Multimodal Multiobject Tracking by Fusing Deep Appearance Features and Motion Information
Multiobject Tracking (MOT) is one of the most important abilities of autonomous driving systems. However, most of the existing MOT methods only use a single sensor, such as a camera, which has the problem of insufficient reliability. In this paper, we propose a novel Multiobject Tracking method by f...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2020-01-01
|
| Series: | Complexity |
| Online Access: | http://dx.doi.org/10.1155/2020/8810340 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832567953066819584 |
|---|---|
| author | Liwei Zhang Jiahong Lai Zenghui Zhang Zhen Deng Bingwei He Yucheng He |
| author_facet | Liwei Zhang Jiahong Lai Zenghui Zhang Zhen Deng Bingwei He Yucheng He |
| author_sort | Liwei Zhang |
| collection | DOAJ |
| description | Multiobject Tracking (MOT) is one of the most important abilities of autonomous driving systems. However, most of the existing MOT methods only use a single sensor, such as a camera, which has the problem of insufficient reliability. In this paper, we propose a novel Multiobject Tracking method by fusing deep appearance features and motion information of objects. In this method, the locations of objects are first determined based on a 2D object detector and a 3D object detector. We use the Nonmaximum Suppression (NMS) algorithm to combine the detection results of the two detectors to ensure the detection accuracy in complex scenes. After that, we use Convolutional Neural Network (CNN) to learn the deep appearance features of objects and employ Kalman Filter to obtain the motion information of objects. Finally, the MOT task is achieved by associating the motion information and deep appearance features. A successful match indicates that the object was tracked successfully. A set of experiments on the KITTI Tracking Benchmark shows that the proposed MOT method can effectively perform the MOT task. The Multiobject Tracking Accuracy (MOTA) is up to 76.40% and the Multiobject Tracking Precision (MOTP) is up to 83.50%. |
| format | Article |
| id | doaj-art-9e7fae2b919f451bbfdb2373b9fa07a2 |
| institution | Kabale University |
| issn | 1076-2787 1099-0526 |
| language | English |
| publishDate | 2020-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Complexity |
| spelling | doaj-art-9e7fae2b919f451bbfdb2373b9fa07a22025-02-03T01:00:08ZengWileyComplexity1076-27871099-05262020-01-01202010.1155/2020/88103408810340Multimodal Multiobject Tracking by Fusing Deep Appearance Features and Motion InformationLiwei Zhang0Jiahong Lai1Zenghui Zhang2Zhen Deng3Bingwei He4Yucheng He5School of Mechanical Engineering and Automation, Fuzhou University, Fuzhou, ChinaSchool of Mechanical Engineering and Automation, Fuzhou University, Fuzhou, ChinaSchool of Mechanical Engineering and Automation, Fuzhou University, Fuzhou, ChinaSchool of Mechanical Engineering and Automation, Fuzhou University, Fuzhou, ChinaSchool of Mechanical Engineering and Automation, Fuzhou University, Fuzhou, ChinaThe T Stone Robotics Institute, Department of Mechanical and Automation Engineering, The Chinese University of Hong Kong, Hong Kong, ChinaMultiobject Tracking (MOT) is one of the most important abilities of autonomous driving systems. However, most of the existing MOT methods only use a single sensor, such as a camera, which has the problem of insufficient reliability. In this paper, we propose a novel Multiobject Tracking method by fusing deep appearance features and motion information of objects. In this method, the locations of objects are first determined based on a 2D object detector and a 3D object detector. We use the Nonmaximum Suppression (NMS) algorithm to combine the detection results of the two detectors to ensure the detection accuracy in complex scenes. After that, we use Convolutional Neural Network (CNN) to learn the deep appearance features of objects and employ Kalman Filter to obtain the motion information of objects. Finally, the MOT task is achieved by associating the motion information and deep appearance features. A successful match indicates that the object was tracked successfully. A set of experiments on the KITTI Tracking Benchmark shows that the proposed MOT method can effectively perform the MOT task. The Multiobject Tracking Accuracy (MOTA) is up to 76.40% and the Multiobject Tracking Precision (MOTP) is up to 83.50%.http://dx.doi.org/10.1155/2020/8810340 |
| spellingShingle | Liwei Zhang Jiahong Lai Zenghui Zhang Zhen Deng Bingwei He Yucheng He Multimodal Multiobject Tracking by Fusing Deep Appearance Features and Motion Information Complexity |
| title | Multimodal Multiobject Tracking by Fusing Deep Appearance Features and Motion Information |
| title_full | Multimodal Multiobject Tracking by Fusing Deep Appearance Features and Motion Information |
| title_fullStr | Multimodal Multiobject Tracking by Fusing Deep Appearance Features and Motion Information |
| title_full_unstemmed | Multimodal Multiobject Tracking by Fusing Deep Appearance Features and Motion Information |
| title_short | Multimodal Multiobject Tracking by Fusing Deep Appearance Features and Motion Information |
| title_sort | multimodal multiobject tracking by fusing deep appearance features and motion information |
| url | http://dx.doi.org/10.1155/2020/8810340 |
| work_keys_str_mv | AT liweizhang multimodalmultiobjecttrackingbyfusingdeepappearancefeaturesandmotioninformation AT jiahonglai multimodalmultiobjecttrackingbyfusingdeepappearancefeaturesandmotioninformation AT zenghuizhang multimodalmultiobjecttrackingbyfusingdeepappearancefeaturesandmotioninformation AT zhendeng multimodalmultiobjecttrackingbyfusingdeepappearancefeaturesandmotioninformation AT bingweihe multimodalmultiobjecttrackingbyfusingdeepappearancefeaturesandmotioninformation AT yuchenghe multimodalmultiobjecttrackingbyfusingdeepappearancefeaturesandmotioninformation |