View in EDS

Autonomous Online Self‐Adaptive Stereo Network.

Saved in:
Bibliographic Details
Title: Autonomous Online Self‐Adaptive Stereo Network.
Authors: Jia, Zihan1 (AUTHOR), Yang, Xiao1 (AUTHOR) 2110293@stu.neu.edu.cn, Zhang, Zheng1 (AUTHOR), Hu, Yige1 (AUTHOR), Jovanovic Dolecek, Gordana1 (AUTHOR) gordana@ieee.org
Source: IET Signal Processing (Wiley-Blackwell). 12/22/2025, Vol. 2025, p1-15. 15p.
Subjects: Stereo vision (Computer science), Stereo image processing, Statistical reliability, Adaptive control systems, Generalization, Trend analysis, Digital technology
Abstract: Existing end‐to‐end stereo matching networks face significant deployment challenges due to their reliance on large training datasets and the limited ability of synthetic data to represent real‐world scenarios, leading to a severe domain shift. To address these challenges, this paper proposes a novel stereo matching network along with an efficient fine‐tuning strategy. First, a lightweight modular stereo matching network is proposed, which incorporates domain knowledge and optimizes specific modules to enhance generalization. Second, a confidence estimation network is developed to generate occlusion masks, which filter erroneous self‐supervised loss. Then, the Mann–Kendall (MK) trend detection method is used to evaluate the loss change trend of the most recent few frames in the online adaptive process to measure the model's adaptation degree to the scene, and control the online operation mode of the model. Finally, online and offline experiments on multiple datasets demonstrate the competitive performance of our method. [ABSTRACT FROM AUTHOR]
Copyright of IET Signal 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: 190416170
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Autonomous Online Self‐Adaptive Stereo Network.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Jia%2C+Zihan%22">Jia, Zihan</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Yang%2C+Xiao%22">Yang, Xiao</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> 2110293@stu.neu.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Zhang%2C+Zheng%22">Zhang, Zheng</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Hu%2C+Yige%22">Hu, Yige</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Jovanovic+Dolecek%2C+Gordana%22">Jovanovic Dolecek, Gordana</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> gordana@ieee.org</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22IET+Signal+Processing+%28Wiley-Blackwell%29%22">IET Signal Processing (Wiley-Blackwell)</searchLink>. 12/22/2025, Vol. 2025, p1-15. 15p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Stereo+vision+%28Computer+science%29%22">Stereo vision (Computer science)</searchLink><br /><searchLink fieldCode="DE" term="%22Stereo+image+processing%22">Stereo image processing</searchLink><br /><searchLink fieldCode="DE" term="%22Statistical+reliability%22">Statistical reliability</searchLink><br /><searchLink fieldCode="DE" term="%22Adaptive+control+systems%22">Adaptive control systems</searchLink><br /><searchLink fieldCode="DE" term="%22Generalization%22">Generalization</searchLink><br /><searchLink fieldCode="DE" term="%22Trend+analysis%22">Trend analysis</searchLink><br /><searchLink fieldCode="DE" term="%22Digital+technology%22">Digital technology</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Existing end‐to‐end stereo matching networks face significant deployment challenges due to their reliance on large training datasets and the limited ability of synthetic data to represent real‐world scenarios, leading to a severe domain shift. To address these challenges, this paper proposes a novel stereo matching network along with an efficient fine‐tuning strategy. First, a lightweight modular stereo matching network is proposed, which incorporates domain knowledge and optimizes specific modules to enhance generalization. Second, a confidence estimation network is developed to generate occlusion masks, which filter erroneous self‐supervised loss. Then, the Mann–Kendall (MK) trend detection method is used to evaluate the loss change trend of the most recent few frames in the online adaptive process to measure the model's adaptation degree to the scene, and control the online operation mode of the model. Finally, online and offline experiments on multiple datasets demonstrate the competitive performance of our method. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of IET Signal 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=190416170
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1049/sil2/8808531
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 15
        StartPage: 1
    Subjects:
      – SubjectFull: Stereo vision (Computer science)
        Type: general
      – SubjectFull: Stereo image processing
        Type: general
      – SubjectFull: Statistical reliability
        Type: general
      – SubjectFull: Adaptive control systems
        Type: general
      – SubjectFull: Generalization
        Type: general
      – SubjectFull: Trend analysis
        Type: general
      – SubjectFull: Digital technology
        Type: general
    Titles:
      – TitleFull: Autonomous Online Self‐Adaptive Stereo Network.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Jia, Zihan
      – PersonEntity:
          Name:
            NameFull: Yang, Xiao
      – PersonEntity:
          Name:
            NameFull: Zhang, Zheng
      – PersonEntity:
          Name:
            NameFull: Hu, Yige
      – PersonEntity:
          Name:
            NameFull: Jovanovic Dolecek, Gordana
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 22
              M: 12
              Text: 12/22/2025
              Type: published
              Y: 2025
          Identifiers:
            – Type: issn-print
              Value: 17519675
          Numbering:
            – Type: volume
              Value: 2025
          Titles:
            – TitleFull: IET Signal Processing (Wiley-Blackwell)
              Type: main
ResultId 1