Training assistance using synthetic images

1. A non-transitory computer readable medium that embodies instructions that cause one or more processors to perform a method comprising:

(A) specifying a 3D model, stored in one or more memories, corresponding to an object;

(B) setting a run-time camera parameter for a run-time camera that is to be used in detecting a pose of the object in a real scene by:

receiving, through a data communication port, information representing an Augmented-Reality (AR) device having the run-time camera; and

acquiring, from one or more memories, the runtime camera parameter corresponding to the AR device in response to the information, a plurality of the run-time camera parameters and corresponding AR devices being associated with each other in one or more memories;

(C) setting a capture camera parameter for a capture camera that is to be used in capturing an image frame containing a reference object;

(D) acquiring, from the capture camera, the image frame of the reference object captured from a first view angle;

(E) generating a 2D synthetic image by rendering the 3D model in a second view angle that is different from the first view angle;

(F) generating training data using (i) the image frame, (ii) the 2D synthetic image, (iii) the run-time camera parameter, and (iv) the capture camera parameter; and

(G) storing the generated training data in one or more memories.

2. The non-transitory computer readable medium according to claim 1 , wherein the AR device is a head-mounted display device having the run-time camera.

3. The non-transitory computer readable medium according to claim 1 , wherein one camera functions as the run-time camera and the capture camera.

4. The non-transitory computer readable medium according to claim 1 , wherein the capture camera is a component of a smartphone or tablet.

5. The non-transitory computer readable medium according to claim 4 , wherein the one or more processors are a component of the smartphone or tablet.

6. The non-transitory computer readable medium according to claim 1 , wherein the method further comprises:

(H) selecting data representing a first view range that includes the first view angle, prior to step (D).

7. The non-transitory computer readable medium according to claim 1 , where the method further comprises:

(H) providing the generated training data to another processor connected to the run-time camera.

8. A non-transitory computer readable medium that embodies instructions that cause one or more processors to perform a method comprising:

(A) specifying a 3D model, stored in one or more memories, corresponding to an object;

(B) setting a run-time camera parameter for a run-time camera that is to be used in detecting a pose of the object in a real scene by acquiring, through a data communication port, the run-time camera parameter from an Augmented-Reality (AR) device having the run-time camera;

(C) setting a capture camera parameter for a capture camera that is to be used in capturing an image frame containing a reference object;

(D) acquiring, from the capture camera, the image frame of the reference object captured from a first view angle;

(E) generating a 2D synthetic image by rendering the 3D model in a second view angle that is different from the first view angle;

(F) generating training data using (i) the image frame, (ii) the 2D synthetic image, (iii) the run-time camera parameter, and (iv) the capture camera parameter; and

(G) storing the generated training data in one or more memories.

9. The non-transitory computer readable medium according to claim 8 , wherein the AR device is a head-mounted display device having the run-time camera.

10. The non-transitory computer readable medium according to claim 8 , wherein one camera functions as the run-time camera and the capture camera.

11. The non-transitory computer readable medium according to claim 8 , wherein the capture camera is a component of a smartphone or tablet.

12. The non-transitory computer readable medium according to claim 11 , wherein the one or more processors are a component of the smartphone or tablet.

13. The non-transitory computer readable medium according to claim 8 , wherein the method further comprises:

(H) selecting data representing a first view range that includes the first view angle, prior to step (D).

14. The non-transitory computer readable medium according to claim 8 , where the method further comprises:

(H) providing the generated training data to another processor connected to the run-time camera.

15. A method of detecting an object in a real scene using a computer, the method comprising:

(A) specifying a 3D model, stored in one or more memories, corresponding to the object, using a processor;

(B) setting a run-time camera parameter for a run-time camera that is to be used in detecting a pose of the object in the real scene, using the processor, by:

receiving, through a data communication port, information representing an Augmented-Reality (AR) device having the run-time camera; and

acquiring, from one or more memories, the runtime camera parameter corresponding to the AR device in response to the information, a plurality of the run-time camera parameters and corresponding AR devices being associated with each other in one or more memories;

(C) setting a capture camera parameter for a capture camera that is to be used in capturing an image frame containing a reference object, using the processor;

(D) acquiring, from the capture camera, the image frame of the reference object captured from a first view angle, using the processor;

(E) generating a 2D synthetic image by rendering the 3D model in a second view angle that is different from the first view angle, using the processor;

(F) generating training data using (i) the image frame, (ii) the 2D synthetic image, (iii) the run-time camera parameter, and (iv) the capture camera parameter, using the processor; and

(G) storing the generated training data in one or more memories, using the processor.

16. The method according to claim 15 , wherein the AR device is a head-mounted display device having the run-time camera.

17. The method according to claim 15 , wherein one camera functions as the run-time camera and the capture camera.

18. The method according to claim 15 , wherein the capture camera is a component of a smartphone or tablet.

19. The method according to claim 18 , wherein the processor is a component of the smartphone or tablet.

20. The method according to claim 15 , further comprising:

(H) selecting data representing a first view range that includes the first view angle, prior to step (D).

21. The method according to claim 15 , further comprising:

(H) providing the generated training data to another processor connected to the run-time camera.

22. A method of detecting an object in a real scene using a computer, the method comprising:

(A) specifying a 3D model, stored in one or more memories, corresponding to the object, using a processor;

(B) setting a run-time camera parameter for a run-time camera that is to be used in detecting a pose of the object in the real scene, using the processor by acquiring, through a data communication port, the run-time camera parameter from an Augmented-Reality (AR) device having the run-time camera;

(C) setting a capture camera parameter for a capture camera that is to be used in capturing an image frame containing a reference object, using the processor;

(D) acquiring, from the capture camera, the image frame of the reference object captured from a first view angle, using the processor;

(E) generating a 2D synthetic image by rendering the 3D model in a second view angle that is different from the first view angle, using the processor;

(F) generating training data using (i) the image frame, (ii) the 2D synthetic image, (iii) the run-time camera parameter, and (iv) the capture camera parameter, using the processor; and

(G) storing the generated training data in one or more memories, using the processor.