View in EDS

Adaptively identify and refine ill-posed regions for accurate stereo matching.

Saved in:
Bibliographic Details
Title: Adaptively identify and refine ill-posed regions for accurate stereo matching.
Authors: Liu, Changlin1,2 (AUTHOR) 2021023511@m.scnu.edu.cn, Sun, Linjun1 (AUTHOR) sunlinjun@semi.ac.cn, Ning, Xin1 (AUTHOR), Xu, Jian1 (AUTHOR), Yu, Lina1 (AUTHOR), Zhang, Kaijie2 (AUTHOR), Li, Weijun1 (AUTHOR) wjli@semi.ac.cn
Source: Neural Networks. Oct2024, Vol. 178, pN.PAG-N.PAG. 1p.
Subjects: Generalization, Pixels, Cost, Design
Abstract: Stereo matching cost constrains the consistency between pixel pairs. However, the consistency constraint becomes unreliable in ill-posed regions such as occluded or ambiguous regions of the images, making it difficult to explore hidden correspondences. To address this challenge, we introduce an Error-area Feature Refinement Mechanism (EFR) that supplies context features for ill-posed regions. In EFR, we innovatively obtain the suspected error region according to aggregation perturbations, then a simple Transformer module is designed to synthesize global context and correspondence relation with the identified error mask. To better overcome existing texture overfitting, we put forward a Dual-constraint Cost Volume (DCV) that integrates supplementary constraints. This effectively improves the robustness and diversity of disparity clues, resulting in enhanced details and structural accuracy. Finally, we propose a highly accurate stereo matching network called Error-rectify Feature Guided Stereo Matching Network (ERCNet), which is based on DCV and EFR. We evaluate our model on several benchmark datasets, achieving state-of-the-art performance and demonstrating excellent generalization across datasets. The code is available at https://github.com/dean7liu/ERCNet_2023. [ABSTRACT FROM AUTHOR]
Copyright of Neural Networks is the property of Pergamon Press - An Imprint of Elsevier Science 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: 178942275
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Adaptively identify and refine ill-posed regions for accurate stereo matching.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Liu%2C+Changlin%22">Liu, Changlin</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<i> 2021023511@m.scnu.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Sun%2C+Linjun%22">Sun, Linjun</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> sunlinjun@semi.ac.cn</i><br /><searchLink fieldCode="AR" term="%22Ning%2C+Xin%22">Ning, Xin</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Xu%2C+Jian%22">Xu, Jian</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Yu%2C+Lina%22">Yu, Lina</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Zhang%2C+Kaijie%22">Zhang, Kaijie</searchLink><relatesTo>2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Li%2C+Weijun%22">Li, Weijun</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> wjli@semi.ac.cn</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Neural+Networks%22">Neural Networks</searchLink>. Oct2024, Vol. 178, pN.PAG-N.PAG. 1p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Generalization%22">Generalization</searchLink><br /><searchLink fieldCode="DE" term="%22Pixels%22">Pixels</searchLink><br /><searchLink fieldCode="DE" term="%22Cost%22">Cost</searchLink><br /><searchLink fieldCode="DE" term="%22Design%22">Design</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Stereo matching cost constrains the consistency between pixel pairs. However, the consistency constraint becomes unreliable in ill-posed regions such as occluded or ambiguous regions of the images, making it difficult to explore hidden correspondences. To address this challenge, we introduce an Error-area Feature Refinement Mechanism (EFR) that supplies context features for ill-posed regions. In EFR, we innovatively obtain the suspected error region according to aggregation perturbations, then a simple Transformer module is designed to synthesize global context and correspondence relation with the identified error mask. To better overcome existing texture overfitting, we put forward a Dual-constraint Cost Volume (DCV) that integrates supplementary constraints. This effectively improves the robustness and diversity of disparity clues, resulting in enhanced details and structural accuracy. Finally, we propose a highly accurate stereo matching network called Error-rectify Feature Guided Stereo Matching Network (ERCNet), which is based on DCV and EFR. We evaluate our model on several benchmark datasets, achieving state-of-the-art performance and demonstrating excellent generalization across datasets. The code is available at https://github.com/dean7liu/ERCNet_2023. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Neural Networks is the property of Pergamon Press - An Imprint of Elsevier Science 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=178942275
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1016/j.neunet.2024.106394
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 1
        StartPage: N.PAG
    Subjects:
      – SubjectFull: Generalization
        Type: general
      – SubjectFull: Pixels
        Type: general
      – SubjectFull: Cost
        Type: general
      – SubjectFull: Design
        Type: general
    Titles:
      – TitleFull: Adaptively identify and refine ill-posed regions for accurate stereo matching.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Liu, Changlin
      – PersonEntity:
          Name:
            NameFull: Sun, Linjun
      – PersonEntity:
          Name:
            NameFull: Ning, Xin
      – PersonEntity:
          Name:
            NameFull: Xu, Jian
      – PersonEntity:
          Name:
            NameFull: Yu, Lina
      – PersonEntity:
          Name:
            NameFull: Zhang, Kaijie
      – PersonEntity:
          Name:
            NameFull: Li, Weijun
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 10
              Text: Oct2024
              Type: published
              Y: 2024
          Identifiers:
            – Type: issn-print
              Value: 08936080
          Numbering:
            – Type: volume
              Value: 178
          Titles:
            – TitleFull: Neural Networks
              Type: main
ResultId 1