Standard

Application of a Neural Network to Predict the Trajectory of Objects Tracked in a Video Sequence. / Savel’ev, S. V.; Kosykh, V. P.

In: Optoelectronics, Instrumentation and Data Processing, Vol. 62, No. 3, 28.09.2026, p. 297-304.

Research output: Contribution to journal › Article › peer-review

Harvard

Savel’ev, SV & Kosykh, VP 2026, 'Application of a Neural Network to Predict the Trajectory of Objects Tracked in a Video Sequence', Optoelectronics, Instrumentation and Data Processing, vol. 62, no. 3, pp. 297-304. https://doi.org/10.3103/S8756699026700329

APA

Savel’ev, S. V., & Kosykh, V. P. (2026). Application of a Neural Network to Predict the Trajectory of Objects Tracked in a Video Sequence. Optoelectronics, Instrumentation and Data Processing, 62(3), 297-304. https://doi.org/10.3103/S8756699026700329

Vancouver

Savel’ev SV, Kosykh VP. Application of a Neural Network to Predict the Trajectory of Objects Tracked in a Video Sequence. Optoelectronics, Instrumentation and Data Processing. 2026 Sept 28;62(3):297-304. doi: 10.3103/S8756699026700329

Author

Savel’ev, S. V. ; Kosykh, V. P. / Application of a Neural Network to Predict the Trajectory of Objects Tracked in a Video Sequence. In: Optoelectronics, Instrumentation and Data Processing. 2026 ; Vol. 62, No. 3. pp. 297-304.

BibTeX

@article{86ee87058fda49d886d539f2c1de3983,
title = "Application of a Neural Network to Predict the Trajectory of Objects Tracked in a Video Sequence",
abstract = "The integration of computationally efficient neural networks into a multiple object tracking (MOT) algorithm in video sequences to reliably predict subsequent trajectories is studied. Several well-known neural networks are tested, and a new neural network is proposed. The neural network models demonstrate improved trajectory prediction performance relative to the ByteTrack baseline algorithm, without significantly impacting performance.",
keywords = "ByteTrack, multiple object tracking, neural network, tracking-by-detection, отслеживание множественных объектов, подход «отслеживание через обнаружение», нейронная сеть, ByteTrack",
author = "Savel{\textquoteright}ev, {S. V.} and Kosykh, {V. P.}",
note = "Savel{\textquoteright}ev, S.V., Kosykh, V.P. Application of a Neural Network to Predict the Trajectory of Objects Tracked in a Video Sequence. Optoelectron.Instrument.Proc. 62, 297–304 (2026). https://doi.org/10.3103/S8756699026700329 This work was supported by the Ministry of Science and Higher Education of the Russian Federation (project no. 124041700103-1).",
year = "2026",
month = sep,
day = "28",
doi = "10.3103/S8756699026700329",
language = "English",
volume = "62",
pages = "297--304",
journal = "Optoelectronics, Instrumentation and Data Processing",
issn = "8756-6990",
publisher = "Allerton Press Inc.",
number = "3",

}

RIS

TY - JOUR

T1 - Application of a Neural Network to Predict the Trajectory of Objects Tracked in a Video Sequence

AU - Savel’ev, S. V.

AU - Kosykh, V. P.

N1 - Savel’ev, S.V., Kosykh, V.P. Application of a Neural Network to Predict the Trajectory of Objects Tracked in a Video Sequence. Optoelectron.Instrument.Proc. 62, 297–304 (2026). https://doi.org/10.3103/S8756699026700329 This work was supported by the Ministry of Science and Higher Education of the Russian Federation (project no. 124041700103-1).

PY - 2026/9/28

Y1 - 2026/9/28

N2 - The integration of computationally efficient neural networks into a multiple object tracking (MOT) algorithm in video sequences to reliably predict subsequent trajectories is studied. Several well-known neural networks are tested, and a new neural network is proposed. The neural network models demonstrate improved trajectory prediction performance relative to the ByteTrack baseline algorithm, without significantly impacting performance.

AB - The integration of computationally efficient neural networks into a multiple object tracking (MOT) algorithm in video sequences to reliably predict subsequent trajectories is studied. Several well-known neural networks are tested, and a new neural network is proposed. The neural network models demonstrate improved trajectory prediction performance relative to the ByteTrack baseline algorithm, without significantly impacting performance.

KW - ByteTrack

KW - multiple object tracking

KW - neural network

KW - tracking-by-detection

KW - отслеживание множественных объектов

KW - подход «отслеживание через обнаружение»

KW - нейронная сеть

KW - ByteTrack

UR - https://www.mendeley.com/catalogue/e2f604a2-85ab-3562-b0bf-1f90b2c0ce61/

UR - https://www.scopus.com/pages/publications/105052129375

U2 - 10.3103/S8756699026700329

DO - 10.3103/S8756699026700329

M3 - Article

VL - 62

SP - 297

EP - 304

JO - Optoelectronics, Instrumentation and Data Processing

JF - Optoelectronics, Instrumentation and Data Processing

SN - 8756-6990

IS - 3

ER -

ID: 85032504