scholarly journals Closure and global shape contributions to contour grouping

2014 ◽  
Vol 14 (10) ◽  
pp. 257-257
Author(s):  
I. Frund ◽  
J. H. Elder
Keyword(s):  
Global Shape ◽  
Contour Grouping

Global precedence and response activation: Evidence from LRPs

2002 ◽  
Vol 55 (1) ◽  
pp. 289-310 ◽  
Author(s):  
Jeff Miller ◽  
David Navon
Keyword(s):  
Global Precedence ◽  
Local Shape ◽  
Global Shape ◽  
Preliminary Information ◽  
Readiness Potentials ◽  
Global Advantage ◽  
Response Activation ◽  
Lateralized Readiness Potentials

Lateralized readiness potentials (LRPs) were measured in left/right/no-go tasks using compound global/local stimuli. In Experiment 1, participants responded to local target shapes and ignored global ones. RTs were affected by the congruence of the global shape with the local one, and LRPs indicated that irrelevant global shapes activated the responses with which they were associated. In Experiment 2, participants responded to conjunctions of target shapes at both levels, withholding the response if a target appeared at only one level. Global shapes activated responses in no-go trials, but local shapes did not. The results are consistent with partial-output models in which preliminary information about global shape can partially activate responses that are inconsistent with the local shape. They also demonstrate that part of the global advantage arises early, before response activation begins and probably before recognition of the local shape.

scholarly journals Allocation of attention to biological motion: Local motion dominates global shape

2011 ◽  
Vol 11 (3) ◽  
pp. 4-4 ◽  
Author(s):  
M. Hirai ◽  
D. R. Saunders ◽  
N. F. Troje
Keyword(s):  
Biological Motion ◽  
Local Motion ◽  
Global Shape

scholarly journals Global shape processing: Which parts form the whole?

2010 ◽  
Vol 10 (6) ◽  
pp. 16-16 ◽  
Author(s):  
J. Bell ◽  
S. Hancock ◽  
F. A. A. Kingdom ◽  
J. W. Peirce
Keyword(s):  
Global Shape ◽  
Shape Processing

Object Class Recognition in Mobile Urban Lidar Data Using Global Shape Descriptors

2013 ◽  
Author(s):  
Salar Awan ◽  
Mustafa Muhamad ◽  
Kresimir Kusevic ◽  
Paul Mrstik ◽  
Michael Greenspan
Keyword(s):  
Lidar Data ◽  
Shape Descriptors ◽  
Object Class ◽  
Global Shape ◽  
Object Class Recognition

Global shape and pH stability of ovorubin, an oligomeric protein from the eggs of Pomacea canaliculata

FEBS Journal ◽  
2008 ◽  
Vol 275 (18) ◽  
pp. 4522-4530 ◽  
Author(s):  
Marcos S. Dreon ◽  
Santiago Ituarte ◽  
Marcelo Ceolín ◽  
Horacio Heras
Keyword(s):  
Ph Stability ◽  
Pomacea Canaliculata ◽  
Global Shape ◽  
Oligomeric Protein

scholarly journals Capsule Networks as Recurrent Models ofGrouping and Segmentation

2019 ◽  
Author(s):  
Adrien Doerig ◽  
Lynn Schmittwilken ◽  
Bilge Sayim ◽  
Mauro Manassi ◽  
Michael H. Herzog
Keyword(s):  
Neural Networks ◽  
Visual System ◽  
Convolutional Neural Networks ◽  
Human Vision ◽  
Global Shape ◽  
Grouping And Segmentation ◽  
Segmentation Process ◽  
Shape Processing ◽  
Visual Crowding ◽  
Recurrent Connections

AbstractClassically, visual processing is described as a cascade of local feedforward computations. Feedforward Convolutional Neural Networks (ffCNNs) have shown how powerful such models can be. However, using visual crowding as a well-controlled challenge, we previously showed that no classic model of vision, including ffCNNs, can explain human global shape processing (1). Here, we show that Capsule Neural Networks (CapsNets; 2), combining ffCNNs with recurrent grouping and segmentation, solve this challenge. We also show that ffCNNs and standard recurrent CNNs do not, suggesting that the grouping and segmentation capabilities of CapsNets are crucial. Furthermore, we provide psychophysical evidence that grouping and segmentation are implemented recurrently in humans, and show that CapsNets reproduce these results well. We discuss why recurrence seems needed to implement grouping and segmentation efficiently. Together, we provide mutually reinforcing psychophysical and computational evidence that a recurrent grouping and segmentation process is essential to understand the visual system and create better models that harness global shape computations.Author SummaryFeedforward Convolutional Neural Networks (ffCNNs) have revolutionized computer vision and are deeply transforming neuroscience. However, ffCNNs only roughly mimic human vision. There is a rapidly expanding body of literature investigating differences between humans and ffCNNs. Several findings suggest that, unlike humans, ffCNNs rely mostly on local visual features. Furthermore, ffCNNs lack recurrent connections, which abound in the brain. Here, we use visual crowding, a well-known psychophysical phenomenon, to investigate recurrent computations in global shape processing. Previously, we showed that no model based on the classic feedforward framework of vision can explain global effects in crowding. Here, we show that Capsule Neural Networks (CapsNets), combining ffCNNs with recurrent grouping and segmentation, solve this challenge. ffCNNs and recurrent CNNs with lateral and top-down recurrent connections do not, suggesting that grouping and segmentation are crucial for human-like global computations. Based on these results, we hypothesize that one computational function of recurrence is to efficiently implement grouping and segmentation. We provide psychophysical evidence that, indeed, grouping and segmentation is based on time consuming recurrent processes in the human brain. CapsNets reproduce these results too. Together, we provide mutually reinforcing computational and psychophysical evidence that a recurrent grouping and segmentation process is essential to understand the visual system and create better models that harness global shape computations.

Sign in / Sign up

Export Citation Format

Share Document