View in EDS HTML Full Text PDF Full Text

Learning Internal Representations of 3D Transformations From 2D Projected Inputs.

Saved in:
Bibliographic Details
Title: Learning Internal Representations of 3D Transformations From 2D Projected Inputs.
Authors: Connor, Marissa1 (AUTHOR) marissa.c.connor@gmail.com, Olshausen, Bruno2 (AUTHOR) baolshausen@berkeley.edu, Rozell, Christopher1 (AUTHOR) crozell@gatech.edu
Source: Neural Computation. Nov2024, Vol. 36 Issue 11, p2505-2539. 35p.
Subjects: Lie groups, Generators of groups, Transformation groups, Biological systems, Task performance
Abstract: We describe a computational model for inferring 3D structure from the motion of projected 2D points in an image, with the aim of understanding how biological vision systems learn and internally represent 3D transformations from the statistics of their input. The model uses manifold transport operators to describe the action of 3D points in a scene as they undergo transformation. We show that the model can learn the generator of the Lie group for these transformations from purely 2D input, providing a proof-of-concept demonstration for how biological systems could adapt their internal representations based on sensory input. Focusing on a rotational model, we evaluate the ability of the model to infer depth from moving 2D projected points and to learn rotational transformations from 2D training stimuli. Finally, we compare the model performance to psychophysical performance on structure-from-motion tasks. [ABSTRACT FROM AUTHOR]
Copyright of Neural Computation is the property of MIT Press 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
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
Text:
  Availability: 1
Header DbId: egs
DbLabel: Engineering Source
An: 180176564
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Learning Internal Representations of 3D Transformations From 2D Projected Inputs.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Connor%2C+Marissa%22">Connor, Marissa</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> marissa.c.connor@gmail.com</i><br /><searchLink fieldCode="AR" term="%22Olshausen%2C+Bruno%22">Olshausen, Bruno</searchLink><relatesTo>2</relatesTo> (AUTHOR)<i> baolshausen@berkeley.edu</i><br /><searchLink fieldCode="AR" term="%22Rozell%2C+Christopher%22">Rozell, Christopher</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> crozell@gatech.edu</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Neural+Computation%22">Neural Computation</searchLink>. Nov2024, Vol. 36 Issue 11, p2505-2539. 35p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Lie+groups%22">Lie groups</searchLink><br /><searchLink fieldCode="DE" term="%22Generators+of+groups%22">Generators of groups</searchLink><br /><searchLink fieldCode="DE" term="%22Transformation+groups%22">Transformation groups</searchLink><br /><searchLink fieldCode="DE" term="%22Biological+systems%22">Biological systems</searchLink><br /><searchLink fieldCode="DE" term="%22Task+performance%22">Task performance</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: We describe a computational model for inferring 3D structure from the motion of projected 2D points in an image, with the aim of understanding how biological vision systems learn and internally represent 3D transformations from the statistics of their input. The model uses manifold transport operators to describe the action of 3D points in a scene as they undergo transformation. We show that the model can learn the generator of the Lie group for these transformations from purely 2D input, providing a proof-of-concept demonstration for how biological systems could adapt their internal representations based on sensory input. Focusing on a rotational model, we evaluate the ability of the model to infer depth from moving 2D projected points and to learn rotational transformations from 2D training stimuli. Finally, we compare the model performance to psychophysical performance on structure-from-motion tasks. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Neural Computation is the property of MIT Press 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=180176564
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1162/neco_a_01695
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 35
        StartPage: 2505
    Subjects:
      – SubjectFull: Lie groups
        Type: general
      – SubjectFull: Generators of groups
        Type: general
      – SubjectFull: Transformation groups
        Type: general
      – SubjectFull: Biological systems
        Type: general
      – SubjectFull: Task performance
        Type: general
    Titles:
      – TitleFull: Learning Internal Representations of 3D Transformations From 2D Projected Inputs.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Connor, Marissa
      – PersonEntity:
          Name:
            NameFull: Olshausen, Bruno
      – PersonEntity:
          Name:
            NameFull: Rozell, Christopher
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 11
              Text: Nov2024
              Type: published
              Y: 2024
          Identifiers:
            – Type: issn-print
              Value: 08997667
          Numbering:
            – Type: volume
              Value: 36
            – Type: issue
              Value: 11
          Titles:
            – TitleFull: Neural Computation
              Type: main
ResultId 1