Image Number 5 for United States Patent #5818461.
The present invention provides methods and apparatus for creating lifelike digital representations of three-dimensional objects. Steps and means are described by which an animator may efficiently select and combine a plurality of animated objects to generate a new animated objects. According to the present invention, an animator may store a plurality of object shapes. To minimize storage, all of the object shapes are stored solely as differences from a base object. To form a new object, an animator may select from the database a number of objects with the desired characteristics and these objects will be blended with one another to create a new object with features that are a combination of the selected objects. The extent to which a selected object is incorporated into the new object may be selected by the animator by assigning each selected object with a weight. The new object is then a sum of each of the differences between each of the selected objects and the base object multiplied by their respective weights. The present invention also includes an improved graphical user interface for animating facial objects.