Abstract: An image processing method includes partitioning an image under test to form a plurality of contiguous image segments having similar image properties, deriving feature data from a subset including one or more of the image segments, and comparing the feature data from the subset of image segments with feature data derived from respective image segments of one or more other images so as to detect a similarity between the image under test and the one or more other images.

Abstract: A method of producing the results of a search for content for display on a client is described.

Abstract: An image processing method includes partitioning an image under test to form a plurality of contiguous image segments having similar image properties, deriving feature data from a subset including one or more of the image segments, and comparing the feature data from the subset of image segments with feature data derived from respective image segments of one or more other images so as to detect a similarity between the image under test and the one or more other images.

Abstract: An image processing method includes partitioning an image under test to form a plurality of contiguous image segments having similar image properties, deriving feature data from a subset including one or more of the image segments, and comparing the feature data from the subset of image segments with feature data derived from respective image segments of one or more other images so as to detect a similarity between the image under test and the one or more other images.

Abstract: An image processing method includes partitioning an image under test to form a plurality of contiguous image segments having similar image properties, deriving feature data from a subset including one or more of the image segments, and comparing the feature data from the subset of image segments with feature data derived from respective image segments of one or more other images so as to detect a similarity between the image under test and the one or more other images.

Abstract: A method comprising: obtaining a disparity map for one of a left eye image and a right eye image, the disparity map indicating the horizontal displacement between corresponding pixels in the left eye image and the right eye image; generating a luminance representation of the disparity map, wherein a disparity unit which defines the resolution of the disparity map, is mapped to a corresponding luminance value; and companding the luminance value for transmission over a luminance channel is described.

Abstract: An image processing method includes partitioning an image under test to form a plurality of contiguous image segments having similar image properties, deriving feature data from a subset including one or more of the image segments, and comparing the feature data from the subset of image segments with feature data derived from respective image segments of one or more other images so as to detect a similarity between the image under test and the one or more other images.

Abstract: A method of producing the results of a search for content for display on a client is described.

Abstract: A method of generating a disparity map is described.

Abstract: A method of cutting between a first image sequence and a second image sequence displayable on a screen which defines a screen plane, wherein the first image sequence includes a stereoscopically viewable object perceivable by a viewer on a first image plane different to the screen plane and the second image sequence including a second object perceivable on a second image plane different to the first image plane, the stereoscopically viewable object comprising a first image at a first screen plane position and a second image at a second screen plane position, the method comprising: identifying the first screen plane position of the first image and the second screen plane position of the second image; inserting into a first post-cut frame of the second image sequence a first copy of the second object substantially at the first screen plane position and inserting a second copy of the second object substantially at the second screen plane position, and inserting the first copy of the second object and the second c

Abstract: A security device and system is disclosed. This security device is particularly useful in a security system where there are many security cameras to be monitored. This device automatically highlights to a user a camera feed in which an incident is occurring. This assists a user in identifying incidents and to make an appropriate decision regarding whether or not to intervene. This highlighting is performed by a trigger signal generated in accordance with a comparison between a sequence of representations of sensory data and other corresponding sequences of representations of sensory data.

Abstract: A method comprising: obtaining a disparity map for one of a left eye image and a right eye image, the disparity map indicating the horizontal displacement between corresponding pixels in the left eye image and the right eye image; generating a luminance representation of the disparity map, wherein a disparity unit which defines the resolution of the disparity map, is mapped to a corresponding luminance value; and companding the luminance value for transmission over a luminance channel is described.

Abstract: An apparatus for determining misalignment between a first image and a second image, the first and second images being viewable stereoscopically, the apparatus comprising: a determiner operable to determine a feature position within the first image and a corresponding feature position within the second image; a definer operable to define, within the first image and the second image, the optical axis of the cameras capturing said respective images; a calculator operable to calculate the misalignment between the feature position within the first image and the corresponding feature position within the second image using a model, the misalignment being determined in dependence upon the location of the feature position of the first image and the corresponding feature position of the second image relative to the defined optical axis of the respective images; and a tester operable to test the validity of the calculated misalignment using a random sample consensus technique, whereby the tested misalignment is valid w

Abstract: A method of determining misalignment between a first image and a second image, the first and second images being viewable stereoscopically, the method comprising: determining a feature position within the first image and a corresponding feature position within the second image; defining, within the first image and the second image, the optical axis of the cameras capturing said respective images; and calculating the misalignment between at least one of scale, roll or vertical translation of the feature position within the first image and the corresponding feature position within the second image, the misalignment being determined in dependence upon the location of the feature position of the first image and the corresponding feature position of the second image relative to the defined optical axis of the respective images is described. A corresponding apparatus is also described.

Abstract: An information handling apparatus in which textual metadata is generated in respect of a current information item in an ensemble of information items including: a mechanism that detects one or more predetermined properties of the current information item; a mechanism that detects a subset of information items from the ensemble of information items the subset being those which have the one or more predetermined properties most similar to those of the current information item; and a mechanism that selects one or more most frequently occurring words and/or phrases within textual metadata associated with the subset of information items, for use in textual metadata to be associated with the current information item.

Abstract: An image processing method includes partitioning an image under test to form a plurality of contiguous image segments having similar image properties, deriving feature data from a subset including one or more of the image segments, and comparing the feature data from the subset of image segments with feature data derived from respective image segments of one or more other images so as to detect a similarity between the image under test and the one or more other images.

Abstract: An information handling method in which information items are mapped to respective nodes in an array of nodes by mutual similarity of the information items, such that similar information items map to nodes at similar positions in the array of nodes; a mapping relationship being defined between feature data relating to an information item and nodes in the array of nodes comprises the steps of: (a) for each of a group of information items: detecting feature data representing each of a set of properties of that information item; (b) for the ensemble of feature data corresponding to the group of information items: detecting more significant and less significant properties within the ensemble; discarding feature data relating to less significant properties, the remaining non-discarded feature data being associated so as to generate a reduced feature vector for each information item; (c) defining a mapping between the reduced feature vectors and nodes in the array of nodes.

Abstract: A method of inserting object data into a stereoscopic image for display on a screen, comprising the steps of: providing a first image having a foreground component and a second image having a foreground component, the second image foreground component being a horizontally displaced version of the first image foreground component; inserting a first opaque section into the first image, and a second opaque section in the second image, the second opaque section being a horizontally displaced version of the first opaque section, wherein the displacement between the first image component and the second image component is less than the displacement between the first opaque section and the second opaque section; and inserting the object data onto the first opaque section for display on the screen.

Abstract: A method of producing a first stereoscopic image is described.

Abstract: An image capturing system for capturing a stereo pair of images comprising a first image and a second image. The system comprises a first camera operable to capture the first image of the stereo pair of images, and a second camera operable to capture the second image of the stereo pair of images. The system further comprises a communication link operable to connect the first camera with the second camera so that the first camera can communicate with the second camera. The first camera is operable to transmit second camera control data to the second camera via the communication link so as to control functions of the second camera which relate to the capture of the second image.