Limits...
Extraction of surface-related features in a recurrent model of V1-V2 interactions.

Weidenbacher U, Neumann H - PLoS ONE (2009)

Bottom Line: The approach is based on feedforward and feedback mechanisms found in visual cortical areas V1 and V2.Unlike previous proposals which treat localized junction configurations as 2D image features, we link them to mechanisms of apparent surface segregation.As a consequence, we demonstrate how junctions can change their perceptual representation depending on the scene context and the spatial configuration of boundary fragments.

View Article: PubMed Central - PubMed

Affiliation: Institute of Neural Information Processing, University of Ulm, Ulm, Germany. ulrich.weidenbacher@uni-ulm.de

ABSTRACT

Background: Humans can effortlessly segment surfaces and objects from two-dimensional (2D) images that are projections of the 3D world. The projection from 3D to 2D leads partially to occlusions of surfaces depending on their position in depth and on viewpoint. One way for the human visual system to infer monocular depth cues could be to extract and interpret occlusions. It has been suggested that the perception of contour junctions, in particular T-junctions, may be used as cue for occlusion of opaque surfaces. Furthermore, X-junctions could be used to signal occlusion of transparent surfaces.

Methodology/principal findings: In this contribution, we propose a neural model that suggests how surface-related cues for occlusion can be extracted from a 2D luminance image. The approach is based on feedforward and feedback mechanisms found in visual cortical areas V1 and V2. In a first step, contours are completed over time by generating groupings of like-oriented contrasts. Few iterations of feedforward and feedback processing lead to a stable representation of completed contours and at the same time to a suppression of image noise. In a second step, contour junctions are localized and read out from the distributed representation of boundary groupings. Moreover, surface-related junctions are made explicit such that they are evaluated to interact as to generate surface-segmentations in static images. In addition, we compare our extracted junction signals with a standard computer vision approach for junction detection to demonstrate that our approach outperforms simple feedforward computation-based approaches.

Conclusions/significance: A model is proposed that uses feedforward and feedback mechanisms to combine contextually relevant features in order to generate consistent boundary groupings of surfaces. Perceptually important junction configurations are robustly extracted from neural representations to signal cues for occlusion and transparency. Unlike previous proposals which treat localized junction configurations as 2D image features, we link them to mechanisms of apparent surface segregation. As a consequence, we demonstrate how junctions can change their perceptual representation depending on the scene context and the spatial configuration of boundary fragments.

Show MeSH

Related in: MedlinePlus

Model activities of V1 and V2 cell pool (right, first row) resulting from a real-world scene image that includes occluding opaque and transparent surfaces.A feature map was extracted from model activities, revealing position and type (colour coded) of junction configurations (right, second row).
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC2691604&req=5

pone-0005909-g013: Model activities of V1 and V2 cell pool (right, first row) resulting from a real-world scene image that includes occluding opaque and transparent surfaces.A feature map was extracted from model activities, revealing position and type (colour coded) of junction configurations (right, second row).

Mentions: In order to examine how the model performs for real-world camera images we used an image taken from a desk scene where several papers and a transparent foil are arranged such that they partly occlude each other (Figure 13). Model activities and the extracted feature map demonstrate that the model is also capable of dealing with real-world images. Note, that one of the papers has a very low contrast ratio with respect to the background. Nevertheless, the model performs excellent in finding the contour and the respective junctions. This also underlines that the model is invariant to contrast changes and thus also stable against changes of illumination conditions.


Extraction of surface-related features in a recurrent model of V1-V2 interactions.

Weidenbacher U, Neumann H - PLoS ONE (2009)

Model activities of V1 and V2 cell pool (right, first row) resulting from a real-world scene image that includes occluding opaque and transparent surfaces.A feature map was extracted from model activities, revealing position and type (colour coded) of junction configurations (right, second row).
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC2691604&req=5

pone-0005909-g013: Model activities of V1 and V2 cell pool (right, first row) resulting from a real-world scene image that includes occluding opaque and transparent surfaces.A feature map was extracted from model activities, revealing position and type (colour coded) of junction configurations (right, second row).
Mentions: In order to examine how the model performs for real-world camera images we used an image taken from a desk scene where several papers and a transparent foil are arranged such that they partly occlude each other (Figure 13). Model activities and the extracted feature map demonstrate that the model is also capable of dealing with real-world images. Note, that one of the papers has a very low contrast ratio with respect to the background. Nevertheless, the model performs excellent in finding the contour and the respective junctions. This also underlines that the model is invariant to contrast changes and thus also stable against changes of illumination conditions.

Bottom Line: The approach is based on feedforward and feedback mechanisms found in visual cortical areas V1 and V2.Unlike previous proposals which treat localized junction configurations as 2D image features, we link them to mechanisms of apparent surface segregation.As a consequence, we demonstrate how junctions can change their perceptual representation depending on the scene context and the spatial configuration of boundary fragments.

View Article: PubMed Central - PubMed

Affiliation: Institute of Neural Information Processing, University of Ulm, Ulm, Germany. ulrich.weidenbacher@uni-ulm.de

ABSTRACT

Background: Humans can effortlessly segment surfaces and objects from two-dimensional (2D) images that are projections of the 3D world. The projection from 3D to 2D leads partially to occlusions of surfaces depending on their position in depth and on viewpoint. One way for the human visual system to infer monocular depth cues could be to extract and interpret occlusions. It has been suggested that the perception of contour junctions, in particular T-junctions, may be used as cue for occlusion of opaque surfaces. Furthermore, X-junctions could be used to signal occlusion of transparent surfaces.

Methodology/principal findings: In this contribution, we propose a neural model that suggests how surface-related cues for occlusion can be extracted from a 2D luminance image. The approach is based on feedforward and feedback mechanisms found in visual cortical areas V1 and V2. In a first step, contours are completed over time by generating groupings of like-oriented contrasts. Few iterations of feedforward and feedback processing lead to a stable representation of completed contours and at the same time to a suppression of image noise. In a second step, contour junctions are localized and read out from the distributed representation of boundary groupings. Moreover, surface-related junctions are made explicit such that they are evaluated to interact as to generate surface-segmentations in static images. In addition, we compare our extracted junction signals with a standard computer vision approach for junction detection to demonstrate that our approach outperforms simple feedforward computation-based approaches.

Conclusions/significance: A model is proposed that uses feedforward and feedback mechanisms to combine contextually relevant features in order to generate consistent boundary groupings of surfaces. Perceptually important junction configurations are robustly extracted from neural representations to signal cues for occlusion and transparency. Unlike previous proposals which treat localized junction configurations as 2D image features, we link them to mechanisms of apparent surface segregation. As a consequence, we demonstrate how junctions can change their perceptual representation depending on the scene context and the spatial configuration of boundary fragments.

Show MeSH
Related in: MedlinePlus