View in EDS

STDC‐MA network for semantic segmentation.

Saved in:
Bibliographic Details
Title: STDC‐MA network for semantic segmentation.
Authors: Lei, Xiaochun1 (AUTHOR), Lu, Linjun2 (AUTHOR), Jiang, Zetao1 (AUTHOR) zetaojiang@guet.edu.cn, Gong, Zhaoting2 (AUTHOR), Lu, Chang2 (AUTHOR), Liang, Jiaming2 (AUTHOR), Xie, Junlin2 (AUTHOR)
Source: IET Image Processing (Wiley-Blackwell). Dec2022, Vol. 16 Issue 14, p3758-3767. 10p.
Subjects: Pixels, Problem solving
Abstract: Semantic segmentation is applied extensively in autonomous driving and intelligent transportation with methods that highly demand spatial and semantic information. Here, an STDC‐MA network is proposed to meet these demands. First, the STDC‐Seg structure is employed in STDC‐MA to ensure a lightweight and efficient structure. Subsequently, the feature alignment module is applied to understand the offset between high‐level and low‐level features, solving the problem of pixel offset related to upsampling on the high‐level feature map. The approach implements the effective fusion between high‐level features and low‐level features. A hierarchical multiscale attention mechanism is adopted to reveal the relationship among attention regions from two different input sizes of one image. Through this relationship, regions receiving much attention are integrated into the segmentation results, thereby reducing the unfocused regions of the input image and improving the effective utilisation of multiscale information. STDC‐MA maintains the segmentation speed as the STDC‐Seg network while improving the segmentation accuracy of small objects. STDC‐MA was verified on the validation dataset of Cityscapes. The segmentation result of STDC‐MA attained 78.32% mIOU with the input of 0.5× scale, 4.92% higher than STDC‐Seg. [ABSTRACT FROM AUTHOR]
Copyright of IET Image Processing (Wiley-Blackwell) is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
FullText Text:
  Availability: 0
Header DbId: egs
DbLabel: Engineering Source
An: 160001630
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: STDC‐MA network for semantic segmentation.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Lei%2C+Xiaochun%22">Lei, Xiaochun</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Lu%2C+Linjun%22">Lu, Linjun</searchLink><relatesTo>2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Jiang%2C+Zetao%22">Jiang, Zetao</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> zetaojiang@guet.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Gong%2C+Zhaoting%22">Gong, Zhaoting</searchLink><relatesTo>2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Lu%2C+Chang%22">Lu, Chang</searchLink><relatesTo>2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Liang%2C+Jiaming%22">Liang, Jiaming</searchLink><relatesTo>2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Xie%2C+Junlin%22">Xie, Junlin</searchLink><relatesTo>2</relatesTo> (AUTHOR)
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22IET+Image+Processing+%28Wiley-Blackwell%29%22">IET Image Processing (Wiley-Blackwell)</searchLink>. Dec2022, Vol. 16 Issue 14, p3758-3767. 10p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Pixels%22">Pixels</searchLink><br /><searchLink fieldCode="DE" term="%22Problem+solving%22">Problem solving</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Semantic segmentation is applied extensively in autonomous driving and intelligent transportation with methods that highly demand spatial and semantic information. Here, an STDC‐MA network is proposed to meet these demands. First, the STDC‐Seg structure is employed in STDC‐MA to ensure a lightweight and efficient structure. Subsequently, the feature alignment module is applied to understand the offset between high‐level and low‐level features, solving the problem of pixel offset related to upsampling on the high‐level feature map. The approach implements the effective fusion between high‐level features and low‐level features. A hierarchical multiscale attention mechanism is adopted to reveal the relationship among attention regions from two different input sizes of one image. Through this relationship, regions receiving much attention are integrated into the segmentation results, thereby reducing the unfocused regions of the input image and improving the effective utilisation of multiscale information. STDC‐MA maintains the segmentation speed as the STDC‐Seg network while improving the segmentation accuracy of small objects. STDC‐MA was verified on the validation dataset of Cityscapes. The segmentation result of STDC‐MA attained 78.32% mIOU with the input of 0.5× scale, 4.92% higher than STDC‐Seg. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of IET Image Processing (Wiley-Blackwell) is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=160001630
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1049/ipr2.12591
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 10
        StartPage: 3758
    Subjects:
      – SubjectFull: Pixels
        Type: general
      – SubjectFull: Problem solving
        Type: general
    Titles:
      – TitleFull: STDC‐MA network for semantic segmentation.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Lei, Xiaochun
      – PersonEntity:
          Name:
            NameFull: Lu, Linjun
      – PersonEntity:
          Name:
            NameFull: Jiang, Zetao
      – PersonEntity:
          Name:
            NameFull: Gong, Zhaoting
      – PersonEntity:
          Name:
            NameFull: Lu, Chang
      – PersonEntity:
          Name:
            NameFull: Liang, Jiaming
      – PersonEntity:
          Name:
            NameFull: Xie, Junlin
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 12
              Text: Dec2022
              Type: published
              Y: 2022
          Identifiers:
            – Type: issn-print
              Value: 17519659
          Numbering:
            – Type: volume
              Value: 16
            – Type: issue
              Value: 14
          Titles:
            – TitleFull: IET Image Processing (Wiley-Blackwell)
              Type: main
ResultId 1