Target Patch Research Articles

Artificially created stimuli produced by a genetic algorithm using a saliency model as its fitness function show that Inattentional Blindness modulates performance in a pop-out visual search paradigm

Salient stimuli are more readily detected than less salient stimuli, and individual differences in such detection may be relevant to why some people fail to notice an unexpected stimulus that appears in their visual field whereas others do notice it. This failure to notice unexpected stimuli is termed 'Inattentional Blindness' and is more likely to occur when we are engaged in a resource-consuming task. A genetic algorithm is described in which artificial stimuli are created using a saliency model as its fitness function. These generated stimuli, which vary in their saliency level, are used in two studies that implement a pop-out visual search task to evaluate the power of the model to discriminate the performance of people who were and were not Inattentionally Blind (IB). In one study the number of orientational filters in the model was increased to check if discriminatory power and the saliency estimation for low-level images could be improved. Results show that the performance of the model does improve when additional filters are included, leading to the conclusion that low-level images may require a higher number of orientational filters for the model to better predict participants' performance. In both studies we found that given the same target patch image (i.e. same saliency value) IB individuals take longer to identify a target compared to non-IB individuals. This suggests that IB individuals require a higher level of saliency for low-level visual features in order to identify target patches.

Variable Scale Compressive Tracking Based on Structural Constraint Sample

A variable scale compressive tracking algorithm based on structural constraint sample is presented to solve the variable scale problem in this paper. A number of scanning windows with different scales and positions are obtained by structural constraint sampling.Some sparse random sensing matrices with different scales that can be computed offline easily are adopted to extract the features of different foreground target and background sample image patches with relevant scales online, the sample patch having a maximal score is regarded as the new tracking result by classifying the compressive features via a naive bayesian classifier,meanwhile,to update the location and scale. Experimental results show the proposed algorithm performs favorably against state-of-the-art algorithms on challenging sequences in terms of the basic attitude and scale change, which is robust and does not depend on the scale selection of the initial tracking area.

Least squares estimation-based adaptive observation model for aerial visual tracking applications

In aerially captured video, low resolution, very small object size and poor image quality may pose tracking as very difficult problem. Based on fundamental principle that single object class lie on a linear subspace, we propose a least squares estimation-based linear model which represents probe candidate as class specific galleries. In this method, tracking is considered as a binary classification problem, which consists of target and background class. Inclusion of background templates improves the discriminative strength of the object model. By projecting target patches on estimated target coefficient vector, all candidates are reconstructed. The squared sum of reconstruction error is used as an observation likelihood function to determine the maximum a posteriori MAP estimation for current frame. To contain the target appearance change during tracking online multiple dictionary learning strategy is proposed. Experiment results of proposed method shows better tracking accuracy and computation when compared to other representative tracking methods.

A color-gradient patch sparsity based image inpainting algorithm with structure coherence and neighborhood consistency

To better maintain structure coherence and neighborhood consistency, an exemplar-based inpainting algorithm is presented by introducing color-gradient patch sparsity (CGPS). Two concepts of CGPS including color-gradient structure sparsity (CGSS) and patch sparse representation are proposed to obtain the filling order, the search region size and the sparse representation of target patch, which are key steps in an exemplar-based inpainting algorithm. Firstly, the CGSS is designed based on weighted color-gradient distance (WCGD) to determine the filling order of all patches located at fill-front. Secondly, the WCGD is applied to search candidate patches and the CGSS is used to limit the search region size. Thirdly, the patch to be filled is sparsely represented under the local patch consistency constraints in color and gradient spaces. Differing from the exemplar-based inpainting approaches in which only color information is used, the proposed algorithm considers both color and gradient information, which ensures a better maintenance of structure coherence, texture clarity and neighborhood consistency. Moreover, the inpainting efficiency can be significantly improved by limiting the search region size via the CGSS. Experimental results on natural images are presented to demonstrate the advantages of the proposed approach for various tasks such as scratch removal, text removal, block removal and object removal.

The effect of feedback on performance and brain activation during perceptual learning

We investigated the role of informative feedback on the neural correlates of perceptual learning in a coherent-motion detection paradigm. Stimulus displays consisted of four patches of moving dots briefly (500 ms) presented simultaneously, one patch in each visual quadrant. The coherence level was varied in the target patch from near threshold to high, while the other three patches contained only noise. The participants judged whether coherent motion was present or absent in the target patch. To guarantee central fixation, a secondary RSVP digit-detection task was performed at fixation. Over six training sessions subjects learned to detect coherent motion in a predefined quadrant (i.e., the learned location). Half of our subjects were randomly assigned to the feedback group, where they received informative feedback after each response during training, whereas the other group received non-informative feedback during training that a response button was pressed. We investigated whether the presence of informative feedback during training had an influence on the learning success and on the resulting BOLD response. Behavioral data of 24 subjects showed improved performance with increasing practice. Informative feedback promoted learning for motion displays with high coherence levels, whereas it had little effect on learning for displays with near-threshold coherence levels. Learning enhanced fMRI responses in early visual cortex and motion-sensitive area MT+ and these changes were most pronounced for high coherence levels. Activation in the insular and cingulate cortex was mainly influenced by coherence level and trained location. We conclude that feedback modulates behavioral performance and, to a lesser extent, brain activation in areas responsible for monitoring perceptual learning.

Image Re-Attentionizing

In this paper, we propose a computational framework, called Image Re-Attentionizing, to endow the target region in an image with the ability of attracting human visual attention. In particular, the objective is to recolor the target patches by color transfer with naturalness and smoothness preserved yet visual attention augmented. We propose to approach this objective within the Markov Random Field (MRF) framework and an extended graph cuts method is developed to pursue the solution. The input image is first over-segmented into patches, and the patches within the target region as well as their neighbors are used to construct the consistency graphs. Within the MRF framework, the unitary potentials are defined to encourage each target patch to match the patches with similar shapes and textures from a large salient patch database, each of which corresponds to a high-saliency region in one image, while the spatial and color coherence is reinforced as pairwise potentials. We evaluate the proposed method on the direct human fixation data. The results demonstrate that the target region(s) successfully attract human attention and in the meantime both spatial and color coherence is well preserved.

Image inpainting based on sparse representations with a perceptual metric

This paper presents an image inpainting method based on sparse representations optimized with respect to a perceptual metric. In the proposed method, the structural similarity (SSIM) index is utilized as a criterion to optimize the representation performance of image data. Specifically, the proposed method enables the formulation of two important procedures in the sparse representation problem, 'estimation of sparse representation coefficients’ and 'update of the dictionary’, based on the SSIM index. Then, using the generated dictionary, approximation of target patches including missing areas via the SSIM-based sparse representation becomes feasible. Consequently, image inpainting for which procedures are totally derived from the SSIM index is realized. Experimental results show that the proposed method enables successful inpainting of missing areas.

A generative probability model of joint label fusion for multi-atlas based brain segmentation.

Automated labeling of anatomical structures in medical images is very important in many neuroscience studies. Recently, patch-based labeling has been widely investigated to alleviate the possible mis-alignment when registering atlases to the target image. However, the weights used for label fusion from the registered atlases are generally computed independently and thus lack the capability of preventing the ambiguous atlas patches from contributing to the label fusion. More critically, these weights are often calculated based only on the simple patch similarity, thus not necessarily providing optimal solution for label fusion. To address these limitations, we propose a generative probability model to describe the procedure of label fusion in a multi-atlas scenario, for the goal of labeling each point in the target image by the best representative atlas patches that also have the largest labeling unanimity in labeling the underlying point correctly. Specifically, sparsity constraint is imposed upon label fusion weights, in order to select a small number of atlas patches that best represent the underlying target patch, thus reducing the risks of including the misleading atlas patches. The labeling unanimity among atlas patches is achieved by exploring their dependencies, where we model these dependencies as the joint probability of each pair of atlas patches in correctly predicting the labels, by analyzing the correlation of their morphological error patterns and also the labeling consensus among atlases. The patch dependencies will be further recursively updated based on the latest labeling results to correct the possible labeling errors, which falls to the Expectation Maximization (EM) framework. To demonstrate the labeling performance, we have comprehensively evaluated our patch-based labeling method on the whole brain parcellation and hippocampus segmentation. Promising labeling results have been achieved with comparison to the conventional patch-based labeling method, indicating the potential application of the proposed method in the future clinical studies.

Optimal spatial management of an invasive plant using a model with above- and below-ground components

Invasive species can disrupt and damage their non-native habitats and are often the focus of control efforts. Motivated by considering the control of kudzu (Pueria montana) by grazers, we model plant growth with above- and below-ground components where grazing pressure is applied in discrete pulses. We compare two strategies, one that manages for eradication and one that minimizes above-ground biomass to maximize a site’s aesthetic and recreational value (beautification). We then consider the spatial allocation of a limited budget to protect a prioritized target patch from invasion from neighboring patches. Optimal management for the beautification goal and eradication goal are similar when budgets are limited and when the initial invasion size is small. Management diverges when budgets or initial invasion size are large. As such, smaller management efforts have less of an imperative to carefully define their management goal. Additionally, our model suggests the ability of the invading plant to respond after control or disturbance, and not overall growth rate, is the important factor when considering differences in optimal management strategies between different goals. Non-target patches were controlled more often when budgets were larger, initial invasion sizes on the target patch were small, and growth rates were low. This pattern suggests that controlling even small non-target patches is only economically favorable when continued control of the target patch passes a threshold of diminishing returns, counter to classic results where small satellite populations are controlled first to prevent new growth.

Regularized Feature Reconstruction for Spatio-Temporal Saliency Detection

Multimedia applications such as image or video retrieval, copy detection, and so forth can benefit from saliency detection, which is essentially a method to identify areas in images and videos that capture the attention of the human visual system. In this paper, we propose a new spatio-temporal saliency detection framework on the basis of regularized feature reconstruction. Specifically, for video saliency detection, both the temporal and spatial saliency detection are considered. For temporal saliency, we model the movement of the target patch as a reconstruction process using the patches in neighboring frames. A Laplacian smoothing term is introduced to model the coherent motion trajectories. With psychological findings that abrupt stimulus could cause a rapid and involuntary deployment of attention, our temporal model combines the reconstruction error, regularizer, and local trajectory contrast to measure the temporal saliency. For spatial saliency, a similar sparse reconstruction process is adopted to capture the regions with high center-surround contrast. Finally, the temporal saliency and spatial saliency are combined together to favor salient regions with high confidence for video saliency detection. We also apply the spatial saliency part of the spatio-temporal model to image saliency detection. Experimental results on a human fixation video dataset and an image saliency detection dataset show that our method achieves the best performance over several state-of-the-art approaches.

Affective engagement and subsequent visual processing: Effects of contrast and spatial frequency.

The present study examined if viewing affective stimuli alters subsequent visual processing, as indexed by steady-state visual potentials (ssVEPs) and behavioral performance in an orientation discrimination task. Participants viewed task-irrelevant but emotionally arousing pictures from the International Affective Picture System (1 s) followed by a target stimulus stream consisting of low (2 cpd) or high-spatial frequency (6 cpd) Gabor patches, flickering at a temporal rate of 14 Hz. Luminance contrast of the patches gradually increased for the first half and decreased for the second half of the total duration, resulting in a waxing-waning pattern of stimulus contrast. The authors found that the waveform envelope of 14 Hz-ssVEPs corresponded to time-varying stimulus contrast. Analyses compared medium- and high-contrast time segments, as a function of emotional content and spatial frequency. Results showed greater ssVEP amplitudes for patches with high compared to medium contrast. Viewing emotionally arousing pictures selectively enhanced the ssVEP amplitudes for low-spatial frequency target patches and attenuated the ssVEP evoked by high-spatial frequency patches, across contrast levels. Response times were slower for patches following unpleasant pictures rather than pleasant and neutral, and error rates mirrored the interaction of emotional content and spatial frequency observed in the ssVEP data. Together, the present results suggest that additive gain mechanisms and early visual pathways may mediate costs and benefits of emotional engagement for subsequent sensory processing.

How may agricultural matrix intensification affect understory birds in an Atlantic Forest landscape? A qualitative model on stochasticity and immigration

Understanding how different approaches of matrix management affect organisms that inhabit natural patches is crucial for biological conservation. Considering that great part of the tropical area is composed of agricultural land and that most of the land use is either intensified or on the verge to be, a relevant question is: how may agricultural intensification of the landscape matrix affect the population dynamics of understory birds? This paper describes a qualitative model based on the Qualitative Process Theory and implemented in Garp3 to provide answers to this question. We built a model using four species of endemic passerine birds in order to evaluate the case in an Atlantic Forest area, a biodiversity hotspot. The model describes a landscape composed of an extinction-resistant source patch and one target forest patch where stochastic events occur. If permeability of the matrix exceeds a given species-specific threshold, propagules coming from the source reach the target patch. Agriculture intensification affects the matrix spatial structure and reduces permeability to forest birds, thus reduces rescue effect. Additionally, we assume that some species may use the matrix as a supplementary habitat for feeding and that matrix management may affect the resource base for those species. Simulation results suggest that, if agriculture intensification continues to threat the Atlantic Forest biome, populations of sensitive species (Xiphorhynchus fuscus and Sclerurus scansor) that exist in the small forest patches (which is true for most forest remnants) will be highly susceptible to local extinctions and without further re-colonization.

Image inpainting with nonsubsampled contourlet transform

A new algorithm for image inpainting based on the searching of redundancy for corner points across different scales and orientations is proposed. The searching utilises the nonsubsampled contourlet transform (NSCT) of the original image. The target region is filled-in following the priority which is given to the corner points that accumulate a high energy at different orientations and scales, and also having a high level of confidence. In each iteration, the source patch chosen is copied onto the target patch producing the minimum dispersion across different scales and orientations of the NSCT within it. This approach – to firstly fill-in corner points of high energy and confidence which minimize the dispersion in the target patch across different scales and orientations – have been tested on two groups of images, one group with broken edges and objects and a second group with large objects to be erased. The proposed algorithm was compared with three other methods and results obtained indicate that the capability of this algorithm is sustainable.

Regional effects of clutter on human target detection performance

Clutter is something that is encountered in everyday life, from a messy desk to a crowded street. Such clutter may interfere with our ability to search for objects in such environments, like our car keys or the person we are trying to meet. A number of computational models of clutter have been proposed and shown to work well for artificial and other simplified scene search tasks. In this paper, we correlate the performance of different models of visual clutter to human performance in a visual search task using natural scenes. The models we evaluate are Feature Congestion (Rosenholtz, Li, & Nakano, 2007), Sub-band Entropy (Rosenholtz et al., 2007), Segmentation (Bravo & Farid, 2008), and Edge Density (Mack & Oliva, 2004) measures. The correlations were performed across a range of target-centered subregions to produce a correlation profile, indicating the scale at which clutter was affecting search performance. Overall clutter was rather weakly correlated with performance (r ≈ 0.2). However, different measures of clutter appear to reflect different aspects of the search task: correlations with Feature Congestion are greatest for the actual target patch, whereas the Sub-band Entropy is most highly correlated in a region 12° × 12° centered on the target.

Local Similarity Number and its Application to Object Tracking

In this paper, we present a tracking technique utilizing a simple saliency visual descriptor. Initially, we define a visual descriptor named local similarity pattern that mimics the famous texture operator local binary patterns. The key difference is that it assigns each pixel a code based on the similarity to the neighbouring pixels. Later, we simplify this descriptor to a local saliency operator which counts the number of similar pixels in a neighbourhood. We name this operator local similarity number ( LSN). We apply the local similarity number operator to measure the amount of saliency in a target patch and model the target. The proposed tracking algorithm uses a joint saliency-colour histogram to represent the target in a mean-shift tracking framework. We will show that the proposed saliency-colour target representation outperforms texture-colour where texture modelled by local binary patterns and colour target representation techniques are used.

Perceptual Organization in Colour Perception: Inverting the Gamut Expansion Effect

In the well-known gamut expansion effect, uniformly coloured target patches are perceived as more colourful when they are embedded in a uniform grey surround than when they are embedded in a variegated one. Here, we provide a demonstration showing that this effect can be inverted when the uniformly coloured target patches are replaced by variegated ones. This observation suggests that the gamut expansion effect is due to mechanisms of transparency perception rather than due to contrast adaptation.

Color perception involves color representations firstly at a semantic level and then at a lexical level

Studies and models have suggested that color perception first involves access to semantic representations of color. This result leads to two questions: (1) is knowledge able to influence the perception of color when associated with a color? and (2) can the perception of color really involve only semantic representations? We developed an experiment where participants have to discriminate the color of a patch (yellow vs. green). The target patch is preceded either by a black-and-white line drawing or by a word representing a natural object associated with the same or a different color (banana vs. frog). We expected a priming effect for pictures because, with a 350-ms SOA, they only involve access to semantic representations of color, whereas words seem only elicit an access to lexical representations. As expected, we found a priming effect for pictures, but also for words. Moreover, we found a general slowdown of response times in the word-prime-condition suggesting the need of an additional processing step to produce priming. In a second experiment, we manipulated the SOA in order to preclude a semantic access in the word-prime-condition that could explain the additional step of processing. We also found a priming effect, suggesting that interaction with perception occurs at a lexical level and the additional step occurs at a color perception level. In the discussion, we develop a new model of color perception assuming that color perception involves access to semantic representations and then access to lexical representations.

Missing Texture Reconstruction Method Based on Error Reduction Algorithm Using Fourier Transform Magnitude Estimation Scheme

A missing texture reconstruction method based on an error reduction (ER) algorithm, including a novel estimation scheme of Fourier transform magnitudes is presented in this brief. In our method, Fourier transform magnitude is estimated for a target patch including missing areas, and the missing intensities are estimated by retrieving its phase based on the ER algorithm. Specifically, by monitoring errors converged in the ER algorithm, known patches whose Fourier transform magnitudes are similar to that of the target patch are selected from the target image. In the second approach, the Fourier transform magnitude of the target patch is estimated from those of the selected known patches and their corresponding errors. Consequently, by using the ER algorithm, we can estimate both the Fourier transform magnitudes and phases to reconstruct the missing areas.

Exemplar-Based Image Inpainting Using Color Distribution Analysis

This paper proposes a novel exemplar-based inpianting algorithm based upon the color distribution analysis. A novel priority scheme is designed to assign priority and guide the filling-in order, which is a crucial step for the performance of the inpainted results. Color distribution is proposed to determine the confidence of the target patches situated on the boundary of the missing region by measuring the variation of variances of neighboring source patches. Our experiments on the natural images and the comparisons with the representative existing approaches show that our proposed scheme can more robustly assign high priority to the structures and enables the salient structures, to which human visual system is sensitive, to be consistent with the surrounding regions.

Muenke Syndrome Mutation, FgfR3P244R, Causes TMJ Defects

Muenke syndrome is characterized by various craniofacial deformities and is caused by an autosomal-dominant activating mutation in fibroblast growth factor receptor 3 (FGFR3P250R). Here, using mice carrying a corresponding mutation (FgfR3P244R), we determined whether the mutation affects temporomandibular joint (TMJ) development and growth. In situ hybridization showed that FgfR3 was expressed in condylar chondroprogenitors and maturing chondrocytes that also expressed the Indian hedgehog (Ihh) receptor and transcriptional target Patched 1(Ptch1). In FgfR3P244R mutants, the condyles displayed reduced levels of Ihh expression, H4C-positive proliferating chondroprogenitors, and collagen type II- and type X-expressing chondrocytes. Primary bone spongiosa formation was also disturbed and was accompanied by increased osteoclastic activity and reduced trabecular bone formation. Treatment of wild-type condylar explants with recombinant FGF2/FGF9 decreased Ptch1 and PTHrP expression in superficial/polymorphic layers and proliferation in chondroprogenitors. We also observed early degenerative changes of condylar articular cartilage, abnormal development of the articular eminence/glenoid fossa in the TMJ, and fusion of the articular disc. Analysis of our data indicates that the activating FgfR3P244R mutation disturbs TMJ developmental processes, likely by reducing hedgehog signaling and endochondral ossification. We suggest that a balance between FGF and hedgehog signaling pathways is critical for the integrity of TMJ development and for the maintenance of cellular organization.