Abstract: A method of generating an image of a scene from a target viewpoint is provided. The method comprises capturing a plurality of multi-plane images of a virtual scene, and determining for each multi-plane image, whether a frontmost clipping plane pair defined for that multi-plane image intersects a virtual object in the virtual scene. Responsive to a positive determination for a respective clipping plane pair, the corresponding image data in the corresponding image plane is identified and assigned an identifier, thus generating a modified version of the corresponding multi-plane image. Camera information is obtained for the one or more virtual cameras that captured the multi-plane images and the target virtual camera. An output image is then generated by combining at least one of the modified multi-plane images with another of the multi-plane images in accordance with the obtained camera information. A corresponding system is also provided.

Abstract: A method is disclosed of detecting whether a first user is a bad actor within a platform network, in which the platform network enables interactions between individuals within video games, the method comprising the steps of measuring for the first user one or more selected from the list consisting of at least a first a game-agnostic behaviour metric that measures an investment of effort by the first user in interactions with other individuals; and at least a first a game-dependent behaviour metric that measures in-game patterns of interaction with other individuals by the first user, comparing the or each metric with an average or reference metric for a typical individual, and where a difference in the comparison or a combination of the comparisons meets a first criterion, treating this as indicative that the first user may be a bad actor.

Abstract: A processing device comprising: an identification unit configured to identify a first action in a given game corresponding to a first input signal; the identification unit is configured to identify a second action in a reference game corresponding to a second input signal, the second action being identified as a corresponding action that corresponds to the first action; and a control unit is configured to interpret the second input signal as the first input signal for the given game in response to the second action being identified as the corresponding action.

Abstract: An image rendering method for a virtual scene includes, for a plurality of IDs, generating a respective mask identifying elements of the scene that are associated with a respective ID; for the resulting plurality of masks, dividing a respective mask into a plurality of tiles; and discarding tiles that do not identify any image elements; for the resulting plurality of remaining tiles, selecting a respective trained machine learning model from among a plurality of machine learning models, the respective machine learning model having been trained to generate data contributing to a render of at least a part of an image, based upon elements of the scene associated with the same respective ID as the elements identified in the mask from which the respective tile was divided; and using the respective trained machine learning model to generate data contributing to a render of at least a part of the image based upon input data at least for the identified elements in the respective tile.

Abstract: A hybrid codec for training a neural coding system, the hybrid codec including a real codec for encoding data pre-processed by a first neural network, and for decoding the encoded data for post-processing by a second neural network; and a proxy codec that is a differentiable representation of the real codec, and is for back-propagating a loss function, representative of a coding bit-rate and a coding distortion between the post-processed data and the original data, for training the weights of at least the first neural network.

Abstract: A hybrid codec for training a neural coding system, the hybrid codec including a real codec for encoding data pre-processed by a first neural network, and for decoding the encoded data for post-processing by a second neural network; and a proxy codec that is a differentiable representation of the real codec, and is for back-propagating a loss function, representative of a coding bit-rate and a coding distortion between the post-processed data and the original data, for training the weights of at least the first neural network.

Abstract: A method of image coding comprises the steps of, for a respective image, training a respective neural network to output pixel data for that image and outputting the trained neural network as the encoded representation of the image; meanwhile a method of image decoding comprises the steps of receiving a respective neural network trained to output pixel data for one image and prompting the neural network to output the pixel data for that image.

Abstract: A method of generating a training set for a neural precomputed light model includes: generating a plurality of candidate viewpoints of a scene, culling candidate viewpoints according to a probability that depends upon a response of the surface of the scene to light at a surface position in the scene corresponding to the viewpoint, and generating training images at the remaining viewpoints.

Abstract: An image rendering method for rendering a pixel at a viewpoint including: for a first element of a virtual scene, having a predetermined surface at a position within that scene, providing the position and a direction based on the viewpoint to a machine learning system previously trained to predict a factor that, when combined with a distribution function that characterises an interaction of light with the predetermined surface, generates a pixel value corresponding to the first element of the virtual scene as illuminated at the position, combining the predicted factor from the machine learning system with the distribution function to generate the pixel value corresponding to the illuminated first element of the virtual scene at the position, and incorporating the pixel value into a rendered image for display.

Abstract: An image rendering method for rendering a pixel at a viewpoint includes, for a first element of a virtual scene, having a predetermined surface at a position within that scene; providing the position and a direction based on the viewpoint to a machine learning system previously trained to predict a factor that, when combined with a distribution function that characterises an interaction of light with the predetermined surface, generates a pixel value corresponding to the first element of the virtual scene as illuminated at the position, combining the predicted factor from the machine learning system with the distribution function to generate the pixel value corresponding to the illuminated first element of the virtual scene at the position, and incorporating the pixel value into a rendered image for display, where the machine learning system was previously trained with a training set based on images comprising multiple lighting conditions.

Abstract: An image rendering method for rendering a pixel at a viewpoint includes: for a first element of a virtual scene, having a predetermined surface at a position within that scene, evaluating whether to render a pixel corresponding to the first element using a machine learning system having been trained to output a value representative of the lighting of the predetermined surface at the position, or using an alternative rendering approach, and rendering the pixel according to which of the machine learning system and the alternative rendering approach are chosen.

Abstract: An image rendering method for rendering a pixel at a viewpoint includes: for a first element of a virtual scene, having a predetermined surface at a position within that scene, evaluating whether to render a pixel corresponding to the first element using at least a first machine learning system having been trained to generate an illuminance output representative of the lighting of the predetermined surface at the position, or using an alternative rendering approach, and rendering the pixel according to which of the at least first machine learning system and the alternative rendering approach are chosen in the evaluating step; where the evaluating step comprises obtaining a confidence value from the at least first machine learning system indicative of the accuracy of the illuminance output, the machine learning system having been trained to generate the confidence value in conjunction with the illuminance output, and the rendering step comprises using the alternative rendering approach if the confidence value

Abstract: A method and system for improving a resolution of an image frame is described. The image frame is part of a data stream. The data stream may be a video game, for example. The method comprising the following steps. Sending an encoded version of the image frame from a server to a client device via a first communication channel between the server and the client device. Detecting a reduced bandwidth of the first communication channel. In response to detecting the reduced bandwidth, determining data to be sent to the client device via a second communication channel. The second communication channel is separate to the first communication channel and the data relates to the image frame. The determined data is sent via the second communication channel.

Abstract: An image rendering method for rendering a pixel of a virtual scene at a viewpoint includes: downloading a machine learning system corresponding to a current or anticipated state an application determining the virtual scene to be rendered from among a plurality of machine learning systems corresponding to a plurality of states of the application; providing a position and a direction based on the viewpoint to the machine learning system previously trained to predict a factor; combining the predicted factor from the machine learning system with a distribution function that characterises an interaction of light with a predetermined surface to generate the pixel value corresponding to an illuminated first element of the virtual scene at the position; and incorporating the pixel value into a rendered image for display.

Abstract: A method of generating an image of a scene from a target viewpoint is provided. The method comprises capturing a plurality of multi-plane images of a virtual scene, and determining for each multi-plane image, whether a frontmost clipping plane pair defined for that multi-plane image intersects a virtual object in the virtual scene. Responsive to a positive determination for a respective clipping plane pair, the corresponding image data in the corresponding image plane is identified and assigned an identifier, thus generating a modified version of the corresponding multi-plane image. Camera information is obtained for the one or more virtual cameras that captured the multi-plane images and the target virtual camera. An output image is then generated by combining at least one of the modified multi-plane images with another of the multi-plane images in accordance with the obtained camera information. A corresponding system is also provided.

Abstract: A method of enhancing a 3D printed model includes generating successive visualisations of a virtual environment comprising a target object, receiving a user input indicating selection of a visualisation of the target object at a particular moment in time, generating visualisation data to enable subsequent visualisation of at least the target object as at the particular moment in time, causing the visualisation data to be stored at a unique location; generating 3D print model data for 3D printing of the target object, and associating data identifying the unique location of the stored visualisation data with the 3D print model data.

Abstract: The present disclosure relates to a computer-implemented method for completing an image, the method comprising the steps of dividing data of an image to be completed into a plurality of image portions. The method entails applying a first filling process to fill a first image portion comprising a first hole, the first hole associated with a first quantity and/or a first quality; and applying a second filling process to fill a second image portion comprising a second hole, the second hole associated with a second quantity different to the first quantity and/or a second quality different to the first quality, the second process being different to first process. The method then includes combining the filled first and second image portions to complete the image.

Abstract: A method of 3D print modelling includes: obtaining a target virtual object for 3D printing, evaluating the target virtual object for 3D printing with respect to one or more printability criteria to detect whether the 3D printed model would have one or more predetermined undesirable physical characteristics, and if so, generating 3D printable model data defining a transparent matrix to be printed surrounding the 3D printed model of the target virtual object.

Abstract: An image rendering method for rendering a pixel at a viewpoint including: for a first element of a virtual scene, having a predetermined surface at a position within that scene, providing the position and a direction based on the viewpoint to a machine learning system previously trained to predict a factor that, when combined with a distribution function that characterises an interaction of light with the predetermined surface, generates a pixel value corresponding to the first element of the virtual scene as illuminated at the position, combining the predicted factor from the machine learning system with the distribution function to generate the pixel value corresponding to the illuminated first element of the virtual scene at the position, and incorporating the pixel value into a rendered image for display.

Abstract: An image rendering method includes: selecting at least a first trained machine learning model from among a plurality of machine learning models, the machine learning model having been trained to generate data contributing to a render of at least a part of an image, where the at least first trained machine learning model has an architecture based learning capability that is responsive to at least a first aspect of a virtual environment for which it is trained to generate the data, and using the at least first trained machine learning model to generate data contributing to a render of at least a part of an image.