Visual based identitiy tracking

1. A method of creating a signature of a person's face for use in a computer based gaming application, comprising:

executing, by a computing system, the computer based gaming application, the computer based gaming application including interaction with a subset of a set of previously enrolled players;

determining, by the computing system, that a person has become detectable in a physical play space of the computer based gaming application, the person not being one of the subset of the set of previously enrolled players currently interacting with the computer based gaming application;

receiving, at the computing system, a depth image of the person obtained from a depth image sensor, the depth image comprises a two dimensional arrangement of pixels, each pixel of the depth image stores a number identifying a distance from the depth image sensor to a body of the person;

receiving, at the computing system, a visual image of the person from a visual image sensor, the visual image being different than the depth image;

correlating, by the computing system, the depth image to the visual image so that pixels in the visual image are identified to correspond to pixels in the depth image;

identifying, by the computing system, foreground pixels and background pixels in the two dimensional arrangement of pixels in the depth image;

removing, by the computing system, the background pixels from the depth image;

assigning, by the computing system, the foreground pixels of the depth image to different parts of the body of the person;

using, by the computing system, the different parts of the body of the person to construct a model;

utilizing, by the computing system, the model to estimate a position of the person's face in the depth image by using one or more criteria to locate eyes and nose of the person;

automatically, by the computing system, determining a bounding box for the depth image that surrounds the person's face;

automatically, by the computing system, identifying pixels in the visual image that correspond to pixels in the depth image inside the bounding box;

automatically, by the computing system, creating and storing a new signature of the person's face that represents the pixels in the visual image that correspond to the pixels in the depth image inside the bounding box;

automatically, by the computing system, identifying the person as a specific previously enrolled player of the set of previously enrolled players based on comparing the new signature of the person's face to stored signatures of the set of previously enrolled players; and

mapping, by the computing system, the person to the determined specific enrolled player and enabling the person to interact with the computer based gaming application based on the mapping.

2. The method of claim 1 , further comprising:

adjusting, by the computing system, the bounding box to more tightly bound the eyes and nose.

3. The method of claim 1 , further comprising:

realigning, by the computing system, the visual image so that the face is at a predetermined reference alignment.

4. The method of claim 1 , further comprising:

smoothing, by the computing system, high variance noise from the depth image.

5. The method of claim 1 , further comprising:

creating, by the computing system, a multi-dimensional vector for pixels of the visual image that correspond to pixels in the depth image inside the bounding box and transforming the vector to Principal Component Analysis space.

6. The method of claim 1 , further comprising:

establishing thresholds, by the computing system, the thresholds are used when automatically identifying the person as the specific enrolled player; and

dynamically updating, by the computing system, the thresholds while and based on performing the computer based gaming application.

7. The method of claim 1 , further comprising:

mapping, by the computing system, the person to an object associated with the computer based gaming application based on the signature of the person's face; and

interacting, by the computing system, with the person using the computer based gaming application based on the mapping.

8. The method of claim 7 , further comprising:

the interacting, by the computing system, with the person includes controlling an avatar in a video game based on motion of the person.

9. The method of claim 1 , further comprising:

determining, by the computing system, that insufficient data exists for identifying the person;

requesting, by the computing system, the person to perform a specific act in response to determining that insufficient data exists for identifying the person; and

obtaining, by the computing system, sufficient data in response to the specific act.

10. A computing apparatus, comprising:

a sensor system for providing one or more depth images and one or more visual images; and

one or more processors for communication with the sensor system, the one or more processors for execution of a computer based gaming application, the computer based gaming application including interaction with a subset of a set of previously enrolled players;

the one or more processors for determining that a person has become detectable in a physical play space of the computer based gaming application, the person not being one of the subset of the set of previously enrolled players currently interacting with the computer based gaming application;

the one or more processors for receiving a depth image of the person from the sensor system, the depth image comprises a two dimensional arrangement of pixels, each pixel of the depth image stores a number identifying a distance from the sensor system to a body of the person;

the one or more processors for identifying foreground pixels and background pixels in the two dimensional arrangement of pixels in the depth image;

the one or more processors for receiving a visual image of the person from the sensor system, the visual image being different than the depth image;

the one or more processors for correlating the depth image to the visual image so that pixels in the visual image are identified to correspond to pixels in the depth image;

the one or more processors for assigning the foreground pixels to different parts of the body of the person;

the one or more processors for using the different parts of the body to construct a model and using the model to estimate a position of a person's face in the visual image that corresponds to the position of the person's face in the depth image by using one or more criteria to locate eyes and nose of the person;

the one or more processors for determining a bounding box for the depth image that surrounds the person's face;

the one or more processors for identifying pixels in the visual image that correspond to pixels in the depth image inside the bounding box;

the one or more processors for creating a new signature of the person's face representing pixels in the visual image that correspond to the pixels in the depth image inside the bounding box;

the one or more processors for identifying the person as a specific previously enrolled player of the set of previously enrolled players based on comparing the new signature of the person's face to stored signatures of the set of previously enrolled players; and

the one or more processors for mapping the person to the determined specific enrolled player and enabling the person to interact with the computer based gaming application based on the mapping.

11. The computing apparatus of claim 10 , wherein:

the one or more processors for creating the new signature of the person's face by creating a multi-dimensional vector for pixels in the visual image of the person at the position of the person's face.

12. The computing apparatus of claim 10 , wherein:

the sensor system includes a depth camera and visual image camera.

13. The computing apparatus of claim 10 , further comprising:

a display, the display is updated based on recognizing the person's face.

14. The computing apparatus of claim 10 , wherein:

the one or more processors for mapping the person to an object associated with the computer based gaming application based on recognizing the person's face.

15. One or more processor readable nonvolatile storage devices having processor readable code embodied on said one or more processor readable non-transitory storage devices, said processor readable code for programming one or more processors to perform a method comprising:

executing a computer based gaming application, the computer based gaming application including interaction with a subset of a set of previously enrolled players;

determining that a person has become detectable in a physical play space of the computer based gaming application, the person not being one of the subset of the set of previously enrolled players currently interacting with the computer based gaming application;

receiving a depth image of the person obtained from a depth image sensor, the depth image comprises a two dimensional arrangements of pixels, each pixel of the depth image stores a number identifying a distance from the depth image sensor to a body of the person;

receiving a visual image of the person from a visual image sensor, the visual image being different than the depth image;

correlating the depth image to the visual image so that pixels in the visual image are identified to correspond to pixels in the depth image;

identifying foreground pixels and background pixels in the two dimensional arrangement of pixels in the depth image;

assigning each of the foreground pixels to a different part of the body of the person;

using the different parts of the body to construct a model;

utilizing the model to estimate a position of a person's face in the visual image that corresponds to the position of the person's face in the depth image by using one or more criteria to locate eyes and nose of the person;

determining a bounding box for the depth image that surrounds the person's face;

identifying pixels in the visual image that correspond to pixels in the depth image inside the bounding box;

determining a new signature for an appearance of the person's face representing pixels in the visual image that correspond to the pixels in the depth image inside the bounding box;

identifying the person as a specific previously enrolled player of the set of previously enrolled players based on comparing the new signature to stored signatures of the set of previously enrolled players; and

mapping the person to the determined specific enrolled player and enabling the person to interact with the computer based gaming application based on the mapping.

16. One or more processor readable nonvolatile storage devices according to claim 15 , wherein:

the method further comprises mapping the person to an object associated with a computer based gaming application based on the signature of the person's face.