Video object detection via space–time feature aggregation and result reuse

Abstract When detecting the objects in videos, motion always leads to object deterioration, like blurring and occlusion, as well as the strange state of the object's shape and posture. Consequently, the detection of video frames will lead to a decline in accuracy by using the image object detec...

Full description

Saved in:
Bibliographic Details
Main Authors: Liang Duan, Rongfei Yang, Kun Yue, Zhengbao Sun, Guowu Yuan
Format: Article
Language:English
Published: Wiley 2024-10-01
Series:IET Image Processing
Subjects:
Online Access:https://doi.org/10.1049/ipr2.13179
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850200432874356736
author Liang Duan
Rongfei Yang
Kun Yue
Zhengbao Sun
Guowu Yuan
author_facet Liang Duan
Rongfei Yang
Kun Yue
Zhengbao Sun
Guowu Yuan
author_sort Liang Duan
collection DOAJ
description Abstract When detecting the objects in videos, motion always leads to object deterioration, like blurring and occlusion, as well as the strange state of the object's shape and posture. Consequently, the detection of video frames will lead to a decline in accuracy by using the image object detection model. This paper proposes an online video object detection method based on the one‐stage detector YOLOx. First, the module for space–time feature aggregation is given, which uses the space–time information of past frames to enhance the feature quality of the current frame. Then, the module for result reuse is given, which incorporates the detection results of past frames to improve the detection stability of the current frame. By these two modules, the trade‐off between accuracy and speed of video object detection could be achieved. Experimental results on the ImageNet VID show the improvement of speed and accuracy of the proposed method.
format Article
id doaj-art-b12e0c4d9fa0406d85ba4a3b2ebc7951
institution OA Journals
issn 1751-9659
1751-9667
language English
publishDate 2024-10-01
publisher Wiley
record_format Article
series IET Image Processing
spelling doaj-art-b12e0c4d9fa0406d85ba4a3b2ebc79512025-08-20T02:12:20ZengWileyIET Image Processing1751-96591751-96672024-10-0118123356336710.1049/ipr2.13179Video object detection via space–time feature aggregation and result reuseLiang Duan0Rongfei Yang1Kun Yue2Zhengbao Sun3Guowu Yuan4School of Information Science and Engineering Yunnan University Kunming ChinaSchool of Information Science and Engineering Yunnan University Kunming ChinaSchool of Information Science and Engineering Yunnan University Kunming ChinaSchool of Engineering Yunnan University Kunming ChinaSchool of Information Science and Engineering Yunnan University Kunming ChinaAbstract When detecting the objects in videos, motion always leads to object deterioration, like blurring and occlusion, as well as the strange state of the object's shape and posture. Consequently, the detection of video frames will lead to a decline in accuracy by using the image object detection model. This paper proposes an online video object detection method based on the one‐stage detector YOLOx. First, the module for space–time feature aggregation is given, which uses the space–time information of past frames to enhance the feature quality of the current frame. Then, the module for result reuse is given, which incorporates the detection results of past frames to improve the detection stability of the current frame. By these two modules, the trade‐off between accuracy and speed of video object detection could be achieved. Experimental results on the ImageNet VID show the improvement of speed and accuracy of the proposed method.https://doi.org/10.1049/ipr2.13179feature extractionobject detection
spellingShingle Liang Duan
Rongfei Yang
Kun Yue
Zhengbao Sun
Guowu Yuan
Video object detection via space–time feature aggregation and result reuse
IET Image Processing
feature extraction
object detection
title Video object detection via space–time feature aggregation and result reuse
title_full Video object detection via space–time feature aggregation and result reuse
title_fullStr Video object detection via space–time feature aggregation and result reuse
title_full_unstemmed Video object detection via space–time feature aggregation and result reuse
title_short Video object detection via space–time feature aggregation and result reuse
title_sort video object detection via space time feature aggregation and result reuse
topic feature extraction
object detection
url https://doi.org/10.1049/ipr2.13179
work_keys_str_mv AT liangduan videoobjectdetectionviaspacetimefeatureaggregationandresultreuse
AT rongfeiyang videoobjectdetectionviaspacetimefeatureaggregationandresultreuse
AT kunyue videoobjectdetectionviaspacetimefeatureaggregationandresultreuse
AT zhengbaosun videoobjectdetectionviaspacetimefeatureaggregationandresultreuse
AT guowuyuan videoobjectdetectionviaspacetimefeatureaggregationandresultreuse