Two layer network flow for fast data association on multi object tracking

Bariqi Abdillah, Grafika Jati, Wisnu Jatmiko, Adi Nurhadiyatna

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Multi object tracking is one interesting topics of computer science that has many applications, such as surveillance system, navigation robot, sports analysis, autonomous driving car, and others. One of the main task of multi-object tracking is data association. This study discusses data association on multi-object tracking and its completion with a two-layer network flow approach. Notice that each object on a frame as a node, then there is an edge connecting each node from one frame to other frame and then finding for the set of edges that provide the greatest probability of transition from one frame to the next, or in the optimization problem better known as max-cost network flow. The probability calculation is based on position distance and similarity feature between objects. The data used in this research is 2DMOT2015. The proposed method obtains highly competitive MOTA of 20.1% compared to existing method with fast computation speed by 215.8 fps.

Original languageEnglish
Title of host publication2018 International Conference on Advanced Computer Science and Information Systems, ICACSIS 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages373-378
Number of pages6
ISBN (Electronic)9781728101354
DOIs
Publication statusPublished - 17 Jan 2019
Event10th International Conference on Advanced Computer Science and Information Systems, ICACSIS 2018 - Yogyakarta, Indonesia
Duration: 27 Oct 201828 Oct 2018

Publication series

Name2018 International Conference on Advanced Computer Science and Information Systems, ICACSIS 2018

Conference

Conference10th International Conference on Advanced Computer Science and Information Systems, ICACSIS 2018
CountryIndonesia
CityYogyakarta
Period27/10/1828/10/18

Keywords

  • Convolutional Neural Network
  • Data Association
  • Multi Object Tracking
  • Network Flow

Fingerprint Dive into the research topics of 'Two layer network flow for fast data association on multi object tracking'. Together they form a unique fingerprint.

  • Cite this

    Abdillah, B., Jati, G., Jatmiko, W., & Nurhadiyatna, A. (2019). Two layer network flow for fast data association on multi object tracking. In 2018 International Conference on Advanced Computer Science and Information Systems, ICACSIS 2018 (pp. 373-378). [8618161] (2018 International Conference on Advanced Computer Science and Information Systems, ICACSIS 2018). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICACSIS.2018.8618161