Method and system for hand pose detection

1. A system for computer human interaction comprising:

a camera configured to generate data of a hand of a user;

an output device;

a memory storing at least a first neural network; and

a processor operatively connected to the camera, the output device, and the memory, the processor being configured to:

receive map data of the hand of the user from the camera;

generate, using the first neural network, a first plurality of activation features based at least in part on the map data, the first plurality of activation features corresponding to a joint angle of a wrist in the hand;

perform a first search in a predetermined plurality of activation features stored in a database stored in the memory to identify a first plurality of hand pose parameters for the wrist;

generate a hand pose model corresponding to the hand of the user based at least in part on the first plurality of hand pose parameters; and

generate an output with the output device in response to input from the user based at least in part on the hand pose model.

2. The system of claim 1 , wherein the processor is further configured to:

identify a second neural network stored in the memory based upon the first plurality of hand pose parameters, the second neural network being one neural network in a plurality of neural networks stored in the memory trained to classify features corresponding to joint angles of a first finger of the hand;

generate a second plurality of activation features using the map data of the hand of the user and the second neural network; and

generate the hand pose model corresponding to the hand of the user based further on the second plurality of activation features.

3. The system of claim 2 , wherein the processor is further configured to:

perform a second search in the predetermined plurality of activation features stored in the database to identify a second plurality of hand pose parameters for the one finger; and

generate the hand pose model corresponding to the hand of the user based on the first plurality of hand pose parameters and the second plurality of hand pose parameters.

4. The system of claim 1 , wherein the processor is further configured to perform the first search by performing spatial nearest neighbor search.

5. The system of claim 1 , wherein the processor is further configured to perform the first search by searching a set of the predetermined plurality of activation features stored in the database that correspond to at least one set of previous map data received within a predetermined time period of generation of the map data.

6. The system of claim 1 , wherein the first plurality of hand pose parameters for the wrist includes angle information about an orientation of the wrist in six degrees of freedom.

7. The system of claim 6 , wherein the six degrees of freedom further includes at least one degree of freedom for a longitudinal rotation that extends along a forearm.

8. The system of claim 7 , wherein the six degrees of freedom further includes at least one degree of freedom for an up-down angle for a raised or lowered palm of the hand.

9. The system of claim 8 , wherein the six degrees of freedom further includes at least one degree of freedom for a left-right rotational angle of the hand.

10. The system of claim 6 , wherein the first plurality of hand pose parameters for the wrist includes angle information about an orientation of the wrist in at least four degrees of freedom.

11. A method for computer human interaction comprising:

using a camera configured to generate data of a hand of a user;

receive map data of the hand of the user from the camera;

using a processor to generating, using the first neural network stored in a memory, a first plurality of activation features based at least in part on the depth map data, the first plurality of activation features corresponding to a joint angle of a wrist in the hand;

using the processor to perform a first search in a predetermined plurality of activation features stored in a database stored in the memory to identify a first plurality of hand pose parameters for the wrist;

using the processor to generate a hand pose model corresponding to the hand of the user based at least in part on the first plurality of hand pose parameters; and

generating an output with the output device in response to input from the user based at least in part on the hand pose model.

12. The method of claim 11 , further comprising:

identifying, using the processor, a second neural network stored in the memory based upon the first plurality of hand pose parameters, the second neural network being one neural network in a plurality of neural networks stored in the memory trained to classify features corresponding to joint angles of a first finger of the hand; and

generating, using the processor, a second plurality of activation features using the map data of the hand of the user and the second neural network.

13. The method of claim 12 , further comprising:

performing, using the processor, a second search in the predetermined plurality of activation features stored in the database to identify a second plurality of hand pose parameters for the one finger using nearest neighbor identification; and

generating, using the processor, the hand pose model corresponding to the hand of the user based on the first plurality of hand pose parameters and the second plurality of hand pose parameters.

14. The method of claim 11 , wherein the first search is a spacial nearest neighbor search.

15. The method of claim 11 , wherein the first search includes a search of a set of the predetermined plurality of activation features stored in a database that correspond to at least one previous depth map received within a predetermined time period of generation of the depth map data.

16. The method of claim 11 , wherein the first plurality of hand pose parameters for the wrist includes angle information about an orientation of the wrist in six degrees of freedom.

17. The method of claim 16 , wherein the six degrees of freedom further includes at least one degree of freedom for a longitudinal rotation that extends along a forearm.

18. The method of claim 17 , wherein the six degrees of freedom further includes at least one degree of freedom for an up-down angle for a raised or lowered palm of the hand.

19. The method of claim 18 , wherein the six degrees of freedom further includes at least one degree of freedom for a left-right rotational angle of the hand.

20. The method of claim 16 , wherein the first plurality of hand pose parameters for the wrist includes angle information about an orientation of the wrist in at least four degrees of freedom.