Abstract: An example embodiment may involve obtaining an a×b pixel macro-cell from an input image. The a×b pixel macro-cell may contain 4 non-overlapping m×n pixel cells. The a×b pixels in the a×b pixel macro-cell may have respective color values and may be associated with respective object type tags. The example embodiment may also include selecting a compression technique to either (i) compress the a×b pixel macro-cell as a whole, or (ii) compress the a×b pixel macro-cell by compressing each of the 4 non-overlapping m×n pixel cells separately. The example embodiment may further include compressing the a×b pixel macro-cell according to the selected compression technique, and writing a representation of the compressed a×b pixel macro-cell to a computer-readable output medium.

Abstract: Methods, systems, and products recognize a gesture in a video signal. A camera generates the video signal as an output. An image comparison is then performed between frames in the video signal to images stored in memory of different gestures. When a frame matches an image, the gesture is determined. A corresponding controller operation may then be selected.

Abstract: Techniques and systems are described for generating an anatomically-constrained local model and for performing performance capture using the model. The local model includes a local shape subspace and an anatomical subspace. In one example, the local shape subspace constrains local deformation of various patches that represent the geometry of a subject's face. In the same example, the anatomical subspace includes an anatomical bone structure, and can be used to constrain movement and deformation of the patches globally on the subject's face. The anatomically-constrained local face model and performance capture technique can be used to track three-dimensional faces or other parts of a subject from motion data in a high-quality manner. Local model parameters that best describe the observed motion of the subject's physical deformations (e.g., facial expressions) under the given constraints are estimated through optimization.

Abstract: Methods, systems and computer program products for automatic color correction of image and video content are provided. A computer-implemented method may include generating an auto-level linear transform and a histogram equalization transform of a color distribution histogram for an original image among a set of sequential images. The auto-level linear transform and the histogram equalization transform are blended to create a modified image. Pixel color is adjusted in the modified image so that a color difference between a pixel and neighboring pixels in the modified image corresponds to a color difference between respective pixels in the original image.

Abstract: A video analytic system includes a depth stream sensor, spatial analysis module, temporal analysis module, and analytics module. The spatial analysis module iteratively identifies objects of interest based on local maximum or minimum depth stream values within each frame, removes identified objects of interest, and repeats until all objects of interest have been identified. The temporal analysis module associates each object of interest in the current frame with an object of interest identified in a previous frame, wherein the temporal analysis module utilizes the association between current frame objects of interest and previous frame objects of interest to generate temporal features related to each object of interest. The analytics module detects events based on the received temporal features.

Abstract: Techniques and systems are described for generating an anatomically-constrained local model and for performing performance capture using the model. The local model includes a local shape subspace and an anatomical subspace. In one example, the local shape subspace constrains local deformation of various patches that represent the geometry of a subject's face. In the same example, the anatomical subspace includes an anatomical bone structure, and can be used to constrain movement and deformation of the patches globally on the subject's face. The anatomically-constrained local face model and performance capture technique can be used to track three-dimensional faces or other parts of a subject from motion data in a high-quality manner. Local model parameters that best describe the observed motion of the subject's physical deformations (e.g., facial expressions) under the given constraints are estimated through optimization.

Abstract: Methods and systems may provide for receiving an input image having an input pixel density and defining a single filter based on a sub-pixel modification value, wherein the single filter has a working lattice density that is greater than the input pixel density. Additionally, the single filter and the input image may be used to generate an output image. In one example, defining the single filter includes converting the input pixel density to the working lattice density at an aperture level of the single filter.

Abstract: A method for connecting an electronic device using an eye-tracking technique and an electronic device that implements the method are provided. The method includes acquiring eye-tracking information, obtaining image information corresponding to the eye-tracking information, comparing the image information with specific information about at least one external device, and based on the comparison, determining a specific external device to be connected from among the at least one external device.

Abstract: It is provided according to the present application an image processing method applied to an electronic device equipped with a binocular camera. The image processing method includes: calculating a motion direction of a first image and a motion direction of a second image, wherein the first image and second image are captured by two cameras of the binocular camera respectively; performing deblurring on the first image and the second image; and combining the images which are already deblurred to obtain a definite three-dimensional perspective image. According to the image processing method, a blurry image generated during imaging is deblurred, the blurring caused by movement or the like may be weakened, the definition of the deblurred image is better than the definition of the blurred image, thereby leading to a better imaging result.

Abstract: A method that includes obtaining, using a processor, image data from a target coating. The method also includes performing, using the processor, an image analysis to determine at least one sparkle point from the image data, and performing, using the processor, a hue analysis to determine a sparkle color from the sparkle point. The method further includes calculating, using the processor, a sparkle color distribution, and generating, using the processor, a coating formulation that is the same or substantially similar in appearance to the target coating.

Abstract: The invention relates to a method for assisting with the search for an element (E) having a predefined hue in an environment, the method comprising the provision of at least one image in color, said or each image comprising pixels, the method being characterized in that the method comprises the selection in said or each image of the pixels having a hue for which value of the parameter associated through the bijection is comprised between first and second values, the first and second values being chosen so that the set of hues, for which value of the parameter associated through the bijection is comprised between both chosen values, comprises the predefined hue.

Abstract: A three dimensional map of a designated area that includes location data and relative height data of vantage points, points of interest, and objects, within the designated area, is received. A line-of-sight view from the vantage points to the points of interest is determined by calculating an angle, a distance, and a direction between a vantage point and a point of interest based, at least in part, on the location data and relative height data associated with the three dimensional map of the designated area. Obstructions of the line-of-sight view between the vantage points and the points of interest are determined, based on the line-of-sight view, the location data, and the relative height data, and responsive to determining that the line-of-sight view is free of obstruction, identifying each point of interest viewable from each vantage points, within the designated area.

Abstract: Results of automatic detection of dirt or other non-steady defects in a sequence of digitized image frames can be unreliable. Here, a determination of a detection of a defective object to be replaced by a replacement pattern in a frame of a sequence of image frames as a misdetection is presented that comprises determining a value of a first similarity measure between a boundary of the detected defective object and a boundary of the replacement pattern and determining a value of a second similarity measure between the detected defective object and the replacement pattern. The detection of the defective object is determined as a misdetection if at least one of the following holds true: the value of the first similarity measure is outside of a corresponding first similarity acceptance range and the value of the second similarity measure is inside of a corresponding second similarity acceptance range.

Abstract: Various embodiments may increase scalability of image representations stored in a database for use in image matching and retrieval. For example, a system providing image matching can obtain images of a number of inventory items, extract features from each image using a feature extraction algorithm, and transform the same into their feature descriptor representations. These feature descriptor representations can be subsequently stored and used to compare against query images submitted by users. Though the size of each feature descriptor representation isn't particularly large, the total number of these descriptors requires a substantial amount of storage space. Accordingly, feature descriptor representations are compressed to minimize storage and, in one example, machine learning can be used to compensate for information lost as a result of the compression.

Abstract: A facial skin mask may be generated based on isolating a head part in a captured image, removing a first pixel that is indicative of non-skin from the head part in the captured image, and removing a second pixel that is indicative of having a high velocity from the head part in the captured image. Heart rate may be detected based on the change of color of the pixels of the generated facial skin mask.

Abstract: Blood flow beneath a user's skin, for example, in a user's face may be visually rendered. In some aspects, a plurality of differences is determined in the intensity of pixels of a first image and the corresponding pixels of a subsequent second image. In some aspects, this plurality of differences is enhanced to accentuate a characteristic associated with the first image and the second image. The enhanced plurality of differences is visually rendered for each subsequent comparison of pixel intensity values.

Abstract: An image processing device according to one embodiment of the invention includes a contour direction estimating unit, a direction evaluating unit, a reference region weighting processing unit, and a composition operation unit.

Abstract: Image processing is performed in view of the difference between the chroma component and the brightness component in an image, with a relatively smaller amount of data processed during the image processing. A frequency decomposing unit 110 performs frequency decomposition directly on Bayer image signals from an imaging element 3. With this, a high frequency component representing color information and a low frequency component representing brightness information are obtained. A filter coefficient obtaining unit 131 of a correction processing unit 130 obtains filter coefficients and the filter coefficients for the high frequency component are different from those for the low frequency component. A filtering processing unit performs a filtering process on subimages based on the filter coefficients obtained by the filter coefficient obtaining unit 131 in such a manner that processing details of the filtering process for the high frequency component are different from those for the low frequency component.

Abstract: There is provided an image processing device including a depth acquisition unit and a smoothing processing unit. The depth acquisition unit acquires a depth to a subject in correlation with a pixel in a captured image of the subject, and the smoothing processing unit designates a pixel in a region excluding a predetermined region in the image as a target pixel, and performs a smoothing process of the degree according to the depth corresponding to the target pixel for a pixel value of the target pixel in a predetermined direction. This causes a pixel in a region excluding the predetermined region in the image to be blurred, thereby generating a panning image.

Abstract: An example embodiment may involve obtaining an a×b pixel macro-cell from an input image. The a×b pixel macro-cell may contain 4 non-overlapping m×n pixel cells. The a×b pixels in the a×b pixel macro-cell may have respective color values and may be associated with respective object type tags. The example embodiment may also include selecting a compression technique to either (i) compress the a×b pixel macro-cell as a whole, or (ii) compress the a×b pixel macro-cell by compressing each of the 4 non-overlapping m×n pixel cells separately. The example embodiment may further include compressing the a×b pixel macro-cell according to the selected compression technique, and writing a representation of the compressed a×b pixel macro-cell to a computer-readable output medium.