Abstract: It is possible to stably reduce an amount of image data. An image processing system includes an image transmitting apparatus and an image receiving apparatus. The image transmitting apparatus includes an image reducing unit that reduces the size of an input image to generate a reduced image, a determining unit that determines, when the reduced image is enlarged to the size of the input image, an enlargement parameter where a difference from the input image becomes minimum and a transmitting unit that transmits the reduced image and the enlargement parameter. The image receiving apparatus includes a receiving unit that receives the reduced image and the enlargement parameter from the image transmitting apparatus and an image restoring unit that enlarges the reduced image using the enlargement parameter to generate a restored image.

Abstract: An image processing apparatus includes a physical property parameter acquisition unit and a texture control unit. The physical property parameter acquisition unit is configured to acquire a physical property parameter on a subject in an image. The texture control unit is configured to control a texture of the subject in the image by using the physical property parameter acquired by the physical property parameter acquisition unit.

Abstract: A digital method for removing optical aberrations from the image is disclosed. The method includes the initial profiling of an optical system and using the obtained information to correct the optical aberrations introduces to the image by the same or identical optical system.

Abstract: Generating an image with a selected level of background blur includes capturing, by a first image capture device, a plurality of frames of a scene, wherein each of the plurality of frames has a different focus depth, obtaining a depth map of the scene, determining a target object and a background in the scene based on the depth map, determining a goal blur for the background, and selecting, for each pixel in an output image, a corresponding pixel from the focus stack.

Abstract: An apparatus for image contrast enhancement and a method, electronic equipment that includes: a dividing unit configured to divide an input image into at least two layers; and detail degrees of the at least two layers are different; an enhancing unit configured to perform contrast enhancement on each of the at least two layers respectively; and degrees of enhancing contrasts of the layers are different; and a combining unit configured to combine the contrast-enhanced at least two layers. By dividing the input image into at least two layers with different detail degrees and performing contrast enhancement on the at least two layers at different degrees, the contrast of the image may be accurately and effectively enhanced, thereby a display effect of the image may be improved.

Abstract: In an example embodiment, a technique is provided for model review using augmented reality. An augmented reality device obtains tiles from a remote computing device for an overview resolution, and augments the model at the overview resolution and an overview view size into a physical environment at a data location. The augmented reality device displays the model augmented into the physical environment to a user disposed at a view location. In response to input requesting a change to the new resolution, the augmented reality device obtains additional tiles from the remote computing device for the new resolution, augments the model at the new resolution and a new view size into the physical environment, and displays the model augmented into the physical environment to the user disposed at the view location. In response to input that requests navigation of the model, the augmented reality device changes at least one of the data location or the view location.

Abstract: An embodiment in accordance with the present invention provides a method for 3D-2D registration (for example, registration of a 3D CT image to a 2D radiograph) that permits deformable motion between structures defined in the 3D image based on a series of locally rigid transformations. This invention utilizes predefined annotations in 3D images (e.g., the location of anatomical features of interest) to perform multiple locally rigid registrations that yield improved accuracy in aligning structures that have undergone deformation between the acquisition of the 3D and 2D images (e.g., a preoperative CT compared to an intraoperative radiograph). The 3D image is divided into subregions that are masked according to the annotations, and the registration is computed simultaneously for each divided region by incorporating a volumetric masking method within the 3D-2D registration process.

Abstract: In an image representing an observed ground surface area, the device calculates an optical-path-radiance with a high degree of accuracy.

Abstract: A method and an apparatus for extracting a facial feature and a method and an apparatus for recognizing a face are provided, in which the apparatus for extracting a facial feature may extract facial landmarks from a current input image, sample a skin region and a facial component region based on the extracted facial landmarks, generate a probabilistic model associated with the sampled skin region and the facial component region, extract the facial component region from a face region included in the input image using the generated probabilistic model, and extract facial feature information from the extracted facial component region.

Abstract: The techniques described herein segment objects represented in images using one or more depth maps, regardless of whether an initial depth map is of a threshold quality. To do so, the techniques determine whether the initial depth map is of the threshold quality. If so, the techniques use this depth map for performing segmentation on an image. If not, then the techniques utilize one or more fallback approaches for generating an additional depth map, from which the segmentation may be performed.

Abstract: A calibration apparatus for a scanning endoscope system having a light scanning device configured to cause light emitted from a light source to scan periodically within a predetermined scanning area, comprising: a relay lens to which light emitted from the light scanning device enters and which magnifies the predetermined scanning area; a light detection unit configured to receive light emerging from the relay lens and detect a scanning trajectory of the received light on a light receiving surface of the light detection unit; and a correction unit configured to correct a scanning parameter of the light emitted from the light scanning device so that the scanning trajectory detected by the light detection unit becomes a reference scanning trajectory.

Abstract: A method for combining array camera images with feature-based ghost removal includes (a) receiving, from an array camera, a first image and a second image respectively captured by a first camera and a second camera of the array camera, (b) rectifying and aligning the first image and the second image, (c) after said rectifying and aligning, identifying features in the first image to produce at least one first feature image each indicating features in the first image, and identifying features in the second image to produce at least one second feature image each indicating features in the second image, (d) comparing the at least one first feature image with the at least one second feature image to determine a ghost mask defining combination weights for combination of the first image with the second image.

Abstract: An image processing apparatus generates a first image by performing a resolution enhancement process for the input image, and a second image by performing a noise reduction process for a noise reduction target image. In the noise reduction process, the image processing apparatus extracts a first partial image containing a target pixel and a plurality of second partial images containing a reference pixel, calculates a correlation value between the first partial image and the plurality of second partial images, provides a weight to each of the plurality of second partial images based on a characteristic of the resolution enhancement process and the correlation value, calculates a pixel value of the target pixel using a pixel value of the reference pixel in the plurality of second partial images and the weight, and generate the second image using the calculated pixel value.

Abstract: A set of input images are acquired sequentially as image tensors. A low-tubal rank tensor and a sparse tensor are initialized using the image tensor, wherein the low-tubal rank tensor is a tensor product of a low-rank spanning tensor basis and corresponding tensor coefficients, and for each image, updating iteratively the image tensor, the tensor coefficients, and the sparse tensor using the image tensor and the low-rank spanning basis from a previous iteration. The spanning tensor basis is updated using the tensor coefficients, the sparse tensor, and the low rank tubal tensor, wherein the low rank tubal tensor represents a set of output images and the sparse tensor representing a set of sparse images.

Abstract: A visibility improvement method using gaze tracking includes detecting a gaze of a user using a camera; determining a focus object at which the user gazes from among at least one object viewed through a display unit of an electronic device; and displaying, on the display unit, an image with high visibility that corresponds to the focus object and has higher visibility than the focus object.

Abstract: A method comprising: receiving a radiograph of at least a portion of a patient's body, wherein the radiograph is a digital grayscale image, and wherein each pixel of the grayscale image corresponds to a localized exposure index (EI) value; generating an HSL (Hue-Saturation-Lightness) image from the radiograph, wherein: a hue channel and a saturation channel of the HSL image are generated based on the localized EI values, a luminance channel of the HSL image is generated based on the intensity values of the pixels; and transforming the HSL image to an RGB image which conveys both the portion of the radiograph and the localized EI values.

Abstract: There is provided an image processing apparatus that includes an ambient-light image obtaining section, a cumulative image generation section, and a high-quality image generation section. The ambient-light image obtaining section obtains an ambient-light image in a first time range, the ambient-light image being an image of an object captured with a predetermined exposure time. The cumulative image generation section generates a cumulative image in a second time range, the cumulative image being obtained by cumulative addition of each pixel value in a plurality of images, the plurality of images being of the object sequentially captured with the predetermined exposure time. The high-quality image generation section generates a high-quality image, the high-quality image being obtained by subtracting a pixel value in the ambient-light image from a corresponding pixel value in a normalized image, the normalized image being the cumulative image normalized based on a total sum of the exposure time.

Abstract: A hybrid CT dataset is obtained from a combination of an integrating detector and a photon-counting detector. The hybrid CT dataset contains sparse spectral energy data and dense energy integration data. The dense panchromatic data sets inherit the resolution properties of the integrating detector while the sparse spectral data sets inherit the spectral information of the photon-counting detector. Subsequently, the sparse spectral energy data sets are pansharpened based upon at least one dense panchromatic data set that lacks spectral information according to a pansharpening algorithm.

Abstract: A system that incorporates the subject disclosure may include, for example, partitioning the image into a group of blocks, calculating principle bilateral filtered image components for a first subset of the group of blocks where the principle bilateral filtered image components are not calculated for a second subset of the group of blocks, and applying an infinite impulse response filter to the image using the principle bilateral filtered image components. Other embodiments are disclosed.

Abstract: Provided is an image processing device, including: a distance calculator that calculates distance information corresponding to at least one image among a plurality of input images; and an image generator that generates an output image with a shallow depth of field based on the distance information, in which the distance calculator calculates distance information from a plurality of contrast calculation regions sizes of which are different, and the image generator calculates a pixel value of an output image by smoothing a pixel value of the input image based on the distance information.

Abstract: Described herein are improved systems and methods for diagnosis and therapy of vision stability dysfunctions in a patient. One embodiment provides a device (1) for obtaining data from a patient (3). Device (1) includes a frame (5) configured for mounting to the patient (3), a target module (8) for providing a drive signal to define a selectively movable target object, an eye tracking module (12) mounted to the frame (5), the module (12) being responsive to the drive signal for obtaining first data indicative of one or more characteristics of the patient's eyes; and a head tracking module (16) mounted to the frame (5), the head tracking module (16) being responsive to the drive signal for obtaining second data indicative of one or more characteristics of the motion of the patient's head.

Abstract: The present disclosure relates to techniques for reconstructing an object in three dimensions that is captured in a set of two-dimensional images. The object is reconstructed in three dimensions by computing depth values for edges of the object in the set of two-dimensional images. The set of two-dimensional images may be samples of a light field surrounding the object. The depth values may be computed by exploiting local gradient information in the set of two-dimensional images. After computing the depth values for the edges, depth values between the edges may be determined by identifying types of the edges (e.g., a texture edge, a silhouette edge, or other type of edge). Then, the depth values from the set of two-dimensional images may be aggregated in a three-dimensional space using a voting scheme, allowing the reconstruction of the object in three dimensions.

Abstract: Innovations in adaptive encoding and decoding for units of a video sequence can improve coding efficiency when switching between color spaces during encoding and decoding. For example, some of the innovations relate to adjustment of quantization or scaling when an encoder switches color spaces between units within a video sequence during encoding. Other innovations relate to adjustment of inverse quantization or scaling when a decoder switches color spaces between units within a video sequence during decoding.

Abstract: Examples of devices, systems, and methods to automatically generate haptics based on visual color features and motion analysis are disclosed. In one example, a video having a plurality of frames is received and masked frames for the video are generated by applying a color mask to the plurality of frames. An event between two of the masked frames is detected and an optical flow estimate is generated for these masked frames. At least one haptic effect corresponding to the event is generated based on the optical flow. The generated haptic effect(s) may be output to a haptic file or a haptic output device, or both.

Abstract: A system for capturing 3D image data of a scene, including three light sources, each configured to emit light at a different wavelength to the other two sources and to illuminate the scene to be captured; a first video camera configured to receive light from the light sources which has been reflected from the scene, to isolate light received from each of the light sources, and to output data relating to the image captured for each of the three light sources; a depth sensor configured to capture depth map data of the scene; and an analysis unit configured to receive data from the first video camera and process the data to obtain data relating to a normal field obtained from the images captured for each of the three light sources, and to combine the normal field data with the depth map data to capture 3D image data of the scene.

Abstract: An item recommendation device includes a context information generator and a ranker. The context information generator generates and outputs context information including user intention by performing semantic analysis on a natural language request which is input. The ranker ranks candidates of items to be presented to a user based on the context information, user information representing an attribute of the user, and history information representing item usage history of the user.

Abstract: A method for applying facial masks to faces in live video. The method includes receiving an image containing a face from a user, wherein the image is a frame of a video and identifying the coordinates of a face in the image. The method also includes identifying the coordinates of facial elements within the face previously identified and synchronizing a bitmap add-on, wherein synchronizing the bitmap add-on includes aligning the bitmap add-on with the identified facial elements. The method further includes applying the bitmap add-ons over the frame of the identified face.

Abstract: An initial high-resolution image (D2) is generated by projection of low-resolution images (DIN) onto a high-resolution image space, an intermediate-resolution image (D31) is generated by projection of the low-resolution images (DIN) onto an intermediate-resolution image space, the intermediate-resolution image (D31) is interpolated and enlarged to generate an intermediate-resolution enlarged image (D3), and the pixel values of the undefined pixels in the initial high-resolution image (D2) are estimated using the pixel values of the corresponding pixels in the intermediate-resolution enlarged image (D3). It is possible to generate a high-resolution image of a high picture quality with higher accuracy even when less low-resolution images are used.

Abstract: A computer-implemented method for denoising image data includes a computer system receiving an input image comprising noisy image data and denoising the input image using a deep multi-scale network comprising a plurality of multi-scale networks sequentially connected. Each respective multi-scale network performs a denoising process which includes dividing the input image into a plurality of image patches and denoising those image patches over multiple levels of decomposition using a threshold-based denoising process. The threshold-based denoising process denoises each respective image patch using a threshold which is scaled according to an estimation of noise present in the respective image patch. The noising process further comprises the assembly of a denoised image by averaging over the image patches.

Abstract: Methods, apparatuses and systems may provide for applying an image to a filter, wherein the filter includes data-adaptive weights. Additionally, an output scaling factor may be determined based on one or more statistical operators applied to the data-adaptive weights and the output scaling factor may be applied to an output of the filter.

Abstract: Aspects of the disclosed apparatuses, methods and systems provide elimination of distortion induced by an optical system that reflects light from an image source. An inverse mapping of the distortion is created for the optical system. The display system applies the inverse mapping to an image prior to display to introduce a distortion to the displayed image that is the inverse of the distortion introduced by the optical system. As a result, the distortion in the displayed image is canceled by the distortion of the optical element providing the user with an image that is substantially distortion free.

Abstract: Disclosed is a display apparatus. The display apparatus performs external compensation by using sensing data generated through actual sensing and calculation data which is calculated based on the sensing data with a median filter.

Abstract: A computer-implemented method for recovering a digital image (x) from a sequence of observed digital images (y1, . . . , yT), includes: obtaining an observed digital image (yt); estimating a point spread function (ft) based on the observed image (yt); estimating the recovered digital image (x), based on the estimated point spread function (ft) and the observed image (yt); and repeating the above steps. In order to correct optical aberrations of a lens, a point spread function of the lens may be used.

Abstract: Method, device, and program for solving the problem where a conventional image restoration method based on Bayesian probability theory uses nonlinear equations and requires a plurality of convolution operations, thereby not being appropriate for a single interaction multiple data-type (SIMD) super parallel computation processing device. This provides a method in which a computer causes a SIMD super parallel computation device to assign operations which include one convolution operation based on a Bayesian probability theory linear equation; a device based on the method; first and third programs for causing a computer to execute the program, and second and fourth programs for configuring the device to the computer and causing the device to run; and storage mediums for storing the programs. Optical degradation contained in TV video and images is reduced thereby restoring the TV video and images to a state close to the pre-degradation state of the TV video and images.

Abstract: A technology is described for identifying hydrometeors. A method includes receiving an image of a hydrometeor captured using a camera. The hydrometeor in the image can be identified and analyzed to determine characteristics associated with the hydrometeor. Environmental measurements recorded substantially contemporaneously with the image can be obtained from environmental sensors located in proximity to the camera. A feature vector can be constructed using the hydrometeor characteristics and the environmental measurements. The feature vector can be input to a classification model used to classify the hydrometeor, and the classification model can output a classification for the hydrometeor using the feature vector.

Abstract: A method for video encoding based on an image super-resolution, the method including: 1) performing super-resolution interpolation on a video image to be encoded using a pre-trained texture dictionary database to yield a reference image; in which the texture dictionary database includes: one or multiple dictionary bases, and each dictionary basis includes a mapping group formed by a relatively high resolution image block of a training image and a relatively low resolution image block corresponding to the relatively high resolution image block; 2) performing motion estimation and motion compensation of image blocks of the video image on the reference image to acquire prediction blocks corresponding to the image blocks of the video image; 3) performing subtraction between the image blocks of the video image and the corresponding prediction blocks to yield prediction residual blocks, respectively; and 4) encoding the prediction residual blocks.

Abstract: The present disclosure relates to a solid-state image-capturing element and electronic device capable of improving the linearity of illuminance values. The dynamic-range expander 118 expands dynamic range of a pixel value for each pixel based on the pixel value having different exposure times of a plurality of pixels. The integrator 119 integrates pixel values having the dynamic range expanded by the dynamic-range expander 118 and generates an illuminance value. The present disclosure is applicable to complementary metal-oxide semiconductor (CMOS) image sensor or the like used in, for example, an illuminometer.

Abstract: Provided is an image processing apparatus capable of applying a pseudo-illumination effect to an image while suppressing a drop in image quality, and a control method for such an image processing apparatus. The image processing apparatus generates a correction signal for correcting sharpness of an image on the basis of illumination characteristics of the virtual light, and applies the correction signal to the image.

Abstract: An object recognition apparatus and method thereof are disclosed. An exemplary apparatus may determine an image feature vector of a first image by applying a convolution network to the first image. The convolution network may extract features from image learning sets that include the first image and a sample segmentation map of the first image. The exemplary apparatus may determine a segmentation map of the first image by applying a deconvolution network to the determined image feature vector. The exemplary apparatus may determine a weight of the convolution network and a weight of the deconvolution network based on the sample segmentation map and the first segmentation map. The exemplary apparatus may determine a second segmentation map of a second image through the convolution network using the determined weight of the convolution network and through the deconvolution network using the determined weight of the deconvolution network.

Abstract: An inpainting device and method using the segmentation of a reference region are disclosed. An inpainting device using the segmentation of a reference region includes a region determination unit configured to determine an inpainting target region and a reference region within video image content, an inpainting group setting unit configured to set inpainting groups using pixel values within the reference region; an inpainting unit configured to perform segmentation inpainting adapted to perform inpainting on a segmented target region using the segmented reference region on an inpainting group basis and non-segmentation inpainting adapted to perform inpainting on the target region using the reference region, and an information provision unit configured to visually provide the result of the segmentation inpainting, the result of the non-segmentation inpainting, and information about the difference between the results of the segmentation inpainting and the non-segmentation inpainting to the user.

Abstract: A passive-tracking system for passively tracking a living being is described herein. The system can include a visible-light sensor to generate visible-light frames based on visible light reflected off an object and a sound transducer to generate sound frames based on sound reflected off the object. The system can also include a processor that can receive the visible-light and sound frames and based on data of these frames, generate a novelty representation of the object through a merging of the data of the frames. Based on the data of the visible-light and sound frames, the processor can determine the object is a living being, and when the processor does, it can generate, based on the novelty representation, a 3D position of the living-being object and can passively track the living-being object over time by, based on the novelty representation, selectively updating the 3D position of the living-being object.

Abstract: Color noise is reduced while suppressing the occurrence of color blurring in the vicinity of an edge. An image processing apparatus including: a first determination unit configured to determine whether or not a first area consisting of a plurality of pixels including a pixel of interest is an area in a uniform color for the pixel of interest in input image data; a second determination unit configured to determine whether or not a second area, different from the first area, consisting of a plurality of pixels including the pixel of interest is an area in a uniform color; a setting unit configured to set a reference area consisting of the pixel of interest and adjacent pixels of the pixel of interest for the pixel of interest; and a first correction unit configured to correct the pixel value of the pixel of interest to a value for which noise has been reduced based on the pixel value of the pixel included in the reference area.

Abstract: Disclosed are an image processing device, an imaging device, an image processing method, and an image processing program capable of, when recovering a deteriorated image due to a point spread function of an optical system, effectively performing phase recovery and suppressing the occurrence of artifact due to frequency recovery processing. The image processing device includes a phase recovery processing unit which subjects image data acquired from an imaging element by capturing an object image using an optical system to phase recovery processing using a phase recovery filter based on a point spread function of the optical system, a gradation correction processing unit which subjects image data subjected to the phase recovery processing to nonlinear gradation correction, and a frequency recovery processing unit which subjects image data subjected to the gradation correction to frequency recovery processing using a frequency recovery filter based on the point spread function of the optical system.

Abstract: Some embodiments include a method of operating a calibration server for a camera module. The method can include: receiving, by the computing server, a first training image taken by the camera module in a mobile device and a corresponding image-context attribute from the mobile device; aggregating, by the computing device, the first training image into a set of contextually similar images based on the image-context attribute; computing a calibration parameter model based on the set of contextually similar images utilizing dimension reduction statistical analysis; and scheduling to update the calibration parameter model to configure an image processor to adjust a raw photograph of the camera module according to the calibration parameter model.

Abstract: A method for replacing image data in a destination region that is divided into sub-pieces along one or more cutting paths, which start and end at two different points on the border, and finding replacement data for the sub-pieces. The cutting paths may be determined as a function of the type of image structured at the start and the end points. The cutting paths may also be determined as a function of the area of the sub-pieces and the lengths of the cutting paths. Optionally, the destination region may be determined by a spot detection algorithm. Further optionally, the spot detection algorithm may comprise calculation of a high pass filter, or detection of areas of luminosity and border-to-volume ratios. A method for moving an image element within an image is also provided.

Abstract: Disclosed are an image processing device, an imaging device, an image processing method, and an image processing program capable of, when recovering a deteriorated image due to a point spread function of an optical system, suppressing the occurrence of artifact and color gradation and achieving reduction in computational costs. The image processing device includes a frequency recovery processing unit which subjects image data acquired from an imaging element by capturing an object image using an optical system to frequency recovery processing using a frequency recovery filter based on a point spread function of the optical system, a gradation correction processing unit which subjects image data subjected to the frequency recovery processing to nonlinear gradation correction, and a phase recovery processing unit which subjects image data subjected to the gradation correction to phase recovery processing using a phase recovery filter based on the point spread function of the optical system.

Abstract: A method for recognizing a binary document image as a table, pure text, or flowchart includes calculating a side profile of the image for each of the four sides, calculating a boundary removal size N corresponding to each side based on widths of lines or strokes closest to that side, and for each side, removing a boundary of size N from the document image, and re-calculating the side profile for each side after the removal. Then, based on a comparison of the side profiles and the re-calculated side profiles, the input document image is recognized as a table if all side profiles change from smooth to non-smooth, as pure text if the side profile changes are small, and as a flowchart if the original side profiles contain multiple sharp changes and wide flat regions and if the side profile changes significantly in the previously wide flat regions.

Abstract: Image backup using a trained image classifier is disclosed. In various embodiments, an image classifier is applied to a plurality of images to identify one or more images to be included in a save set of images. The save set of images are selectively stored to a second location according to one or more policies.

Abstract: A map creation apparatus includes an image receiver configured to receive images in time series while moving; a first calculator configured to extract an image area indicating an object from the images and calculate a coordinate of the object in a world coordinate system; a second calculator configured to track the object in the extracted image area with the images and calculate an optical flow of the object; an eliminator configured to calculate a difference in coordinate between vanishing points generated by the movement of the image receiver and the object based on the optical flow, and eliminate the image area of the object from the images when determining the object as a moving object based on the calculated difference; and a storage controller configured to store map information including the coordinate of the object, not eliminated by the moving object eliminator, in the world coordinate system.

Abstract: A method implemented in a video editing device comprises retrieving media content and generating a user interface comprising a graphical representation of the retrieved media content on a first timeline component. The method further comprises analyzing the retrieved media content to extract attributes associated with the media content and generating a second timeline component in the user interface. At least a portion of the extracted attributes is arranged along the second timeline component with respect to time, and each of the portion of extracted attributes is represented by a corresponding graphical representation. Furthermore, each attribute corresponds to a segment in the media content. The method further comprises retrieving, based on the displayed attributes arranged along the graphical timeline component, a selection of at least one segment of the media content.