Abstract: Apparatus, methods, and other embodiments associated with image morphology operations that provide fractional pixel dilation and erosion are described. According to one embodiment, an apparatus includes scaling logic configured to scale up an input data structure of image pixel data from a first number of pixels to a second number of pixels to generate a scaled-up data structure of image pixel data. The apparatus also includes morphology logic configured to perform a morphological operation on the scaled-up data structure of image pixel data to generate a morphed data structure of image pixel data. The apparatus further includes sub-sampling logic configured to sub-sample the morphed data structure of image pixel data to generate an output data structure of image pixel data. The output data structure of image pixel data has a same number of pixels as the input data structure of image pixel data.

Abstract: Disclosed herein are techniques and systems for computing geodesic saliency of images using background priors. An input image may be segmented into a plurality of patches, and a graph associated with the image may be generated, the graph comprising nodes and edges. The nodes of the graph include nodes that correspond to the plurality of patches of the image plus an additional virtual background node that is added to the graph. The graph further includes edges that connect the nodes to each other, including internal edges between adjacent patches and boundary edges between those patches at the boundary of the image and the virtual background node. Using this graph, a saliency value, called the “geodesic” saliency, for each patch of the image is determined as a length of a shortest path from a respective patch to the virtual background node.

Abstract: A method and device for processing a picture comprises estimating a color mapping between a first and a second color-graded version of the picture by estimating a color mapping function that maps the color values of the first color-graded version of the picture onto the color values of the second color-graded version of the picture.

Abstract: A computer determines one or more brightness levels associated with a first set of pixels within a light source in an image. The computer determines a brightness effect on a second set of pixels in the image based on the one or more brightness levels associated with the first set of pixels within the light source. The computer receives input to alter at least one of the one or more brightness levels associated with the first set of pixels within the light source. The computer alters one or more brightness levels associated with the second set of pixels in the image based on the received input and the determined brightness effect.

Abstract: Methods and devices for video coding are described. When encoding a non-zero coefficient group (CG) prior to the last such group in a group scan ordering within a transform block, data representative of the true (x,y)-position of the last non-zero transform coefficient in a scan ordering within that CG is modified, to improve coding efficiency, based upon at least one of (a) an intra-prediction mode used to encode the block and (b) at least one coefficient group neighboring that CG. Upon decoding a CG thus encoded, the modification is un-done to retrieve the true (x,y)-position.

Abstract: Various techniques are disclosed for smart surveillance camera systems and methods using thermal imaging to intelligently control illumination and monitoring of a surveillance scene. For example, a smart camera system may include a thermal imager, an IR illuminator, a visible light illuminator, a visible/near IR (NIR) light camera, and a processor. The camera system may capture thermal images of the scene using the thermal imager, and analyze the thermal images to detect a presence and an attribute of an object in the scene. In response to the detection, various light sources may be selectively operated to illuminate the object only when needed or desired, with a suitable type of light source, with a suitable beam angle and width, or in otherwise desirable manner. The visible/NIR light camera may also be selectively operated based on the detection to capture or record surveillance images containing objects of interest.

Abstract: An area partitioning method for partitioning an image into a foreground and a background includes a foreground designation step of causing a user to designate a part of pixels in an area that should be the foreground in the image as a foreground designating pixel, an estimation step of estimating a foreground color distribution and a background color distribution based on a color of the foreground designating pixel designated by the user, and an area partition step of partitioning the image into a foreground area and a background area with the estimated foreground color distribution and the estimated background color distribution as a condition. The estimation step includes dividing the color distribution of the image into a plurality of clusters, and selecting at least one cluster having a large relationship with the foreground designating pixel designated by the user in the plurality of clusters as the foreground color distribution.

Abstract: A technique for visualizing elements in images by applying a color coding procedure to data which comprises an initial image and segmentation results based on N labels. The segmentation results comprise information on segmentation uncertainty. The color coding procedure constructs a resulting colored image based on the initial image and N pre-selected base colors, and in such a manner that colors in the resulting colored image are modified by using intensity of the initial image and the information on segmentation uncertainty.

Abstract: Systems, apparatuses, and/or methods may provide for segmenting an image by generating a histogram of its pixel values, dividing the histogram into class intervals, and then iteratively computing new, shifted weighted means and shifted class interval boundaries for the class intervals until a predetermined level of convergence to a limit is obtained. The pixels may then be updated to a last weighted mean for class intervals to which they belong, providing segmentation. Similarly, any data may be segmented to provide computational efficiency.

Abstract: A method for estimating occlusion in a virtual environment comprising at least two objects and observed from a viewpoint. Thereafter, a second item of information representative of the distance separating a second fragment of a second object from the viewpoint according to an observation direction is computed; the second item of information is compared with a first item of information representative of the distance separating a first fragment of a first object from the viewpoint according to the observation direction and is stored in a first buffer memory. If the second item of information is less than the first item of information, then a third item of information representative of attributes associated with the second fragment is estimated, the second item of information is stored in a second buffer memory and of the third item of information is stored in a third buffer memory.

Abstract: An image processing apparatus (104) includes an image restoration filter storing unit configured to store an image restoration filter corresponding to image capturing condition information, an optical transfer function storing unit configured to store optical transfer function information, and an image restoration filter generating unit configured to determine whether a first image restoration filter corresponding to first image capturing condition information of a captured image is stored in the image restoration filter storing unit, select the first image restoration filter from the image restoration filter storing unit when the first image restoration filter is stored in the image restoration filter storing unit, and generate the first image restoration filter using the optical transfer function information when the first image restoration filter is not stored in the image restoration filter storing unit.

Abstract: A method executed by a server to provide a mobile device with relevant information about an object in a captured image received from an image reproducing device, the method including: receiving, from the image reproducing device, an image captured by the image reproducing device; receiving, from the mobile device, information corresponding to an application usage history of the mobile device; searching, based on the application usage history, for the relevant information about the object in the captured image; and transmitting to the mobile device the relevant information found based on the searching.

Abstract: A method for tracking objects across a number of image frames includes tracking objects in the frames based on appearance models of each foreground region corresponding to each of the objects and determining if a plurality of the tracked objects overlap. Where a plurality of the tracked objects overlap, the method creates compensated appearance models for each of the plurality of overlapping objects by attenuating common appearance features among the corresponding appearance models; and tracks the plurality of overlapping objects based on the created compensated appearance models.

Abstract: Image cropping suggestion using multiple saliency maps is described. In one or more implementations, component scores, indicative of visual characteristics established for visually-pleasing croppings, are computed for candidate image croppings using multiple different saliency maps. The visual characteristics on which a candidate image cropping is scored may be indicative of its composition quality, an extent to which it preserves content appearing in the scene, and a simplicity of its boundary. Based on the component scores, the croppings may be ranked with regard to each of the visual characteristics. The rankings may be used to cluster the candidate croppings into groups of similar croppings, such that croppings in a group are different by less than a threshold amount and croppings in different groups are different by at least the threshold amount. Based on the clustering, croppings may then be chosen, e.g., to present them to a user for selection.

Abstract: According to an embodiment, a density measuring device includes a first calculator, a second calculator, and a first generator. The first calculator calculates, from an image including objects of a plurality of classes classified according to a predetermined rule, for each of a plurality of regions formed by dividing the image, density of the objects captured in the region. The second calculator calculates, from the density of the objects captured in each of the regions, likelihood of each object class captured in each of the regions. The first generator generates density data, in which position corresponding to each of the regions in the image is assigned with the density of the object class having at least the higher likelihood than the lowest likelihood from among likelihoods calculated for object classes captured in the corresponding region.

Abstract: An image processing apparatus includes a gain calculation unit that calculates a plurality of first gains that respectively correspond to a plurality of predetermined edges based on a target pixel value and values of pixels adjacent to the target pixel, a blend ratio calculation unit that calculates a plurality of characteristic amounts that represent an accuracy of detecting the plurality of edges based on the target pixel value and the adjacent pixel values and calculates a plurality of weights that respectively correspond to the plurality of first gains based on the plurality of characteristic amounts, a gain adjustment unit that calculates a second gain based on the plurality of first gains and the plurality of weights and an output image generation unit that applies the second gain to the target pixel value to generate output image data.

Abstract: Provided are a method and an apparatus for image processing. A scene includes a low resolution depth image and a high resolution color image. A depth image up-scaler up-scales the depth image using an interpolation so as to improve the resolution of the depth image. A color-depth similarity based filter removes noise from the up-scaled depth image using the color image and the up-scaled depth image. A boundary sharpening filter converts a boundary between objects in the up-scaled depth image into a clearer boundary. The resolution of the depth image may increase to the level of the resolution of the color image by means of the depth image up-scaler, the color-depth similarity based filter and the boundary sharpening filter, and the depth of each pixel in the depth image may be adjusted.

Abstract: Relightable free-viewpoint rendering allows a novel view of a scene to be rendered and relit based on multiple views of the scene from multiple camera viewpoints. Cast shadows in the scene are handled to allow local occlusion to be taken into account when determining irradiance estimates for the sample positions in the scene. This is achieved by determining an initial estimate of irradiance by ignoring local occlusion, then using that initial irradiance estimate with knowledge of the scene geometry to determine an angle-dependent radiance estimate for the scene. This radiance estimate can then be used with the knowledge of the scene geometry to determine local irradiance estimates at the sample positions within the scene.

Abstract: A computer determines one or more brightness levels associated with a first set of pixels within a light source in an image. The computer determines a brightness effect on a second set of pixels in the image based on the one or more brightness levels associated with the first set of pixels within the light source. The computer receives input to alter at least one of the one or more brightness levels associated with the first set of pixels within the light source. The computer alters one or more brightness levels associated with the second set of pixels in the image based on the received input and the determined brightness effect.

Abstract: Systems and methods are disclosed for performing local adaptive weighting upsampling. An image may be converted into a color space representation. The color space representation of the image may then be divided into the individual color spaces. Upsampling may then be performed on the individual color planes. The upsampling may include resizing the color planes by adding pixels to the color planes. Local processing of the different color spaces may be performed to estimate values for the added pixels. Upon generating the estimated values, the estimated values may be refined by determining which known pixel values from the original image have the highest degree of influence on a newly added pixel.