Relation of Descriptor Completeness and Distinctiveness to Localization of the Most Informative Regions in Images

Allocation of informative regions in images depends on a visual task. We propose two criteria for allocation of informative regions in images considering visual recognition of objects in images using a space-variant foveal filter. The first criterion relates to descriptor completeness. This criterion is formalized as a measure of similarity of an original image to an image reconstructed from foveal filter responses. The second criterion relates to descriptor distinctiveness. This criterion relates to the problem of retrieving discriminative and repeatable features in images of objects. It is formalized as a variation of descriptor elements of different objects within the same class. We reveal that regions of interest produced by these criteria are distributed in the areas of high variety of brightness gradients orientations.

Differences of eye movement pattern in natural and man-made scenes and image categorization with the help of these patterns

In this paper, we investigated the parameters related to eye movement patterns of individuals while viewing images that consist of natural and man-made scenes. These parameters are as follows: number of fixations and saccades, fixation duration, saccade amplitude and distribution of fixation locations. We explored the way in which individuals look at images of different semantic categories, and used this information for automatic image classification. We showed that the eye movements and the contents of eye fixation locations of observers differ for images of different semantic categories. These differences were used effectively in automatic image categorization. Another goal of this study was to find the answer of this question that "whether the image patches of fixation points have sufficient information for image categorization?" To achieve this goal, a number of patches with different sizes from two different image categories was extracted. These patches, which were selected at the location of eye fixation points, were used to form a feature vector based on K-means clustering algorithm. Then, different statistical classifiers were trained for categorization purpose. The results showed that it is possible to predict the image category by using the feature vectors derived from the image patches. We found significant differences in parameters of eye movement pattern between the two image categories (average across subjects). We could categorize images by using these parameters as features. The results also showed that it is possible to predict the image category by using image patches around the subjects' fixation points.

Scanpaths of Complex Image Viewing: Insights From Experimental and Modeling Studies

From the first works of Buswell, Yarbus, and Noton and Stark, the scan path for viewing complex images has been considered as a possible key to objective estimation of cognitive processes and their dynamics. However, evidences both pro and con were revealed in the modern research. In this article, the results supporting the Yarbus-Stark concept are presented. In psychophysical tests, two types of images (three paintings from Yarbus` works and four textures) were used with two instructions, namely, “free viewing” and “search for modified image regions.” The focus of the analysis of experimental results and modeling has been given to local elements of the scan path. It was shown that each parameter used (square of viewing area, S; distance between center of mass of viewing area and image center, R; parameter Xi, based on duration of the current fixation and angle between preceding and following saccades), reflects the specificity of both visual task and image properties. Additionally, the return gaze fixations which have a set of specific properties and mainly address to the areas of interest on image were revealed. Evidently these facts can be formalized in an advanced mathematical model as additional instrument to study the mechanisms of complex image viewing.

Distractor effect at initial stages of recognition depends on visual image properties

In the present study, we describe a new method for estimation of distractor effect at image viewing. Dynamics of gaze fixation duration was compared at presentation of distractor during recognition of two types of target stimuli: fading-in contour images and black static cross. Circular blue distractor was applied to a target stimulus at some fixation points during initial stages of image recognition. All stimuli were presented in the center of computer screen inside the foveal part of the vision field. Subjects received the following instructions: to fixate their gaze on target image, recognize target stimulus as quick as possible and press the keyboard button to stop a trial. Our experiments have shown that (i) the median of the distractor effect (increase of fixation duration) was significantly less in trials with fading-in contour images, than in trials with cross image presentation (415 and 610 ms, p < 0.05); (ii) fixation duration without distractors had no significant difference between fading-in image trials (390 ± 10 ms) and cross image trials (438 ± 22 ms); (iii) mean recognition time for fading-in images was significantly more than for cross image (13.9 ± 4.3 s, and 2.47 ± 0.84 s). Possible prospective to use the presented method are discussed.