On the study of predictors in Single Shot Multibox Detector

Xuemei Xie, Xun Xu, Lihua Ma, Guangming Shi, Pengfei Chen
{"title":"On the study of predictors in Single Shot Multibox Detector","authors":"Xuemei Xie, Xun Xu, Lihua Ma, Guangming Shi, Pengfei Chen","doi":"10.1145/3177404.3177412","DOIUrl":null,"url":null,"abstract":"Single shot multibox detector (SSD) is a state-of-the-art network for real-time object detection. It is originally designed for general datasets. While, for specific datasets, their distribution of ground truth boxes is somehow different and thus, SSD shows unsatisfying performance. In this paper, we improve the performance of SSD on specific datasets. We first dissect the mechanism of predictors, the predicting parameters of a potential detection, in two aspects: classification and localization. Then we reveal the relationship between default boxes and predictors. With this point we finally make an improvement on default box setting and achieve a higher mAP over the original SSD on specific datasets.","PeriodicalId":133378,"journal":{"name":"Proceedings of the International Conference on Video and Image Processing","volume":"116 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the International Conference on Video and Image Processing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3177404.3177412","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3

Abstract

Single shot multibox detector (SSD) is a state-of-the-art network for real-time object detection. It is originally designed for general datasets. While, for specific datasets, their distribution of ground truth boxes is somehow different and thus, SSD shows unsatisfying performance. In this paper, we improve the performance of SSD on specific datasets. We first dissect the mechanism of predictors, the predicting parameters of a potential detection, in two aspects: classification and localization. Then we reveal the relationship between default boxes and predictors. With this point we finally make an improvement on default box setting and achieve a higher mAP over the original SSD on specific datasets.
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
单发多盒探测器中预测器的研究
单镜头多盒探测器(SSD)是一种先进的实时目标检测网络。它最初是为通用数据集设计的。然而,对于特定的数据集,它们的地面真值盒的分布在某种程度上是不同的,因此,SSD表现出不令人满意的性能。在本文中,我们改进了SSD在特定数据集上的性能。我们首先从分类和定位两个方面剖析了预测器的机制,即潜在检测的预测参数。然后我们揭示默认框和预测器之间的关系。在这一点上,我们最终对默认框设置进行了改进,并在特定数据集上实现了比原始SSD更高的mAP。
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
自引率
0.00%
发文量
0
期刊最新文献
Exposing Video Forgeries by Detecting Misaligned Double Compression Weld Defect Detection based on Completed Local Ternary Patterns Faster R-CNN for Marine Organism Detection and Recognition Using Data Augmentation Application Research of the Laser and Digital Image Processing in Bridge Monitoring Signal Processing of Ultrasonic Testing of Hardened Layer Depth Based on Wavelet Transform Theory
×
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