Abstract: A method and an apparatus for filtering an array of pixels. A Domain Transforming unit applies a Domain Transform to the array of pixels to obtain a Domain Transform signal. A weight determining unit determines weights associated with samples of the Domain Transform signal. A filter then filters the Domain Transform signal. The filter takes the determined weights into account during filtering.

Abstract: A method and an apparatus for removing outliers from a main view of a scene during 3D reconstruction of a scene from multiple views of the scene. A 3D projection unit projects a 3D point of a pixel of the main view into neighboring views. A comparator then compares the distance of each of the projected 3D points in the neighboring views to the 3D point of the main view with a defined distance threshold. Based on the comparison a flagging unit assigns flags to the pixel in the main view. Finally, depending on values of the flags a rejecting unit rejects the pixel in the main view as an outlier.

Abstract: A method and an apparatus for filtering an image using a guided image filter are described. A filter computes a filtering output using a guidance image. For calculating the filtering output confidence values associated to a filter input are taken into account. A confidence-weighted averaging is performed using the confidence values.

Abstract: A method and an apparatus for generating a confidence map for a disparity map associated to a set of two or more images are described. Motion between at least two subsequent sets of two or more images is determined. Based on the determined motion information static and dynamic regions in the images of the sets of two or more images are detected and separated. A disparity change between a disparity value determined for a static region of a current image and a motion compensated disparity value of a previous image is determined. The result of the determination is taken into account for generating or refining a confidence map.

Abstract: A circuit for correcting the output of an audio level meter comprises input means for generating the square or the absolute value of an input signal, a low pass filter having a predetermined attack time and release time, and output means for converting the output signal from a linear scale to a logarithmic scale. The circuit further comprises a correction means to which an information about whether the input signal to the audio level meter was subject to squaring or converting into an absolute value at the input, as well as the attack and release time of the low pass filter, are supplied as input values, and which provides, at its output, a value representing the difference between the output of the audio level meter and the true signal power of the input signal.

Abstract: An apparatus and a method of determining a sequence of transitions for a varying state of a system, wherein the system is described by a finite number n of states, and wherein a transition from a current state to a next state causes a cost in dependence of a distance that is dependent on a previous state, the current state, and the next state.

Abstract: A method for determining an alpha value for a candidate pixel of an image in an alpha matting process and an apparatus configured to perform the method are described. A selecting unit selects two or more pairs of foreground pixels and background pixels. A calculating unit calculates for each pair of foreground pixel and background pixel an associated alpha value based on a color of the foreground pixel and a color of the background pixel, as well as a probability function based on the associated alpha value. The probability function is modeled by assuming two or more characteristics of the pixels as statistically independent. A determining unit determines the pair of foreground pixel and background pixel with the highest value of the probability function. A setting unit then sets the alpha value for the candidate pixel to the associated alpha value of the determined pair of foreground pixel and background pixel.

Abstract: A method for generating a confidence map comprising a plurality of confidence values, each being assigned to a respective disparity value in a disparity map assigned to at least two stereo images each having a plurality of pixels, wherein a single confidence value is determined for each disparity value, and wherein for determination of the confidence value at least a first confidence value based on a match quality between a pixel or a group of pixels in the first stereo image and a corresponding pixel or a corresponding group of pixels in the second stereo image and a second confidence value based on a consistency of the corresponding disparity estimates is taken into account.

Abstract: A method and an apparatus for recovering a component of a distortion field of an image of a set of multi-view images are described. Also described are a method and an apparatus for determining a disparity field of an image of a set of multi-view images, which makes use of such method. In a first step pixel correspondences between the image and another image of the set of multi-view images are determined. A disparity field is then determined from at least a subset of the determined pixel correspondences. The disparity field is smoothed and the component of the distortion field is estimated from the smoothed disparity field. When determining a second disparity field the estimated component of the distortion field is taken into account when determining pixel correspondences between the image and another image of the set of multi-view images.

Abstract: A method and an apparatus for bilayer image segmentation are described. A set of segmentation seeds for the image is generated by analyzing a depth histogram of the image. Then a segmentation map is generated by minimizing an objective function, which models a directed flow from the foreground segmentation seeds towards the background segmentation seeds.

Abstract: A method and an apparatus for bi-layer segmentation of an image or a sequence of images are described. A classifier is derived based on depth data of the image and another classifier is derived based on color data of the image. The image is then segmented by maximizing a weighted sum of matching scores derived from the classifiers based on depth data and color data of the image. The classifier based on color data of the image is derived using color sampling subsequent to generating an initial segmentation of the image.

Abstract: A method and an apparatus for multi-label segmentation of an image are described. First an energy function is determined for the image. Then for a homogeneous region of the image variables of the energy function are grouped to a single variable. Subsequently the energy function is minimized and labels are assigned to regions of the image based on the minimized energy function.

Abstract: In one embodiment, it is proposed a method for processing at least one disparity map associated to at least one left view and one right view of stereovision images. Such method is remarkable in that it comprises determining at least one modified disparity map that comprises, for a given pixel or group of pixels, a modified disparity value determined in function of disparity values of the at least one disparity map associated to pixels that belong to a neighborhood of the given pixel or group of pixels, said disparity values being weighted in function of a value obtained from the at least one disparity map and at least one other disparity map.

Abstract: A method and an apparatus for performing alpha matting on an image are described. A contour retrieving unit retrieves object contour information for the image. An edge decision unit then determines hard edges based on the retrieved object contour information using an edge model. Finally, an alpha assignment unit assigns alpha values to pixels of the image in a vicinity of a determined hard edge based on the edge model.

Abstract: A method and an apparatus for filtering an image using a guided image filter are described. A filter computes a filtering output using a guidance image. For calculating the filtering output confidence values associated to a filter input are taken into account. A confidence-weighted averaging is performed using the confidence values.

Abstract: A video processing apparatus and a method for detecting a temporal synchronization mismatch between at least a first and a second video stream of a stereoscopic video content are described. An eye blink of a creature that is imaged in the video content is detected. The temporal synchronization mismatch is determined by determination of a temporal offset between the reproduction of an eye blink in the first video stream and the reproduction of said eye blink in the second video stream.

Abstract: A method and an apparatus for determining a confidence value of a disparity estimate for a pixel or a group of pixels of a selected image of at least two stereo images are described, the confidence value being a measure for an improved reliability value of the disparity estimate for the pixel or the group of pixels. First an initial reliability value of the disparity estimate for the pixel or the group of pixels is determined, wherein the reliability is one of at least reliable and unreliable. Then a distance of the pixel or the group of pixels to a nearest pixel or group of pixels with an unreliable disparity estimate is determined. Finally, the confidence value of the disparity estimate for the pixel or the group of pixels is obtained from the determined distance.

Abstract: A solution for determining a similarity or dissimilarity measure for a selected pixel of a first image relative to another selected pixel in a second image is described. The first image and the second image form a stereoscopic image pair or part of a multi-view image group. In a first step a first support window containing the selected pixel in the first image is determining. Then a second support window containing the selected pixel in the second image is determining. Subsequently one or more statistical properties of the selected pixel in the first image are calculated to define a probability distribution for the selected pixel in the first image. Finally, pixel similarity or dissimilarity between the first support window and the second support window is aggregated using only those pixels belonging to the probability distribution for the selected pixel in the first image with a probability above a defined minimum.

Abstract: A method for determining an alpha value for a candidate pixel of an image in an alpha matting process and an apparatus configured to perform the method are described. A selecting unit selects two or more pairs of foreground pixels and background pixels. A calculating unit calculates for each pair of foreground pixel and background pixel an associated alpha value based on a color of the foreground pixel and a color of the background pixel, as well as a probability function based on the associated alpha value. The probability function is modeled by assuming two or more characteristics of the pixels as statistically independent. A determining unit determines the pair of foreground pixel and background pixel with the highest value of the probability function. A setting unit then sets the alpha value for the candidate pixel to the associated alpha value of the determined pair of foreground pixel and background pixel.

Abstract: An apparatus and a method for refining a value of a similarity measure are described. A similarity measure is assigned to a pixel or a group of pixels of a disparity map, which is assigned to at least two stereo images each having a plurality of pixels. The similarity measure constitutes an estimate for a match quality of the pixel or the group of pixels. For refinement of the value of the similarity measure, the similarity measure between a pixel or a group of pixels in a first stereo image and a corresponding pixel or a group of corresponding pixels in a second stereo image is determined. A contrast value for the pixel or the group of pixels of the first or the second stereo image is determined and the value of the similarity measure is corrected by a correction value that is a function of the determined contrast value.