Contour integration and segmentation with self-organized lateral connections

Choe, Yoonsuck; Miikkulainen, Risto

doi:10.1007/s00422-003-0435-5

Cited by 29 publications

(27 citation statements)

References 68 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Neural models that concentrate on one of these aspects of grouping are bound to have different features or to exhibit different behaviors. Models that start by using correlated firing of neural units as a hypothesis for the emergence of dynamic binding (Choe & Miikkulainen, 2004;Li, 1998;Yen & Finkel, 1998) might have more potential as an explanation of transient global grouping (but note the critical evaluation by Shadlen & Movshon, 1999, and the recent data on curve tracing by Roelfsema, Lamme, & Spekreijse, 2004). In their specifications of neural architecture, the authors of these models often rely on the now well-established finding that long-range connections between pyramidal neurons in area V1 tend to connect groups of cells with a similar orientation tuning (Gilbert, 1992;Gilbert et al, 1996;Stettler et al, 2002).…”

Section: Discussionmentioning

confidence: 99%

Perceptual grouping in Gabor lattices: Proximity and alignment

Claessens

Wagemans

2005

Perception & Psychophysics

View full text Add to dashboard Cite

We propose the Gabor lattice as a new stimulus designed to deal with multiple organizations in perceptual grouping, allowing both comparison between psychophysical data and neural findings and a systematic investigation of grouping based on several low-level characteristics and their interactions. A Gabor lattice is a geometric lattice with Gabor patches, evoking a multistable global orientation percept. Visual grouping in Gabor lattices with elements aligned in a global orientation was compared with grouping of nonaligned Gabor patches and of Gaussian blobs. The effect sizes of proximity and alignment were estimated in logistic regression analyses. The results confirmed the importance of proximity and local element alignment as factors in dynamic grouping. We also found a small but consistent enhancement of grouping along the global vector orthogonal to the local patch orientations. In light of these results, we further motivate the relevance of these stimuli and the associated experimental paradigm.

show abstract

Section: Discussionmentioning

confidence: 99%

Perceptual grouping in Gabor lattices: Proximity and alignment

Claessens

Wagemans

2005

Perception & Psychophysics

View full text Add to dashboard Cite

show abstract

“…Two approaches are generally taken when trying to model contour integration. The first is the biological route (Yen and Finkel 1998;Li 1998;Grigorescu et al 2003;Mundhenk and Itti 2003;Choe and Miikkulainen 2004;Ben-Shahar and Zucker 2004). In this method, the idea is to create a model of contour integration that explores how the brain may perform such activities.…”

Section: Computationmentioning

confidence: 99%

“…Thus, adding to the previous argument, there seems to be some ability for neurons in visual cortex to enhance a contour's perceptibility at locations represented by neurons that they are not directly connected to. Several theories have been advanced to explain how that can happen, for instance neural synchronization (Yen and Finkel 1998;Choe and Miikkulainen 2004), potential propagation (Li 1998) and fast plasticity (Braun 1999;Mundhenk and Itti 2003).…”

Section: Introductionmentioning

confidence: 99%

Computational modeling and exploration of contour integration for visual saliency

Mundhenk

Itti

2005

Biol Cybern

View full text Add to dashboard Cite

We propose a computational model of contour integration for visual saliency. The model uses biologically plausible devices to simulate how the representations of elements aligned collinearly along a contour in an image are enhanced. Our model adds such devices as a dopamine-like fast plasticity, local GABAergic inhibition and multi-scale processing of images. The fast plasticity addresses the problem of how neurons in visual cortex seem to be able to influence neurons they are not directly connected to, for instance, as observed in contour closure effect. Local GABAergic inhibition is used to control gain in the system without using global mechanisms which may be non-plausible given the limited reach of axonal arbors in visual cortex. The model is then used to explore not only its validity in real and artificial images, but to discover some of the mechanisms involved in processing of complex visual features such as junctions and end-stops as well as contours. We present evidence for the validity of our model in several phases, starting with local enhancement of only a few collinear elements. We then test our model on more complex contour integration images with a large number of Gabor elements. Sections of the model are also extracted and used to discover how the model might relate contour integration neurons to neurons that process end-stops and junctions. Finally, we present results from real world images. Results from the model suggest that it is a good current approximation of contour integration in human vision. As well, it suggests that contour integration mechanisms may be strongly related to mechanisms for detecting end-stops and junction points. Additionally, a contour integration mechanism may be involved in finding features for objects such as faces. This suggests that visual cortex may be more information efficient and that neural regions may have multiple roles.

show abstract

“…Furthermore, recurrent (artificial) neural network models are able to exhibit rich temporal dynamics [4], [5], [6]. Thus, time becomes an essential factor in neural network operation, whether it is natural or artificial (also see [7], [8], [9], [10]). …”

Section: Introductionmentioning

confidence: 99%

Evolution of recollection and prediction in neural networks

Chung

Kwon

Choe

2009

2009 International Joint Conference on Neural Networks

Self Cite

View full text Add to dashboard Cite

Abstract-A large number of neural network models are based on a feedforward topology (perceptrons, backpropagation networks, radial basis functions, support vector machines, etc.), thus lacking dynamics. In such networks, the order of input presentation is meaningless (i.e., it does not affect the behavior) since the behavior is largely reactive. That is, such neural networks can only operate in the present, having no access to the past or the future. However, biological neural networks are mostly constructed with a recurrent topology, and recurrent (artificial) neural network models are able to exhibit rich temporal dynamics, thus time becomes an essential factor in their operation. In this paper, we will investigate the emergence of recollection and prediction in evolving neural networks. First, we will show how reactive, feedforward networks can evolve a memory-like function (recollection) through utilizing external markers dropped and detected in the environment. Second, we will investigate how recurrent networks with more predictable internal state trajectory can emerge as an eventual winner in evolutionary struggle when competing networks with less predictable trajectory show the same level of behavioral performance. We expect our results to help us better understand the evolutionary origin of recollection and prediction in neuronal networks, and better appreciate the role of time in brain function.

show abstract

Contour integration and segmentation with self-organized lateral connections

Cited by 29 publications

References 68 publications

Perceptual grouping in Gabor lattices: Proximity and alignment

Perceptual grouping in Gabor lattices: Proximity and alignment

Computational modeling and exploration of contour integration for visual saliency

Evolution of recollection and prediction in neural networks

Contact Info

Product

Resources

About