British Telecommunications Public Limited Company

British Telecommunications Corporation

Daniel Ballin

Marco Gillies

Charles Nightingale

An object model of a first type such as an avatar is analyzed, and an object model of a second type, such as a clothing model, is generated. The object model of the first type is analyzed by comparing it to a set of pre-defined object models of the same type and deriving a set of weight values which when used as coefficients in a summation of the predefined object models result in an approximation to the input object model. The derived set of weight values can then be used to sum together a set of predefined object models of a second type to give a new predefined object model of a second type which corresponds to the input object model of the first type. Preferably, the input object model is a user avatar, and the output object model is a virtual item of clothing. The user avatar may then be displayed to a user 'wearing' the generated item of clothing.

Method and system for virtual object generation

A hierarchical behavioral framework is used to generate and control autonomous and semi-autonomous behavior in an articulate object. A behavioral controller is arranged to receive input associated with a behavioral action, to infer a plurality of behavioral parameter values using the framework, and to generate equivalent behavior in the articulate object using the parameter values when loaded in the behavioral controller to generate output corresponding to the equivalent behavior. The equivalent behavior may reproduce the inputted behavioral action, and/or include one or more other behavioral actions, which may be performed simultaneously or as part of a sequence of actions.

Apparatus and method for generating behaviour in an object

The invention provides an image processing method and system wherein an image is conceptually textured onto the surface of a three dimensional shape via a projection thereonto. The shape and/or the image position are then moved relative to each other, preferably by a rotation about one or more axes of the shape, and a second projection taken of the textured surface back to the image position to obtain a second, processed image. The view displayed within the processed image will be seen to have undergone an aspect ratio change as a result of the processing. The invention is of particular use in simulating the small movements of humans when speaking, and in particular of processing viseme images to simulate such small movements when displayed as a sequence.

Growing community of inventors

Ipswich, United Kingdom

Daniel Ballin