|
NUMERICAL METHODS AND DATA ANALYSIS
Development of a multi-object tracking algorithm with untrained features of object matching
A. G. Vadim, V. F. Kalugin State Research Institute of Aviation Systems
Abstract:
The problem of multiple object tracking is one of the most difficult tasks in computer vision. The article is devoted to a task of multiple object tracking on video footage received from an unmanned aerial vehicle. Unlike a static camera platform, the mobile platform causes an accidental camera movement, which leads to sudden changes in the position, angle and scale of objects. Such aspects considerably hinder efficient object tracking. In this paper, we explore the possibilities of improving the tracking quality in the case of camera movements. We significantly outperform ByteTrack algorithm, one of the best tracking algorithms for the MOT Challenge dataset, on the Visdrone 2019 dataset.
Keywords:
multiple object tracking, YOLO v5, ByteTrack, Kalman filter, Visdrone 2019, UAV
Received: 11.01.2023 Accepted: 18.06.2023
Citation:
A. G. Vadim, V. F. Kalugin, “Development of a multi-object tracking algorithm with untrained features of object matching”, Computer Optics, 47:6 (2023), 1002–1010
Linking options:
https://www.mathnet.ru/eng/co1204 https://www.mathnet.ru/eng/co/v47/i6/p1002
|
Statistics & downloads: |
Abstract page: | 7 | Full-text PDF : | 14 |
|