Abstract: In a method of and an apparatus for processing image data representing an image saliency data for the image is generated by determining a series of features of the image and the determined features are used to generate a probability measure for each point of the image representative of a location of a subject of the image. The saliency data is processed using respective ones of weighting functions of a plurality of spatial scales in order to determine the positions of regions of interest of the image at respective ones of the scales. Response data is generated for each scale representing the relative strength of response of the saliency data at the positions of the determined regions of interest to the function at that scale.

Abstract: A method for processing an image sequence captured using an image capture device, wherein a captured sequence to be stabilized by estimating motion of the device, abstracting data from the image sequence, stabilizing the image sequence in response to the estimated motion of the device, and forming an output image sequence from the stabilized image sequence and the abstracted data.

Abstract: A method of determining a regular grid pattern from a surface coded pattern that comprises the regular grid pattern interleaved with a further data carrying pattern wherein the surface coded pattern is subject to perspective distortion, the method comprising: extracting a set of straight line hypotheses from the coded surface pattern; clustering the straight line hypotheses by orientation; for each cluster, extracting a set of line pencil hypotheses; generating a set of regular grid hypotheses from pairs of the line pencil hypotheses; and determining the best regular grid hypothesis.

Abstract: A video stabilization scheme which uses a dynamic programming algorithm in order to determine a best path through a notional stabilization band defined in terms of a trajectory of an image capture element. The best path is used in order to produce an optimally stabilized video sequence within the bounds defined by the image capture area of the image capture element.

Abstract: A variable skew correction system comprises a de-skew application executable to transform scanned image data of a document exhibiting a variable skew condition to an output model representing a de-skewed image of the document by transferring pixel data for each of a plurality of raster lines of the scanned image data to the output model, wherein a variable spacing between at least two adjacent raster lines of the scanned image data is modified in the output model to correct non-linear distortion of the scanned image data.

Abstract: A method of displaying a digital image comprises the steps of: providing a processor with image data for generating a displayable image, which image data is from a plurality of static digital images, the processor generating a displayable rendered image from the image data, determining a rostrum path to display part of the displayable rendered image and generating a novel viewpoint simulating a different viewing angle. A corresponding computer system is also disclosed.

Abstract: The present invention provides an apparatus and method which facilitates the use of a printed or scribed document as an interface with a computer. The apparatus in one aspect comprises: a printed or scribed document bearing specialized calibration marks, the document being positioned on a work surface; a camera focused on the document for generating video signals representing the document in electronic form; and a processor linked to the camera for processing an image captured by the camera and configured to identify the calibration marks of the document in the captured image and then determine from the location of the calibration marks in the image a transformation between co-ordinates of features. In the image and corresponding co-ordinates of features in the document that compensates for the freely variable positioning of the document on the work surface.

Abstract: A method of displaying digital images comprises the steps of using a processor to determine an extent of similarity between a first image part and a second image part, displaying a first image part from a viewpoint and transitioning to displaying a second image part from a viewpoint, the second image part being selected at least partly based on a determined extent of similarity between the first image part and the second image part, in which the viewpoint of an image part is moved either during the transition or after the transition takes place. An appropriate system is provided.

Abstract: Zoomed sequential images including a first image at a first magnification, a later image at a second magnification, and at least one intermediate image at a magnification between the first and second magnifications are processed by applying an electronic zoom to each of the intermediate images to magnify each of the intermediate images to the same size as the image size at the higher magnification level of the first and second magnifications.

Abstract: A method for computing apparatus to create a page with a plurality of information items laid out upon it is described. A plurality of information items and a definition of one or more groups of information items to be formed on the page are received, as is a candidate layout of the information items on the page. Perceptual distances between information items in the candidate layout are determined, the perceptual distances representing both physical distance and graphical variation between relevant information items. It is determined from the perceptual distances whether the candidate layout satisfies predetermined criteria such that the candidate layout is perceived as visually representing the one or more groups. If the candidate layout is not perceived as visually representing the one or more groups, it is modified or regenerated until it does so.

Abstract: Detectors capable of accurately detecting and tracking moving features of such as faces within a video stream are sometimes too slow to be run in real-time. The present invention rapidly scans video footage in real-time and generates a series of preattemptive triggers indicating the frames and locations within the frames at our deserving of further investigation by a sub real-time detector. The triggers are generated by looking for peaks in a time variant measure such as the amount of symmetry within a frame or portion of a frame.

Abstract: Apparatus for providing electronic data associated with an item to a user, the apparatus comprising a memory tag scanner arranged to communicate with a memory tag associated with an item, a user profile memory arranged to store user profile data therein, an output interface arranged to provide data to a user, and a data processor arranged to cause data to be read from the memory tag associated with an item and the data to be selectively provided to a user via the output interface in accordance with the stored user profile data.

Abstract: A detector for use with a data encoding pattern, the pattern comprising a plurality of markings, wherein the detector is operable to generate pattern data representing a portion of the pattern imaged by an image capture device of the detector, process the pattern data in order to generate a model representing a form of the markings for the pattern, and use at least one of the pattern data and model data representing the generated model to determine a position for the detector, a corresponding method and apparatus.

Abstract: A data encoding pattern comprising a plurality of markings respectively composed from at least two sub-markings, wherein the colour of respective ones of the sub-markings and their arrangement for a marking is reserved.

Abstract: An image capture device electronically captures and selects a desired text portion. A visible light image is projected onto text on a document to enable a user to position the light image relative to the desired text portion. At least some of the text is captured in accordance with a position of the light image. A text selector selects the desired text portion from the captured text.

Abstract: A method for selectively adjusting the amount of specular reflections appearing in the final image output by an image capturing device by blending two substantially perfectly overlapping images, one with natural specular reflections and one without specular reflections. In the combined image, the amount of specular reflection is lower than in the image having natural specular reflections, to the extent in many cases that the underlying image content can be seen. In any event, the apparatus provides means for selectively adjusting the intensity of specular reflection, or even eliminating it altogether, in each region of the image containing such reflection.

Abstract: An attention detection system detects a condition of shared attention of plural persons. The system includes plural body language detectors each associated with a different person for detecting body language of the associated person. An analyzer receives body language information from the body language detectors, analyzes the body language of the persons and determines when said body language information indicates shared attention between the persons. The attention detection system generates a signal that captures an image of the shared attention.

Abstract: In a method of and an apparatus for processing image data representing an image saliency data for the image is generated by determining a series of features of the image and the determined features are used to generate a probability measure for each point of the image representative of a location of a subject of the image. The saliency data is processed using respective ones of weighting functions of a plurality of spatial scales in order to determine the positions of regions of interest of the image at respective ones of the scales. Response data is generated for each scale representing the relative strength of response of the saliency data at the positions of the determined regions of interest to the function at that scale.

Abstract: A memory tag is a transponder device with a memory for storing digital content. Purchase-related information for a product is stored in such a memory tag. A write capability of the memory tag is disabled.

Abstract: A method of processing image data representing a captured image, the data generated using an image capture element of an image capture device, the method comprising generating state data representing at least one characteristic of motion of the image capture device and a status of at least one setting of the device during a predetermined time period, and processing the image data on the basis of the state data in order to provide a measure of the accuracy of framing of the captured image within a field of view of the device, and corresponding apparatus.