is proposed, which can show better performance in computation, FPS, and model complexity. Meanwhile, the separation of convolution is used to improve the basic convolution layer, which can achieve better results under the same number of filters. In the experiments, we verified the capability of our methods based on ablation experiment and model evaluation, which demonstrates the superiority of our method. Moreover, we have also conducted deep network and multichannel experiments on MS-COCO2014 datasets and achieved 20.9% mAP performance.https://ieeexplore.ieee.org/document/9217510/Object detectionlight-netprior boxanchorobject detector |
| spellingShingle | Yunbo Rao Guo Yi Junmin Xue Jiansu Pu Jianping Gou Qiujie Wang Qifei Wang Light-Net: Lightweight Object Detector IEEE Access Object detection light-net prior box anchor object detector |
| title | Light-Net: Lightweight Object Detector |
| title_full | Light-Net: Lightweight Object Detector |
| title_fullStr | Light-Net: Lightweight Object Detector |
| title_full_unstemmed | Light-Net: Lightweight Object Detector |
| title_short | Light-Net: Lightweight Object Detector |
| title_sort | light net lightweight object detector |
| topic | Object detection light-net prior box anchor object detector |
| url | https://ieeexplore.ieee.org/document/9217510/ |
| work_keys_str_mv | AT yunborao lightnetlightweightobjectdetector AT guoyi lightnetlightweightobjectdetector AT junminxue lightnetlightweightobjectdetector AT jiansupu lightnetlightweightobjectdetector AT jianpinggou lightnetlightweightobjectdetector AT qiujiewang lightnetlightweightobjectdetector AT qifeiwang lightnetlightweightobjectdetector |