Multi-Object Tracking with Predictive Information Fusion and Adaptive Measurement Noise
Multi-object tracking (MOT) aims to detect objects in video sequences and associate them across frames. Currently, the mainstream research direction regarding MOT is the tracking-by-detection (TBD) framework. Tracking results are highly sensitive to detection outputs, and challenges from object occl...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/15/2/736 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832589168246521856 |
|---|---|
| author | Xiaohui Cheng Haoyi Zhao Yun Deng Shuangqin Shen |
| author_facet | Xiaohui Cheng Haoyi Zhao Yun Deng Shuangqin Shen |
| author_sort | Xiaohui Cheng |
| collection | DOAJ |
| description | Multi-object tracking (MOT) aims to detect objects in video sequences and associate them across frames. Currently, the mainstream research direction regarding MOT is the tracking-by-detection (TBD) framework. Tracking results are highly sensitive to detection outputs, and challenges from object occlusion and complex motion present significant obstacles in the field of MOT. To reduce dependence on detection outputs, we propose a method that integrates predictive information to improve Non-Maximum Suppression (NMS). By applying secondary modulation to the suppression scores and dynamically adjusting the suppression threshold using tracking information, our method better retains candidate boxes for occluded objects. Furthermore, to track occluding and overlapping objects more effectively, we introduce an adaptive measurement noise method that adjusts the measurement noise to mitigate the impact of object occlusion or overlap on tracking accuracy. Additionally, we enhance the affinity matrix in the association algorithm by incorporating height information, thereby improving the stability of complex moving objects. Our method outperforms the baseline model ByteTrack on the DanceTrack dataset, increasing Higher Order Tracking Accuracy (HOTA), Multi-Object Tracking Accuracy (MOTA), and the ID F1 Score (IDF1) by 10.2%, 3.0%, and 4.8%, respectively. |
| format | Article |
| id | doaj-art-1e882235c3b9480e9b0e67f4c328a186 |
| institution | Kabale University |
| issn | 2076-3417 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj-art-1e882235c3b9480e9b0e67f4c328a1862025-01-24T13:20:39ZengMDPI AGApplied Sciences2076-34172025-01-0115273610.3390/app15020736Multi-Object Tracking with Predictive Information Fusion and Adaptive Measurement NoiseXiaohui Cheng0Haoyi Zhao1Yun Deng2Shuangqin Shen3School of Computer Science and Engineering, Guilin University of Technology, Guilin 541004, ChinaSchool of Computer Science and Engineering, Guilin University of Technology, Guilin 541004, ChinaSchool of Computer Science and Engineering, Guilin University of Technology, Guilin 541004, ChinaSchool of Physics and Electronic Information Engineering, Guilin University of Technology, Guilin 541004, ChinaMulti-object tracking (MOT) aims to detect objects in video sequences and associate them across frames. Currently, the mainstream research direction regarding MOT is the tracking-by-detection (TBD) framework. Tracking results are highly sensitive to detection outputs, and challenges from object occlusion and complex motion present significant obstacles in the field of MOT. To reduce dependence on detection outputs, we propose a method that integrates predictive information to improve Non-Maximum Suppression (NMS). By applying secondary modulation to the suppression scores and dynamically adjusting the suppression threshold using tracking information, our method better retains candidate boxes for occluded objects. Furthermore, to track occluding and overlapping objects more effectively, we introduce an adaptive measurement noise method that adjusts the measurement noise to mitigate the impact of object occlusion or overlap on tracking accuracy. Additionally, we enhance the affinity matrix in the association algorithm by incorporating height information, thereby improving the stability of complex moving objects. Our method outperforms the baseline model ByteTrack on the DanceTrack dataset, increasing Higher Order Tracking Accuracy (HOTA), Multi-Object Tracking Accuracy (MOTA), and the ID F1 Score (IDF1) by 10.2%, 3.0%, and 4.8%, respectively.https://www.mdpi.com/2076-3417/15/2/736multi-object trackingtracking-by-detectionocclusioncomplex motion |
| spellingShingle | Xiaohui Cheng Haoyi Zhao Yun Deng Shuangqin Shen Multi-Object Tracking with Predictive Information Fusion and Adaptive Measurement Noise Applied Sciences multi-object tracking tracking-by-detection occlusion complex motion |
| title | Multi-Object Tracking with Predictive Information Fusion and Adaptive Measurement Noise |
| title_full | Multi-Object Tracking with Predictive Information Fusion and Adaptive Measurement Noise |
| title_fullStr | Multi-Object Tracking with Predictive Information Fusion and Adaptive Measurement Noise |
| title_full_unstemmed | Multi-Object Tracking with Predictive Information Fusion and Adaptive Measurement Noise |
| title_short | Multi-Object Tracking with Predictive Information Fusion and Adaptive Measurement Noise |
| title_sort | multi object tracking with predictive information fusion and adaptive measurement noise |
| topic | multi-object tracking tracking-by-detection occlusion complex motion |
| url | https://www.mdpi.com/2076-3417/15/2/736 |
| work_keys_str_mv | AT xiaohuicheng multiobjecttrackingwithpredictiveinformationfusionandadaptivemeasurementnoise AT haoyizhao multiobjecttrackingwithpredictiveinformationfusionandadaptivemeasurementnoise AT yundeng multiobjecttrackingwithpredictiveinformationfusionandadaptivemeasurementnoise AT shuangqinshen multiobjecttrackingwithpredictiveinformationfusionandadaptivemeasurementnoise |