Three Dimensional (3D) Modeling of a Complex Control Object

A non-transitory computer readable medium having instructions to initialize predictive information that models a complex control object in a three dimensional (3D) sensory space stored thereon, which instructions when executed by a processor, perform actions including:

• accessing observed information including a set of points corresponding to surface points on a surface of a complex control object in a three dimensional (3D) sensory space;
• sensing an actual position of at least one complex control object in a first reference frame of the 3D sensory space;
• sensing an apparent position of the complex control object, wherein the apparent position as sensed is different from the actual position as sensed;
• obtaining a second reference frame calculated to account for apparent position of the complex control object;
• obtaining (i) the actual and apparent positions of the complex control object into a common reference frame using a transformation that renders the actual position in the first reference frame and the apparent position in the second reference frame into a common reference frame, transforming, (ii) an actual position transformed from the sensed apparent position and (iii) a normalized orientation of the complex control object transformed from the set of points;
• obtaining a selection an archetype that represents a pose of the complex control object in the normalized orientation selected from a plurality of observed information archetypes;
• and
• initializing predictive information that models the complex control object from initialization parameters associated with the archetype selected.

A system including:

• an optical sensor;
• a processor;
• and a memory storing instructions to initialize predictive information that models a complex control object in a three dimensional (3D) sensory space, which instructions when executed by the processor perform:
• accessing observed information including a set of points corresponding to surface points on a surface of a complex control object in a three dimensional (3D) sensory space;
• sensing an actual position of at least one complex control object in a first reference frame of the 3D sensory space;
• sensing an apparent position of the complex control object, wherein the apparent position as sensed is different from the actual position as sensed;
• obtaining a second reference frame calculated to account for apparent position of the complex control object;
• obtaining (i) the actual and apparent positions of the complex control object into a common reference frame using a transformation that renders the actual position in the first reference frame and the apparent position in the second reference frame into a common reference frame, transforming, (ii) an actual position transformed from the sensed apparent position and (iii) a normalized orientation of the complex control object transformed from the set of points;
• obtaining a selection of an archetype that represents a pose of the complex control object in the normalized orientation selected from a plurality of observed information archetypes;
• and
• initializing predictive information that models the complex control object from initialization parameters associated with the archetype selected.