Abstract not found

... This article models how these interactions help visual cortex to realize: (1) the binding process whereby cortex groups distributed data into coherent object representations; (2) the attentional process whereby cortex selectively processes important events; and (3) the developmental and learning processes whereby cortex shapes its circuits to match environmental constraints. New computational ideas about feedback systems suggest how neocortex develops and learns in a stable way, and why top-down attention requires converging bottom-up inputs to fully activate cortical cells, whereas perceptual groupings do not.

... showed that a model of bottom-up visual attention can account in part for the spatial locations fixated by humans while free-viewing complex natural and artificial scenes. That study used a definition of salience based on local detectors with coarse global surround inhibition. Here, we use a similar framework to investigate the roles of several types of non-linear interactions known to exist in visual cortex, and of eccentricity-dependent processing. For each of these, we added a component to the salience model, including richer interactions among orientation-tuned units, both at spatial short range (for clutter reduction) and long range (for contour facilitation), and a detailed model of eccentricity-dependent changes in visual processing. Subjects free-viewed naturalistic and artificial images while their eye movements were recorded, and the resulting fixation locations were compared with the modelsÕ predicted salience maps. We found that the proposed interactions indeed play a significant role in the spatiotemporal deployment of attention in natural scenes; about half of the observed inter-subject variance can be explained by these different models. This suggests that attentional guidance does not depend solely on local visual features, but must also include the effects of interactions among features. As models of these interactions become more accurate in predicting behaviorally-relevant salient locations, they become useful to a range of applications in computer vision and human-machine interface design.

A shape can be more difficult to identify when other shapes are near it. For example, when several grating patches (see Fig. 1) are viewed parafoveally, observers are unable to report the orientation of the central patch. This phenomenon, known as ‘crowding, ’ has historically been confused with lateral masking, in which one stimulus attenuates signals generated by another stimulus. Here we show that despite their inability to report the orientation of an individual patch, observers can reliably estimate the average orientation, demonstrating that the local orientation signals are combined rather than lost. Our results imply that crowding is distinct from ordinary masking, and is perhaps related to texture perception. Under crowded conditions, the orientation signals in primary visual cortex are pooled before they reach consciousness.

We present a cortical-based model for computing the perceptual salience of contours embedded in noisy images. It has been suggested (Gilbert, 1992; Field, Hayes & Hess, 1993) that horizontal intra-cortical connections in primary visual cortex may modulate contrast detection thresholds and pre-attentive "popout ". In our model, horizontal connections mediate context-dependent facilitatory and inhibitory interactions among oriented cells. Strongly facilitated cells undergo temporal synchronization; and perceptual salience is determined by the level of synchronized activity. The model accounts for a range of reported psychophysical and physiological effects of contour salience (Polat & Sagi, 1993, 1994; Kapadia, Ito, Gilbert & Westheimer, 1995; Field et al., 1993; Kovács, Polat & Norcia, 1996; Pettet, McKee & Grzywacz, 1996). In particular, the model proposes that intrinsic properties of synchronization account for the increased salience of smooth, closed contours (Kovács & Julesz, 1993, ...

We describe a neurobiological model of visual attention and eye/head movements in primates, and its application to the automatic animation of a realistic virtual human head watching an unconstrained variety of visual inputs. The bottom-up (image-based) attention model is based on the known neurophysiology of visual processing along the occipito-parietal pathway of the primate brain, while the eye/head movement model is derived from recordings in freely behaving Rhesus monkeys. The system is successful at autonomously saccading towards and tracking salient targets in a variety of video clips, including synthetic stimuli, real outdoors scenes and gaming console outputs. The resulting virtual human eye/head animation yields realistic rendering of the simulation results, both suggesting applicability of this approach to avatar animation and reinforcing the plausibility of the neural model. 1.

Abstract not found

We present a recurrent neural network for feature binding and sensory segmentation, the competitive layer model (CLM). The CLM uses topographically structured competitive and cooperative interactions in a layered network to partition a set of input features into salient groups. The dynamics is formulated within a standard additive recurrent network with linear threshold neurons. Contextual relations among features are coded by pairwise compatibilities which define an energy function to be minimized by the neural dynamics. Due to the usage of dynamical winner-take-all circuits the model gains more flexible response properties than spin models of segmentation by exploiting amplitude information in the grouping process. We prove analytic results on the convergence and stable attractors of the CLM, which generalize earlier results on winner-take-all networks, and incorporate deterministic annealing for robustness against local minima. The piecewise linear dynamics of the CLM allows a linear eigensubspace analysis which we use to analyze the dynamics of binding in conjunction with annealing. For the example of contour detection we show how the CLM can integrate figure-ground segmentation and grouping into a unified model.

Two relevant dimensions are revealed within which developmental patterns of perceptual organization might be investigated. Within the local -- integrative dimension, employing a contour integration task, we found indications that spatial integration develops slowly. We also found reduced contextual modulation of a local target in children employing the Ebbinghaus illusion. Within the action -- perception dimension, we hypothesize a relatively slow development of the perceptual system (mediated by the ventral visual stream), as compared to the development of the action system (mediated by the dorsal visual stream). Taken together, the data indicate that long-range neuronal connectivity supporting perceptual organization in the posterior pole of the brain, and in the ventral visual pathway is not fully developed in young children.

We consider the role of contextual guidance in learning and processing within multi-stream neural networks. Earlier work (Kay & Phillips, 1994, 1996; Phillips et al., 1995) showed how the goals of feature discovery and associative learning could be fused within a single objective, and made precise using information theory, in such a way that local binary processors could extract a single feature that is coherent across streams. In this paper we consider multi-unit local processors with multivariate binary outputs that enable a greater number of coherent features to be extracted. Using the Ising model, we define a class of information-theoretic objective functions and also local approximations, and derive the learning rules in both cases. These rules have similarities to, and differences from, the celebrated BCM rule. Local and global versions of Infomax appear as by-products of the general approach, as well as multivariate versions of Coherent Infomax. Focussing on the more biologicall...