STARNet: spatio-temporal aware recurrent network for efficient video object detection on embedded devices

IF 2.4 4区 计算机科学 Q3 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE Machine Vision and Applications Pub Date : 2024-01-31 DOI:10.1007/s00138-023-01504-0
Mohammad Hajizadeh, Mohammad Sabokrou, Adel Rahmani
{"title":"STARNet: spatio-temporal aware recurrent network for efficient video object detection on embedded devices","authors":"Mohammad Hajizadeh, Mohammad Sabokrou, Adel Rahmani","doi":"10.1007/s00138-023-01504-0","DOIUrl":null,"url":null,"abstract":"<p>The challenge of converting various object detection methods from image to video remains unsolved. When applied to video, image methods frequently fail to generalize effectively due to issues, such as blurriness, different and unclear positions, low quality, and other relevant issues. Additionally, the lack of a good long-term memory in video object detection presents an additional challenge. In the majority of instances, the outputs of successive frames are known to be quite similar; therefore, this fact is relied upon. Furthermore, the information contained in a series of successive or non-successive frames is greater than that contained in a single frame. In this study, we present a novel recurrent cell for feature propagation and identify the optimal location of layers to increase the memory interval. As a result, we achieved higher accuracy compared to other proposed methods in other studies. Hardware limitations can exacerbate this challenge. The paper aims to implement and increase the efficiency of the methods on embedded devices. We achieved 68.7% <i>mAP</i> accuracy on the ImageNet VID dataset for embedded devices in real-time and at a speed of 52 <i>fps</i>.</p>","PeriodicalId":51116,"journal":{"name":"Machine Vision and Applications","volume":"32 1","pages":""},"PeriodicalIF":2.4000,"publicationDate":"2024-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Machine Vision and Applications","FirstCategoryId":"94","ListUrlMain":"https://doi.org/10.1007/s00138-023-01504-0","RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}
引用次数: 0

Abstract

The challenge of converting various object detection methods from image to video remains unsolved. When applied to video, image methods frequently fail to generalize effectively due to issues, such as blurriness, different and unclear positions, low quality, and other relevant issues. Additionally, the lack of a good long-term memory in video object detection presents an additional challenge. In the majority of instances, the outputs of successive frames are known to be quite similar; therefore, this fact is relied upon. Furthermore, the information contained in a series of successive or non-successive frames is greater than that contained in a single frame. In this study, we present a novel recurrent cell for feature propagation and identify the optimal location of layers to increase the memory interval. As a result, we achieved higher accuracy compared to other proposed methods in other studies. Hardware limitations can exacerbate this challenge. The paper aims to implement and increase the efficiency of the methods on embedded devices. We achieved 68.7% mAP accuracy on the ImageNet VID dataset for embedded devices in real-time and at a speed of 52 fps.

Abstract Image

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
STARNet:用于嵌入式设备上高效视频对象检测的时空感知递归网络
将各种物体检测方法从图像转换到视频的难题仍未解决。将图像方法应用于视频时,由于图像模糊、位置不同和不清晰、质量低等相关问题,经常无法有效推广。此外,视频对象检测缺乏良好的长期记忆也是一个额外的挑战。在大多数情况下,众所周知连续帧的输出非常相似,因此需要依赖这一事实。此外,一系列连续或非连续帧所包含的信息量要大于单帧所包含的信息量。在本研究中,我们提出了一种用于特征传播的新型递归单元,并确定了层的最佳位置以增加记忆间隔。因此,与其他研究中提出的方法相比,我们实现了更高的准确度。硬件限制会加剧这一挑战。本文的目的是在嵌入式设备上实现这些方法并提高其效率。我们在嵌入式设备的 ImageNet VID 数据集上以 52 fps 的速度实时实现了 68.7% 的 mAP 准确率。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Machine Vision and Applications
Machine Vision and Applications 工程技术-工程:电子与电气
CiteScore
6.30
自引率
3.00%
发文量
84
审稿时长
8.7 months
期刊介绍: Machine Vision and Applications publishes high-quality technical contributions in machine vision research and development. Specifically, the editors encourage submittals in all applications and engineering aspects of image-related computing. In particular, original contributions dealing with scientific, commercial, industrial, military, and biomedical applications of machine vision, are all within the scope of the journal. Particular emphasis is placed on engineering and technology aspects of image processing and computer vision. The following aspects of machine vision applications are of interest: algorithms, architectures, VLSI implementations, AI techniques and expert systems for machine vision, front-end sensing, multidimensional and multisensor machine vision, real-time techniques, image databases, virtual reality and visualization. Papers must include a significant experimental validation component.
期刊最新文献
A novel key point based ROI segmentation and image captioning using guidance information Specular Surface Detection with Deep Static Specular Flow and Highlight Removing cloud shadows from ground-based solar imagery Underwater image object detection based on multi-scale feature fusion Object Recognition Consistency in Regression for Active Detection
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1