Dynamic Object Detection Algorithm Based on Lightweight Shared Feature Pyramid

Current object detection algorithms perform inference on all samples at a fixed computational cost in the inference stage, which wastes computing resources and is not flexible. To solve this problem, a dynamic object detection algorithm based on a lightweight shared feature pyramid is proposed, whic...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Li Zhu, Zihao Xie, Jing Luo, Yuhang Qi, Liman Liu, Wenbing Tao
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	object detection adaptive inference shared feature pyramid Science Q
Acceso en línea:	https://doaj.org/article/3598857755d24172bce0127b86ff8af9
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:3598857755d24172bce0127b86ff8af9
record_format	dspace
spelling	oai:doaj.org-article:3598857755d24172bce0127b86ff8af92021-11-25T18:54:43ZDynamic Object Detection Algorithm Based on Lightweight Shared Feature Pyramid10.3390/rs132246102072-4292https://doaj.org/article/3598857755d24172bce0127b86ff8af92021-11-01T00:00:00Zhttps://www.mdpi.com/2072-4292/13/22/4610https://doaj.org/toc/2072-4292Current object detection algorithms perform inference on all samples at a fixed computational cost in the inference stage, which wastes computing resources and is not flexible. To solve this problem, a dynamic object detection algorithm based on a lightweight shared feature pyramid is proposed, which performs adaptive inference according to computing resources and the difficulty of samples, greatly improving the efficiency of inference. Specifically, a lightweight shared feature pyramid network and lightweight detection head is proposed to reduce the amount of computation and parameters in the feature fusion part and detection head of the dynamic object detection model. On the PASCAL VOC dataset, under the two conditions of “anytime prediction” and “budgeted batch object detection”, the performance, computation amount and parameter amount are better than the dynamic object detection models constructed by networks such as ResNet, DenseNet and MSDNet.Li ZhuZihao XieJing LuoYuhang QiLiman LiuWenbing TaoMDPI AGarticleobject detectionadaptive inferenceshared feature pyramidScienceQENRemote Sensing, Vol 13, Iss 4610, p 4610 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	object detection adaptive inference shared feature pyramid Science Q
spellingShingle	object detection adaptive inference shared feature pyramid Science Q Li Zhu Zihao Xie Jing Luo Yuhang Qi Liman Liu Wenbing Tao Dynamic Object Detection Algorithm Based on Lightweight Shared Feature Pyramid
description	Current object detection algorithms perform inference on all samples at a fixed computational cost in the inference stage, which wastes computing resources and is not flexible. To solve this problem, a dynamic object detection algorithm based on a lightweight shared feature pyramid is proposed, which performs adaptive inference according to computing resources and the difficulty of samples, greatly improving the efficiency of inference. Specifically, a lightweight shared feature pyramid network and lightweight detection head is proposed to reduce the amount of computation and parameters in the feature fusion part and detection head of the dynamic object detection model. On the PASCAL VOC dataset, under the two conditions of “anytime prediction” and “budgeted batch object detection”, the performance, computation amount and parameter amount are better than the dynamic object detection models constructed by networks such as ResNet, DenseNet and MSDNet.
format	article
author	Li Zhu Zihao Xie Jing Luo Yuhang Qi Liman Liu Wenbing Tao
author_facet	Li Zhu Zihao Xie Jing Luo Yuhang Qi Liman Liu Wenbing Tao
author_sort	Li Zhu
title	Dynamic Object Detection Algorithm Based on Lightweight Shared Feature Pyramid
title_short	Dynamic Object Detection Algorithm Based on Lightweight Shared Feature Pyramid
title_full	Dynamic Object Detection Algorithm Based on Lightweight Shared Feature Pyramid
title_fullStr	Dynamic Object Detection Algorithm Based on Lightweight Shared Feature Pyramid
title_full_unstemmed	Dynamic Object Detection Algorithm Based on Lightweight Shared Feature Pyramid
title_sort	dynamic object detection algorithm based on lightweight shared feature pyramid
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/3598857755d24172bce0127b86ff8af9
work_keys_str_mv	AT lizhu dynamicobjectdetectionalgorithmbasedonlightweightsharedfeaturepyramid AT zihaoxie dynamicobjectdetectionalgorithmbasedonlightweightsharedfeaturepyramid AT jingluo dynamicobjectdetectionalgorithmbasedonlightweightsharedfeaturepyramid AT yuhangqi dynamicobjectdetectionalgorithmbasedonlightweightsharedfeaturepyramid AT limanliu dynamicobjectdetectionalgorithmbasedonlightweightsharedfeaturepyramid AT wenbingtao dynamicobjectdetectionalgorithmbasedonlightweightsharedfeaturepyramid
_version_	1718410515485556736

Dynamic Object Detection Algorithm Based on Lightweight Shared Feature Pyramid

Ejemplares similares