Cargando…

Spatial–Semantic and Temporal Attention Mechanism-Based Online Multi-Object Tracking

Multi-object tracking (MOT) plays a crucial role in various platforms. Occlusion and insertion among targets, complex backgrounds and higher real-time requirements increase the difficulty of MOT problems. Most state-of-the-art MOT approaches adopt the tracking-by-detection strategy, which relies on...

Descripción completa

Detalles Bibliográficos
Autores principales: Meng, Fanjie, Wang, Xinqing, Wang, Dong, Shao, Faming, Fu, Lei
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2020
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7146429/
https://www.ncbi.nlm.nih.gov/pubmed/32188090
http://dx.doi.org/10.3390/s20061653
_version_ 1783520200453259264
author Meng, Fanjie
Wang, Xinqing
Wang, Dong
Shao, Faming
Fu, Lei
author_facet Meng, Fanjie
Wang, Xinqing
Wang, Dong
Shao, Faming
Fu, Lei
author_sort Meng, Fanjie
collection PubMed
description Multi-object tracking (MOT) plays a crucial role in various platforms. Occlusion and insertion among targets, complex backgrounds and higher real-time requirements increase the difficulty of MOT problems. Most state-of-the-art MOT approaches adopt the tracking-by-detection strategy, which relies on compute-intensive sliding windows or anchoring schemes to detect matching targets or candidates in each frame. In this work, we introduce a more efficient and effective spatial–temporal attention scheme to track multiple objects in various scenarios. Using a semantic-feature-based spatial attention mechanism and a novel Motion Model, we address the insertion and location of candidates. Some online-learned target-specific convolutional neural networks (CNNs) were used to estimate target occlusion and classify by adapting the appearance model. A temporal attention mechanism was adopted to update the online module by balancing current and history frames. Extensive experiments were performed on Karlsruhe Institute of Technologyand Toyota Technological Institute (KITTI) benchmarks and an Armored Target Tracking Dataset (ATTD) built for ground-armored targets. Experimental results show that the proposed method achieved outstanding tracking performance and met the actual application requirements.
format Online
Article
Text
id pubmed-7146429
institution National Center for Biotechnology Information
language English
publishDate 2020
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-71464292020-04-15 Spatial–Semantic and Temporal Attention Mechanism-Based Online Multi-Object Tracking Meng, Fanjie Wang, Xinqing Wang, Dong Shao, Faming Fu, Lei Sensors (Basel) Article Multi-object tracking (MOT) plays a crucial role in various platforms. Occlusion and insertion among targets, complex backgrounds and higher real-time requirements increase the difficulty of MOT problems. Most state-of-the-art MOT approaches adopt the tracking-by-detection strategy, which relies on compute-intensive sliding windows or anchoring schemes to detect matching targets or candidates in each frame. In this work, we introduce a more efficient and effective spatial–temporal attention scheme to track multiple objects in various scenarios. Using a semantic-feature-based spatial attention mechanism and a novel Motion Model, we address the insertion and location of candidates. Some online-learned target-specific convolutional neural networks (CNNs) were used to estimate target occlusion and classify by adapting the appearance model. A temporal attention mechanism was adopted to update the online module by balancing current and history frames. Extensive experiments were performed on Karlsruhe Institute of Technologyand Toyota Technological Institute (KITTI) benchmarks and an Armored Target Tracking Dataset (ATTD) built for ground-armored targets. Experimental results show that the proposed method achieved outstanding tracking performance and met the actual application requirements. MDPI 2020-03-16 /pmc/articles/PMC7146429/ /pubmed/32188090 http://dx.doi.org/10.3390/s20061653 Text en © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Meng, Fanjie
Wang, Xinqing
Wang, Dong
Shao, Faming
Fu, Lei
Spatial–Semantic and Temporal Attention Mechanism-Based Online Multi-Object Tracking
title Spatial–Semantic and Temporal Attention Mechanism-Based Online Multi-Object Tracking
title_full Spatial–Semantic and Temporal Attention Mechanism-Based Online Multi-Object Tracking
title_fullStr Spatial–Semantic and Temporal Attention Mechanism-Based Online Multi-Object Tracking
title_full_unstemmed Spatial–Semantic and Temporal Attention Mechanism-Based Online Multi-Object Tracking
title_short Spatial–Semantic and Temporal Attention Mechanism-Based Online Multi-Object Tracking
title_sort spatial–semantic and temporal attention mechanism-based online multi-object tracking
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7146429/
https://www.ncbi.nlm.nih.gov/pubmed/32188090
http://dx.doi.org/10.3390/s20061653
work_keys_str_mv AT mengfanjie spatialsemanticandtemporalattentionmechanismbasedonlinemultiobjecttracking
AT wangxinqing spatialsemanticandtemporalattentionmechanismbasedonlinemultiobjecttracking
AT wangdong spatialsemanticandtemporalattentionmechanismbasedonlinemultiobjecttracking
AT shaofaming spatialsemanticandtemporalattentionmechanismbasedonlinemultiobjecttracking
AT fulei spatialsemanticandtemporalattentionmechanismbasedonlinemultiobjecttracking