Skip to main navigation Skip to search Skip to main content

Research on Video Pedestrian Tracking Based on the Combination of Optical Flow Method and Target Tracking Network

  • Harbin Institute of Technology
  • St. Petersburg State University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Target tracking is designed to predict the position of a given target in each video frame, and it has a wide range of applications in the fields of robot vision, video surveillance, and unmanned driving. This paper mainly studies the challenges of interference and target occlusion of similarities in target tracking, takes pedestrians as the main research target, combines the optical flow algorithm with the target tracking network, and improves the optical flow algorithm and the target tracking model, thereby improving the robustness of the tracking algorithm. Extensive experiments and comparative tests show the effectiveness of our method.

Original languageEnglish
Title of host publicationProceedings of the 7th International Scientific Conference on Intelligent Information Technologies for Industry (IITI’23) - Volume 1
EditorsSergey Kovalev, Andrey Sukhanov, Igor Kotenko
PublisherSpringer Science and Business Media Deutschland GmbH
Pages120-135
Number of pages16
ISBN (Print)9783031437885
DOIs
StatePublished - 2023
Event7th International Scientific Conference on Intelligent Information Technologies for Industry, IITI 2023 - St. Petersburg, Russian Federation
Duration: 25 Sep 202330 Sep 2023

Publication series

NameLecture Notes in Networks and Systems
Volume776 LNNS
ISSN (Print)2367-3370
ISSN (Electronic)2367-3389

Conference

Conference7th International Scientific Conference on Intelligent Information Technologies for Industry, IITI 2023
Country/TerritoryRussian Federation
CitySt. Petersburg
Period25/09/2330/09/23

Keywords

  • Diffusion equation
  • Self-attention mechanism
  • Target tracking
  • Variational optical flow

Fingerprint

Dive into the research topics of 'Research on Video Pedestrian Tracking Based on the Combination of Optical Flow Method and Target Tracking Network'. Together they form a unique fingerprint.

Cite this