Are we ready to tackle perceptual segmentation of natural scenes?
Saved in:
| Title: | Are we ready to tackle perceptual segmentation of natural scenes? |
|---|---|
| Authors: | Coen-Cagli, Ruben1,2,3 (AUTHOR) ruben.coen-cagli@einsteinmed.edu, Mamassian, Pascal4 (AUTHOR) |
| Source: | Vision Research. Mar2026, Vol. 240, pN.PAG-N.PAG. 1p. |
| Subjects: | Visual perception, Computer vision, Pattern perception, Machine learning, Landscapes, Image segmentation, Psychophysics |
| Abstract: | • Segmentation is fundamental to perceptual organization but poorly understood for natural stimuli. • Limitations of classical experimental methods and computer vision databases are discussed. • New measurements and algorithms for segmentation of natural images are highlighted. • Progress hinges on paradigms that integrate those computational and experimental innovations. • Recent studies that exemplify encouraging initial steps towards this goal are reviewed. Processes of perceptual segmentation and integration (PSI) are fundamental to perceptual organization. Although PSI of visual stimuli has been studied for over a century, we have only a rudimentary understanding of PSI of natural visual stimuli. This is due to limitations of traditional experimental methods in visual psychophysics of PSI; to the exclusive focus of computer-vision research for image segmentation on performance benchmarks; and to the scarcity of meaningful interactions between those two communities. The recent literature discussed in this paper presents a compelling argument that the field is starting to overcome those barriers. One important example of such an interaction between visual psychophysics and machine learning is given by the literature on the crowding phenomenon, which calls for revised models of summary statistics to explain some uncrowding results. Other examples reviewed here include studies of the perceptual uncertainty and dynamics of segmentation of natural stimuli, which call for computational models with probabilistic representations and dynamic computations. Conversely, contemporary machine learning algorithms produce impressive segmentation maps that still need to be aligned with human maps as measured with objective tasks such as the same/different segment paradigm reviewed here. Therefore, the time is ripe to move vision science forward by bridging new computational and experimental paradigms for PSI of natural stimuli. [ABSTRACT FROM AUTHOR] |
| Copyright of Vision Research 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: 191265326 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Are we ready to tackle perceptual segmentation of natural scenes? – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Coen-Cagli%2C+Ruben%22">Coen-Cagli, Ruben</searchLink><relatesTo>1,2,3</relatesTo> (AUTHOR)<i> ruben.coen-cagli@einsteinmed.edu</i><br /><searchLink fieldCode="AR" term="%22Mamassian%2C+Pascal%22">Mamassian, Pascal</searchLink><relatesTo>4</relatesTo> (AUTHOR) – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22Vision+Research%22">Vision Research</searchLink>. Mar2026, Vol. 240, pN.PAG-N.PAG. 1p. – Name: Subject Label: Subjects Group: Su Data: <searchLink fieldCode="DE" term="%22Visual+perception%22">Visual perception</searchLink><br /><searchLink fieldCode="DE" term="%22Computer+vision%22">Computer vision</searchLink><br /><searchLink fieldCode="DE" term="%22Pattern+perception%22">Pattern perception</searchLink><br /><searchLink fieldCode="DE" term="%22Machine+learning%22">Machine learning</searchLink><br /><searchLink fieldCode="DE" term="%22Landscapes%22">Landscapes</searchLink><br /><searchLink fieldCode="DE" term="%22Image+segmentation%22">Image segmentation</searchLink><br /><searchLink fieldCode="DE" term="%22Psychophysics%22">Psychophysics</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: • Segmentation is fundamental to perceptual organization but poorly understood for natural stimuli. • Limitations of classical experimental methods and computer vision databases are discussed. • New measurements and algorithms for segmentation of natural images are highlighted. • Progress hinges on paradigms that integrate those computational and experimental innovations. • Recent studies that exemplify encouraging initial steps towards this goal are reviewed. Processes of perceptual segmentation and integration (PSI) are fundamental to perceptual organization. Although PSI of visual stimuli has been studied for over a century, we have only a rudimentary understanding of PSI of natural visual stimuli. This is due to limitations of traditional experimental methods in visual psychophysics of PSI; to the exclusive focus of computer-vision research for image segmentation on performance benchmarks; and to the scarcity of meaningful interactions between those two communities. The recent literature discussed in this paper presents a compelling argument that the field is starting to overcome those barriers. One important example of such an interaction between visual psychophysics and machine learning is given by the literature on the crowding phenomenon, which calls for revised models of summary statistics to explain some uncrowding results. Other examples reviewed here include studies of the perceptual uncertainty and dynamics of segmentation of natural stimuli, which call for computational models with probabilistic representations and dynamic computations. Conversely, contemporary machine learning algorithms produce impressive segmentation maps that still need to be aligned with human maps as measured with objective tasks such as the same/different segment paradigm reviewed here. Therefore, the time is ripe to move vision science forward by bridging new computational and experimental paradigms for PSI of natural stimuli. [ABSTRACT FROM AUTHOR] – Name: AbstractSuppliedCopyright Label: Group: Ab Data: <i>Copyright of Vision Research 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=191265326 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1016/j.visres.2025.108749 Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 1 StartPage: N.PAG Subjects: – SubjectFull: Visual perception Type: general – SubjectFull: Computer vision Type: general – SubjectFull: Pattern perception Type: general – SubjectFull: Machine learning Type: general – SubjectFull: Landscapes Type: general – SubjectFull: Image segmentation Type: general – SubjectFull: Psychophysics Type: general Titles: – TitleFull: Are we ready to tackle perceptual segmentation of natural scenes? Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Coen-Cagli, Ruben – PersonEntity: Name: NameFull: Mamassian, Pascal IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 03 Text: Mar2026 Type: published Y: 2026 Identifiers: – Type: issn-print Value: 00426989 Numbering: – Type: volume Value: 240 Titles: – TitleFull: Vision Research Type: main |
| ResultId | 1 |
