arrow
返回

Learning multi-scale features for foreground segmentation

delete2019-08-31
delete133
delete
OA
AI
L
Long Ang Lim
H
Hacer Yalım Keleş *
DOI:10.1007/s10044-019-00845-9delete
delete原文链接
delete分享
delete收藏
查看原文
摘要

摘要

En 中文
Foreground segmentation algorithms aim at segmenting moving objects from the background in a robust way under various challenging scenarios. Encoder-decoder-type deep neural networks that are used in this domain recently perform impressive segmentation results. In this work, we propose a variation of our formerly proposed method (Anonymous 2018) that can be trained end-to-end using only a few training examples. The proposed method extends the feature pooling module of FgSegNet by introducing fusion of features inside this module, which is capable of extracting multi-scale features within images, resulting in a robust feature pooling against camera motion, which can alleviate the need of multi-scale inputs to the network. Sample visualizations highlight the regions in the images on which the model is specially focused. It can be seen that these regions are also the most semantically relevant. Our method outperforms all existing state-of-the-art methods in CDnet2014 datasets by an average overall F-measure of 0.9847. We also evaluate the effectiveness of our method on SBI2015 and UCSD Background Subtraction datasets. The source code of the proposed method is made available at.
Keyword:
Foreground segmentation
Convolutional neural networks
Feature pooling module
Background subtraction
Video surveillance
AI总结

AI总结

对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

Pattern Analysis and Applications 封面图
Pattern Analysis and Applications
IF:
2
论文数:
1.9K
被引数:
1.9K

机构

A
Ankara University
学者数:
8.5K
论文数: 7.2K
被引数: 5.9K
引用论文

引用论文

In0.53Ga0.47As p-i-n photodiodes with transparent cadmium tin oxide contactsIn0.53Ga0.47As p-i-n光电二极管,具有透明镉锡氧化物接触层
err1992-10-05
err0
PREAI
errPaul R. Berger; Niloy K. Dutta; George Zydzik; H. M. O’Bryan; Ursula Keller; Peter R. Smith; John Lopata; D. Sivco; A. Y. Cho
err分享
err收藏
Bidirectional stretching features on the flow and heat transport of Burgers nanofluid subject to modified heat and mass fluxes
err2022-03-30
err0
PREAI
errFuzhang Wang; Zahoor Iqbal; Juan Zhang; Mostafa A. H. Abdelmohimen; Abdulrazak H. Almaliki; Ahmed M. Galal
err分享
err收藏
err分享
err收藏
TNFR1 Signaling and IFN-γ Signaling Determine whether T Cells Induce Tumor Dormancy or Promote Multistage CarcinogenesisTNFR1信号和ifn-γ 信号决定T细胞是否诱导肿瘤休眠或促进多阶段癌变
err2008-06-01
err0
errOAAI
errNele Müller-Hermelink; Heidi Braumüller; Bernd Pichler; Thomas Wieder; Reinhard Mailhammer; Katrin Schaak; Kamran Ghoreschi; Amir Yazdi; Roland Haubner; Christian A. Sander; Ralph Mocikat; Markus Schwaiger; Irmgard Förster; Ralph Huss; Wolfgang A. Weber; Manfred Kneilling; Martin Röcken
err分享
err收藏
err分享
err收藏
End-to-end video background subtraction with 3d convolutional neural networks
err2017-12-11
err80
PREAI
errSakkos, Dimitrios; Liu, Heng; Han, Jungong; Shao, Ling
err分享
err收藏
学者 查看更多内容